UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fractures in perspectives on good student writing Wiebe, Sunita 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-319604.pdf [ 12.54MB ]
Metadata
JSON: 831-1.0055903.json
JSON-LD: 831-1.0055903-ld.json
RDF/XML (Pretty): 831-1.0055903-rdf.xml
RDF/JSON: 831-1.0055903-rdf.json
Turtle: 831-1.0055903-turtle.txt
N-Triples: 831-1.0055903-rdf-ntriples.txt
Original Record: 831-1.0055903-source.json
Full Text
831-1.0055903-fulltext.txt
Citation
831-1.0055903.ris

Full Text

FRACTURES IN PERSPECTIVES ON GOOD STUDENT WRITING  by  SUNITA WIEBE B.A. ( 1 Class Hons.), University of Malaya, 1994 M.A. University of British Columbia, 1998 st  A T H E S I S SUBMITTED IN PARTIAL F U L F I L L M E N T O F T H E REQUIREMENTS FOR THE D E G R E E O F DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Educational Studies)  T H E UNIVERSITY OF BRITISH C O L U M B I A July 2007 © Sunita Wiebe, 2007  II  ABSTRACT While university students are expected to be good academic writers, there is little consensus on what constitutes good writing. The purpose of this study was to document fractures in instructors' and students' perspectives on good academic writing by surveying 157 instructors and 523 students about first-year writing at the University of British Columbia. The survey instrument consisted of a four-part questionnaire. The author used three composition pedagogies (Current Traditional Rhetoric, Expressivism and Social Construction) to ascertain how instructors and students ranked and graded three sample paragraphs of first-year student writing and assessed the importance of 45 writing attributes. Respondents' scores of the 45 attributes were aggregated into seven attribute families (Mechanics, Author's Voice, Social Analysis, Paragraph Structure, Academic Inquiry, Figurative Language and Academic Conventions). The author also measured the extent to which assessments of good writing were shaped by faculty's world views and personal characteristics, or their academic situations. Of the 14 measures used (three ranking options, three grading options, and assessments of importance of seven attribute families and combined attributes), there were nine fractures dividing instructors from students. The biggest involved Academic Inquiry (which instructors favoured), Social Analysis and Author's Voice (both of which students favoured). There was no consensus among instructors. All three paragraphs received a wide range of grades. Every paragraph was ranked top, middle or bottom with no majority opinion for any one paragraph about how best to write. Of the 14 measures, there were six fractures between instructor operations (paragraph ranking and grading) and preferences (importance of different writing attributes). Situational variables had more influence on instructors' paragraph assessments while personological characteristics were more predictive of the importance they assigned writing attributes. Instructors were most divided by employment status and world view. They were also divided by gender, country of birth and first language. Fractures in perspectives on good writing divide instructors from students as well as faculty themselves. Centralized Writing Departments that use a three-pronged research/pedagogical/ administrative approach should therefore be established to investigate fracture points; navigate students through such fractures; and provide writing researchers, instructors, and program planners administrative and funding support.  iii TABLE OF CONTENTS  ii  ABSTRACT TABLE OF CONTENTS  i"  LIST O F T A B L E S  vi  LIST O F F I G U R E S  ix  ACKNOWLEDGEMENTS  .  C H A P T E R O N E : THE MYTH OF GOOD ACADEMIC WRITING Warning: Writing Collisions Ahead Cultural Clashes in the University Purpose of the Study Compositional Collisions World Views Socio-Demographic Characteristics Framework of the Study Overview of the Study  X 1 3 6 8 9 11 13 14 15  C H A P T E R T W O : COMPOSITIONAL COLLISIONS  17  Historical Foundations of Composition Studies Current Traditional Rhetoric Epistemological Underpinnings Primary Features Criticisms Expressivism Epistemological Underpinnings Primary Features Criticisms Social Construction Epistemological Underpinnings Primary Features Criticisms Conclusion  18 20 21 22 22 23 24 25 26 27 28 29 30 31  C H A P T E R T H R E E : PROCEDURES Rationale for the Questionnaire Procedures for Data Collection  32 32 33  C H A P T E R F O U R : INSTRUMENT DEVELOPMENT  35  Section A: Sample Student Writing Section B: Attributes of Writing Pilot Testing the Attributes of Writing Final Version of Attributes of Writing Validity of the Attributes of Writing Section C: Differing World Views Operationalizing the Four World Views  36 38 39 42 44 49 54  iv Testing the Validity of the Eighty Adjectives Testing the Validity of the Twelve Adjective Families Section D: Background C H A P T E R FIVE: RESPONSES TO SAMPLE STUDENT WRITING Ranks Respondents Assigned the Paragraphs Instructors' Ranking of the Paragraphs Students' Ranking of the Paragraphs Grades Respondents Assigned to the Paragraphs Range of Grades Assigned to the Paragraphs Relationships between Paragraph Ranks and Paragraph Grades Correlations between Paragraph Ranks and Paragraph Grades.... Relationships between Paragraph Grade and Bottom, Middle and Top Paragraph Ranks Effects of Paragraph Ranking on Grading Analysis and Implications Dissenting Opinions of How to Rank and Grade the Paragraphs Varying Opinions of Writing Quality Overlapping Grade Ranges Consistency in Ranking and Grading Relationships C H A P T E R SIX: RESPONSES TO ATTRIBUTES OF WRITING Importance Assigned the Forty Five Writing Attributes Levels of Critical Importance Among the Forty-Five Attributes Attribute Families Instructors' and Students' Scoring of the Seven Attribute Families Analysis and Implications C H A P T E R S E V E N : RESPONDENTS' WORLD VIEWS How Respondents Scored the Eighty Adjectives Respondents' World View Profiles Analysis of Respondents' World Views  61 65 66 70 71 71 72 72 .75 78 79 80 83 84 84 85 86 86 88 89 :...94 98 102 105 108 108 113 115  C H A P T E R EIGHT: RESPONDENTS' SOCIO-DEMOGRAPHIC CHARACTERISTICS . . 116 Socio-Demographic Characteristics of Respondents Respondents' Roles Respondents' Languages Respondents' Academic Qualifications Summary of Respondents' Socio-Demographic Characteristics C H A P T E R NINE: RIVAL HYPOTHESES  117 118 121 124 126 ....127  Personological vs. Situational Rival Hypotheses .128 Rival Hypotheses: Statistically Significant Fractures in Instructors' Paragraph Rank and Grade Assessments ...131 Rival Hypotheses: Statistically Significant Fractures in Instructors' Attribute Family Assessments 136 Summary of Key Findings 142 Analysis and Implications 144 C H A P T E R T E N : FRACTURES IN PERSPECTIVES ON GOOD STUDENT WRITING .: 147 Fractures of Writing between and among Respondents  147  V  Perspectives on Good Student Writing Conclusions  153 157  C H A P T E R E L E V E N : THE REVOLUTION BEGINS WITH WRITING - CONCLUSION, DISCUSSION AND REGOMMENDATIONS 158 The Globalized University and Fractures of Good Writing 158 Alternate Perspectives on Good Academic Writing 163 Recommendations for Administrators, Instructors and Researchers within the Writing Department 167 Final Thoughts and Future Research Projects 168 REFERENCES  171  APPENDICES  179  vi LIST O F T A B L E S Table 1: Initial 30 Pilot Test Attributes According to Pedagogy  39  Table 2: Importance of Thirty Pilot Test Attributes in Descending Order  41  Table 3: Final Version of the Writing Attributes Categorized According to Pedagogy  43  Table 4: Correlations between Importance Respondents Assigned Writing Attributes and Composition Pedagogies  45  Table 5: Correlations between Pedagogies and Importance Assigned Attributes by Instructors and Students  48  Table 6: The regulation-radical change dimension  51  Table 7: Pilot Test Adjectives According to World View  56  Table 8: Correlations of 30 Functionalist Adjectives with four World View Indices in Descending Order of Functionalism  57  Table 9: Correlations of 30 Interpretive Adjectives with Four World View Indices in Descending Order of Interpretivism  58  Table 10: Correlations of 30 Radical Humanist Adjectives with four World View Indices in descending order of Radical Humanism  59  Table 11: Correlations of 30 Radical Structuralism Adjectives with four World View Indices in Descending Order of Radical Structuralism Table 12: Eighty Adjectives Listed Alphabetically According to Paradigm  60 61  Table 13: Correlations of 20 Functionalist Adjectives with World View Indices in Descending Order of Functionalism  62  Table 14: Correlations of 20 Interpretive Adjectives with Four World View Indices in Descending Order of Interpretivism  62  Table 15: Correlations of 20 Radical Humanism Adjectives with four World View Indices in Descending Order of Radical Humanism  63  Table 16: Correlations of 20 Radical Structuralist Adjectives with four World View Indices in descending order of Radical Structuralism Table 17: Correlations between Adjective Families and World Views  63 66  Table 18: Instructors' and Students' Preference Ranks for Paragraphs in Three Pedagogical Styles  71  Table 19: Instructors' and Students' Assessments of the Paragraphs (Grades and Standard Deviations  73  Table 20: Grades Instructors and Students Assigned Current Traditional, Expressivist and Social Construction Paragraphs by Rank  77  VII  Table 21: Grade/Rank Correlations for Current Traditional, Expressivist and Social Construction Paragraphs  79  Table 22: Instructors' and Students' Paragraph Assessments by Rank..  81  Table 23: Variance Accounted for in Instructors' and Students' Paragraph Grades by Paragraph Rank  83  Table 24: Writing Attributes According to Pedagogy  89  Table 25: Instructors' Assessments of the 45 Writing Attributes in Descending Order of Importance  90  Table 26: Instructors' and Students' Distributions of Critically Important Points Among the 45 Writing Attributes  95  Table 27: Rotated Component Matrix for the Seven Attribute Families  99  Table 28: Overall, Instructors' and Students' Assessments of Importance of Seven Attribute Families. Differences in Instructors' and Students' Assessments of Importance. Variances Accounted for in Assessments of Importance by Level....  103  Table 29: Respondents' Endorsements of the 80 Adjectives in Descending Order  109  Table 30: Socio-Demographic Research Areas and Sub-Areas  116  Table 31: Instructor and Student Tallies by Region of Birth  117  Table 32: Instructor and Student Tallies by Year of Arrival in Canada  118  Table 33: Instructor Tallies by Length of UBC Employment  119  Table 34: Instructor Tallies by Department  120  Table 35: Student Tallies by Faculty of Enrolment  ..120  Table 36: Students' Intended Major by Faculty  121  Table 37: Instructor and Student Tallies by First Language  122  Table 38: Instructor and Student Tallies by Years Writing Academic Papers in English  122  Table 39: Instructor and Student Tallies by Language of Highest Academic Qualification  123  Table 40: Instructor Tallies by Year of Marking Training.  123  Table 41: Instructor and Student Tallies by Highest Academic Qualification  124  Table 42: Instructor and Student Tallies by Year of Highest Academic Graduation  125  Table 43: Instructor and Student Tallies by Geographic Region in which Highest Academic Qualification was Earned  125  Table 44: Variances Accounted for in Instructors' Paragraph Assessments and Importance of Writing Quality Attributes by Collective Personological and Situational Factors  129  Table 45: Average Variances Accounted for in Instructors' Paragraph Assessments and Importance of Writing Quality Attributes by Collective Personological and Situational Factors  :  130  VIII  Table 46: Correlations between Instructors' Paragraph Assessments (Ranks and Grades) and Importance of Attribute Families  131  Table 47: Personological and Situational Influence on Instructors' Paragraph Assessments (Ranks and Grades)  132  Table 48: Correlations between Instructors' Four World Views (Functionalist, Interpretive, Radical Humanist and Radical Structuralist) and Paragraph Assessments (Ranks and Grades)  133  Table 49: Differences in Instructors' Paragraph Assessments by Situational Characteristics (Employment Status, Departmental Grouping, and Length of UBC Employment)  134  Table 50: Personological and Situational Impact on Instructors' Assessments of Importance of Combined Attributes of Writing  136  Table 51: Personological and Situational Influence on Instructors' Assessments of Importance of Seven Attribute Families  138  Table 52: Correlations between Instructors' Scoring of the Seven Attribute Families' Importance and World Views (Functionalist, Interpretive, Radical Humanist and Radical Structuralist) 139 Table 53: Instructors' Assessments of Seven Attribute Families' Importance by Gender, Country of Birth, First Language, UBC Role, Part/Full-Time Employment Status, Departmental Grouping, Length of UBC Employment, and Marking Training  141  LIST OF FIGURES  Figure 1 : Burrell and Morgan's scheme for analysing assumptions about the nature of social science  51  Figure 2: Four paradigms for the analysis of social theory  52  Figure 3: Instructors' Grade Ranges for the Three Paragraphs  76  Figure 4: Students' Grade Ranges for the Three Paragraphs  76  Figure 5: Respondents' Endorsements of the Functionalist, Interpretive, Radical Humanist and Radical Structuralist World Views  112  Figure 6: World View Profiles for Respondents on Average, Instructors, Students and Two Sample Respondents Figure 7: Fractures in Assessments of Student Writing between Instructors and Students  114 148  Figure 8: Fractures in Students' Assessments of Attribute Family Importance by Gender, Country of Origin and First Language  151  Figure 9: Fractures in Instructors' Assessments of the Paragraphs and Attribute Family Importance  154  X  ACKNOWLEDGEMENTS This study is the result of long-standing collaborations with some truly generous and remarkable people. Roger Boshier, John Collins, Carl Leggo and Dan Pratt - the four gentlemen of my thesis committee - have been unwavering in their support and enthusiasm for this project. They saw promise and potential even when concepts were at their murkiest and most nebulous. I am especially grateful to Roger Boshier for reading every word of this thesis in its many incarnations, and for reminding me of the importance of brevity. Steven Wiebe, my husband and all-round inspiration, provided sanity and comfort through the many ups and downs of graduate school. Thank you, Steve, for being my rock. Raveen Kumaran provided invaluable help with the formatting of this thesis and defense presentation. I am grateful for his technical expertise. Seetha Friar, best friend and sister, reminded me to stay focused and to laugh. Finally, I would like to dedicate this work to the three people who made it possible. My parents, Jeya and Kumaran, taught me the value of prayer and the glory of words. They are living embodiments of the wondrous power of faith. John Collins - research supervisor, guardian angel and friend - held my hand and never let me go through the labyrinth of doctoral studies. Thank you, John, for your patience and compassion. This work would not exist without you.  1 CHAPTER ONE: THE MYTH OF GOOD ACADEMIC WRITING While the pursuit of good academic writing is a crucial component of higher education, there is frequently confusion about it. Some years ago, the English Department where I was a teaching assistant decided to host a series of marking meetings. The aim was to give instructors, tenured faculty, sessionals and teaching assistants the chance to establish standards for first-year writing. What happened was less than inspiring. Tales abounded of hurt feelings, shouting matches and rivers of tears. Teaching assistants accused tenured faculty of ridiculing their opinions; faculty members berated younger teaching staff for violating departmental standards. Instructors who graded high were labeled "soft" and castigated for betraying the principles of the teaching profession; those who graded lower were charged with being "draconian" and "elitist." Far from being a bonding experience, marking meetings exposed deep and bitter divisions. This was a large department of nearly one hundred instructors of different age groups, ethnicities and intellectual backgrounds. Yet, this heterogeneous teaching body was expected to adhere to the same standards, even though they were not described. Reasons for fractures surrounding perspectives on good academic writing were never investigated, but side-stepped as if explosive disagreements were embarrassments that could never be referred to again. These unresolved issues form the impetus for this study. How prevalent are differing perspectives on good first-year student writing, and at what points are disagreements strongest? What factors influence writing preferences? Because the teaching and learning of first-year English does not exist in a vacuum, any examination of good academic writing must consider the context within which such activity occurs. Thus, this study is not limited to instructors' and students' perspectives on academic writing to composition classrooms; instead, it probes dominant themes of the University at the beginning of the twenty-first century. North American universities are among the most diverse places on Earth, comprising individuals of disparate age and ethnic groups, sexual orientations, religious and political beliefs, intellectual training, philosophical and cultural affiliations. Top-tier schools also face tremendous pressure to become internationalized. Over the past three decades, the number of students leaving home to study abroad has grown at an  2 annual rate of 3.9 percent, from 800,000 in 1975 to 2.5 million in 2004 (Newsweek, 2006). This academic migration involves students from one developed country to another, as well as an increasing number of students from developing countries. Canada receives more than 130,000 international students every year and foreignstudent enrolment increased by more than 15 percent across the country in 2003, with many provinces showing jumps of 20 percent or greater (Association of Universities and Colleges of Canada, 2006). This mixture is further enriched by recent immigrants, and by non-native speakers of English expected to succeed in institutions where English is the lingua franca. Apart from multicultural education, however, there is limited discussion of globalization on the teaching and learning of academic writing. Instead, there is temptation to constrain campus diversity to a clever marketing ploy and glossy campus brochures of a happy, multi-ethnic student body. "Diversity" and "globalization" become convenient catch phrases used on web-sites, press releases and mission statements to suggest a tolerance of difference. Yet, within the context of first-year writing, differences of opinion are not likely to pose for shiny, smiling pictures. First-year English is mandatory for most incoming university students, regardless of academic discipline or experience with. Instructors and students have to deal with classrooms filled with individuals of different age groups, language proficiencies, interests, ethnicities, cultural backgrounds, academic orientations, pet peeves, prejudices, expectations, goals, life experiences and handicaps. First-year English instructors I spoke to mentioned challenges posed by newly arrived, E S L students but offered little by way of solutions (one instructor condescendingly referred to his East Asian, E S L female students as Hello Kitty Girls). In a training seminar I attended, teaching assistants were instructed to watch out for dozens of ESL-type errors in their students' writing, but provided no guidance on how to teach writing in a multicultural, multilinguistic and multiethnic classroom. Somehow within this environment of oblique references, guessing games and half-spoken assumptions, instructors and students were expected to share a common understanding of academic writing. As the marking meetings revealed on a small scale, the globalized, open and tolerant university is the site of deep, acute and infinite fractures concerning writing. The word "fracture" refers to a literal, frequently physical rupture or break. It evokes sharp, jagged edges and hurt. The metaphor is deliberate and apt: the global  3 university is constantly rent by arguments over what constitutes good writing, and students are in danger of falling into crevasses of confusion and debate. Students I spoke to during this study - particularly those born outside Canada - frequently expressed anguish, frustration and despair. They were desperate to become good writers but had little idea how to produce writing their instructors would like. Worse, they were aware that what one instructor wanted, another would reject. Instead of smooth, melded consensus, their stories revealed conflicting perspectives on good writing. These fractures are airbrushed from campus brochures but exert a painful and tangible influence. Fractures abound because there is no fail-proof perspective on good student writing. Instead, there are multiple legitimate approaches, depending on context and discursive purpose. The marking meetings uncovered disjunctures of opinion but little was done to deepen awareness of why they existed.  Warning: Writing Collisions Ahead The phrase "good academic writing" is paradoxical: postmodern scholarship opposes such simplistic binary oppositions as good/bad, positing such arbitrary terms as empty signifiers. On the other hand, student writing is frequently categorized as either "good" or "bad," resulting in dismal grades, course failures, rejections from desired programs, overheated arguments between faculty and students, and even conflict among instructors. Any student who has ever received a "D" in first-year English can probably attest to feeling hurt and rejected. Good writing depends on constraints which specific circumstances impose upon writers (Leki, 1995). It is subjective, as numerous studies have revealed.  Diederich's  (1974) study involved 53 readers commenting on and ranking 300 essays on a scale of one to nine. The results were inconsistent: every text earned every ranking between one and nine. Coles and Vopat's (1985) research revealed a comparable range. The authors asked 48 composition scholars to contribute a piece of student writing at the college level that represented excellence, however flawed or unfinished. The submitted pieces came in a multitude of forms: grammatically unconventional and highly emotional first-person pieces alongside formulaic essays with clear thesis statements and a complete avoidance of overt authorial voice. A s Coles and Vopat demonstrated, what one deems excellent, another may consider mediocre, suggesting "no judgments about  4 writing can be taken as final or absolute" (p. viii). Leki's study produced similar results. When she asked 29 university instructors to rank four student essays according to the criteria they might apply when looking for "good student writing," every one of the four pieces received every rating from "best" to "worst." Li's (1996) research further problematized the quest for "good writing" by introducing a cross-cultural twist: 60 writing instructors from two countries, China and the U S , were asked to rank four student essays. Each piece received at least three different rankings from participants of each country. Other research indicates fractures occur among institutions, academic departments, as well as faculty members within a particular department or program (Johns, 1997; Lea, 1994; Lea & Street, 1998). Where one instructor might demand a structured approach and an impersonal tone, another might prefer a looser, more exploratory format and use of the personal pronoun; where some faculty members validate terse bullet-headings, to-the-point paragraphs, and simple diction, others may expect convoluted sentence structures and liberal amounts of jargon. Vardi (2000) discovered that instructors within the same business faculty of a large Australian university each had divergent perspectives on student writing. While an accounting instructor wanted thesis statements in students' opening paragraphs, a statistics lecturer preferred headings and a content page. Vardi determined that these expectations were rooted in four factors: the instructor's reason for setting the essay task; the thinking of the discipline; the lecturer's personal belief about "good writing" in relation to teaching objectives; and need to assess students' understanding. These variables collectively contributed to what instructors expected from students. Yet another fracture involves inconsistencies within the instructor's compositional and pedagogical philosophies. There are four distinct philosophies of composition: formalist, expressive, mimetic and rhetoric (Fulkerson, 1979). Formalist instructors judge student writing according to internal form, i.e. display of grammatical and syntactical correctness, while adherents of expressive views value writing that features an honest and personal voice. In contrast, the mimetic approach validates logic and reason while rhetoricists argue that "good writing" is adapted to achieve a desired effect on an audience. Though the four philosophies are not mutually exclusive, a problem emerges when instructors move mindlessly from one approach to another. If an instructor asks students to state their opinions on topic X , the assignment is expressive  5 in nature and the essay must be considered successful so long as students state their opinions. It should not matter if the opinion is logically flawed, unpersuasive or incorrect since the assignment is not mimetic, rhetoric or formalist in nature. Unfortunately, instructors who are less than mindful of contradictions of their practice might fall into the trap of assigning an essay that fits the requirements of one philosophy, but use different lenses when evaluating writing. Perspectives on good academic writing are also the result of social forces. A s Li argued, "['good writing'] is a splice of multiple linguistic and nonlinguistic, cultural and historical strands; of what is written in a piece and the manner in which the piece is written; of ideology and aesthetic; of society and individuals" (1996, p.111). Rather than being a transparent medium which provide writers with discursive skills, language and literacy are closely linked to social values and practices and are used to construct knowledge and ways of looking at the world (Lea, 1999). Word choices and forms of expression are indicative of rules that govern what can and cannot be said, and of systems of acceptable thought. Language appropriate in one context can appear awkward and unsophisticated in another. Writing plays a crucial role at all levels of academia, from the first-year undergraduate to the tenured professor. Entrance examinations, application essays, term papers, grant applications, proposals and published papers, determine students' and faculty members' acceptance into, and successful progression through academia. Consequently, good writers, as implied by "publish or perish," gain entry into the clubhouse of the academy, and have opportunities to thrive. There is intense pressure to produce acceptable writing as demands for university degrees have led to increased student enrollment and limited teacher-student interaction. Since instructors have few chances to develop close relationships with their students, writing is the primary means of assessing student achievement and of determining which students earn high grades, qualify for scholarships, or become eligible for competitive academic programs. The lack of consensus concerning good writing can result in fractures between instructor expectations and student perspectives (Lea, 1994; Lea & Street, 2000; Street, 1999). Beginning students bring literacy practices and modes of expression used successfully in other situations, only to discover they are unsuitable in an academic forum. A s Lea's (1994) aptly titled essay "I Thought I Could Write Until I Came Here" suggests, students often feel they have to learn a new language with unfamiliar rules  6 (see also Bartholomae, 1988; Bizzell, 1982; Lillis, 1997). Mature students who may be accomplished writers outside the university discover their skills do not mesh with the academy. The gulf between prior understandings of writing and academic literacy can have a debilitating effect on identity and confidence (Clark & Ivanic, 1997; Ivanic, 1998). Because perspectives on good writing are rarely articulated, there is also limited discussion about the rules of academic writing. One persistent theme in writing is that writing conventions are implicit rather than explicit (Flower, 1994; Hounsell, 1997; Lea 1994; Lillis, 1997; Scott, 1996; Taylor, 1988). Instructors do not think it necessary to explicate writing conventions because they assume students already possess such information. Such assumptions are increasingly problematic when university classrooms are multicultural and more mature students with varied professional backgrounds are returning to tertiary education. What the instructor expects the student to know might be in conflict with what he or she already knows, and vice versa. Furthermore, not being privy to the philosophies that shape their instructor's standards of assessments, students are often at a loss. Contradictions are compounded as students move from department to department, or program to program. Street (1999) described the frustration of student-instructor miscommunication during a doctoral hearing in which it was apparent the candidate did not share the meanings attributed to key phrases being used by the examiners. There was considerable confusion, he wrote, over such terms as "tease out" and "elaborate" thesis statements, "making generalizations," "pitching it at a more abstract and analytic level," "making themes more explicit" and "pulling them out from the embedded text." "As the examiners struggled to communicate our meanings," Street noted, "the student felt we were either asking for 'repetition' and 'redundancy' or that she had already done everything we were referring to (pp. 194-5)."  Cultural Clashes in the University Debates on good academic writing are symptomatic of cultural clashes that percolate North American universities. Snow (1959) described the university as being split by two cultures: one scientific, the other humanist. Two divisions that have gained prominence on university campuses in recent years are the Culture of Excellence vs. the Culture of Thought, and the Culture of Inclusion vs. the Culture of Elitism. The Culture of Excellence stems from a phenomenon sweeping contemporary university  7 campuses: the transformation of institutions of higher learning into profitable, bottomline-efficient corporations (Altbach, 1999; Giroux & Myrsiades, 2001; Readings, 1996) complete with the language and mission of big business. Like successful corporations, universities must establish themselves as competitive securers of the brightest and the best and actively jostle for favourable rankings in such college guides as Maclean's University Guide, US News and World Report's annual college rankings, and The Princeton  Review.  The guides judge universities according to a variety of elements, ranging from library acquisitions, to added value, from financial assistance to campus food. The successful university is defined as a consumer-driven enterprise offering a product (education) to its customers (students) in as pleasant, non-threatening, and as physically comfortable a manner as possible. Students are to operate as passive agents, selecting universities in the manner of car-shopping drivers (Readings, 1996). The University of Excellence focuses on commodities that can be measured, bought or sold (e.g. faculty/student ratio) and on transmitting knowledge quickly, cheaply and efficiently. There is little room in the University of Excellence for the Culture of Thought. Within the Corporate University, students are not encouraged to ask the difficult questions Thought involves, but accept the conveniently empty signifier of Excellence as a replacement for learning (Readings, 1996). Another concern is that universities aiming for Excellence rather than Thought fail to equip students with the capacity for critiquing social and political structures (Giroux, 2001); instead, the University becomes a venue for producing job-seekers with marketable skills and a life-long desire to become the perfect consumer. Another division that has had dramatic repercussions on university campuses involves the conflict between the Culture of Inclusion and the Culture of Exclusion. North American universities are more heterogeneous than ever with increasing participation from women, ethnic minorities, recent immigrants, foreign students, E S L speakers, and the economically disadvantaged. While such inclusion is strongly advocated by multiculturalists and proponents of social justice, it has received criticism from groups arguing for universities to retain a culture of exclusivity. One argument claims affirmative action and diversity policies have caused intellectual mediocrity (Bercuson, Bothwell & Granatstein, 1997) because universities admit too many ill-prepared students. Such policies, critics of diversity policies allege,  8 pit white (majority) instructors and students against minority counterparts. Affirmative action, they allege, favours under-prepared minority students over majority individuals with higher scholastic scores, and engenders feelings of shame among white instructors because they do not sufficiently address culture and ethnicity in their classrooms (Browne-Miller, 1996). Proponents of exclusivity also claim the Western canon is under attack. Right-wing critics like D'Souza, Cheney, Will, Bloom and Bennett argue that the University is controlled by leftists dismantling curricula based on centuries of Eurocentric thought and culture for relativistic, ethnic fluff. The left, they contend, is motivated by deep contempt for Western culture (Hollander, 2000), and, in attempting to destroy the canon, offer their students an inferior education.  Purpose of the Study Inspired by writing fractures and cultural clashes, the purpose of this study was to document and explain fractures in instructors' and students' perspectives on good academic writing within a globally oriented and prestigious research university in western Canada. An additional aim was to determine if these fractures were more powerfully informed by who instructors were personologically (i.e. their world views and personal characteristics) or academically (i.e. their academic situations). In order to accomplish these goals, I used these variables: •  Compositional Collisions: How did instructors and students perceive good academic writing? How did they assess three sample texts as well as attributes of writing derived from three major composition pedagogies? Were there statistically significant fractures among and between instructors and students?  •  World Views: How did respondents perceive the world around them? Did they espouse different views which influenced perspectives on good writing?  •  Socio-Demographic Characteristics: What were respondents' personal characteristics and academic situations? Was there a relationship between their personological or situational characteristics and perspectives on good academic writing? If so, which set of characteristics exerted greater influence?  I discuss each of these themes below.  9 Compositional Collisions Themes that gird the cultures of Excellence vs. Thought, and Inclusion vs. Exclusion play a vital role in composition pedagogy. A central debate revolves around the purpose of first-year English. Is it functional, i.e. to train undergraduates for future careers and the writing of memos and reports free from spelling and syntax errors? Or, is it a medium for thought, to provoke incoming students to think critically and challenge established precepts? Is the role of composition instruction to transform a polyglot, motley horde into linguistically streamlined grammarians? Or to prepare beginning students for entry into the scholarly community? Closely related to issues of purpose are questions about the type of discourse firstyear English is meant to promote. Is first-year writing about the protection and exaltation of standardized English: clear, transparent prose objectively measured as either "excellent" or "poor"? Alternatively, is it about subverting the exclusivity of correct, university-level English by validating the quirks of expression that characterize multicultural, multiethnic, multigenerational first-year classrooms? Or is it about polishing the established discourse of the academy to proceed in intellectually smooth, familiar patterns? Debates surrounding Excellence vs. Thought and Inclusion vs. Exclusion also percolate within composition studies. Composition pedagogy is a textured and complex field with a number of rhetorical approaches; consequently, a major challenge was identifying which approaches to writing or composition pedagogies would best capture these debates. In the interests of manageability, I selected three pedagogies based on the following criteria: 1) each pedagogy had to be distinct from the other two in terms of philosophical orientation and intellectual sources; 2) each pedagogy had to advocate an idiosyncratic form of good academic writing that could be clearly identified as belonging to that particular rhetorical approach; 3) each pedagogy had to champion a unique set of writing attributes that were integral to its conception of good academic writing; 4) each pedagogy needed to have had significant impact on the teaching and assessing of first-year English; 5) each pedagogy had to have an established history. One means of classifying composition strands was to categorize them by the following rhetorical approaches: objective rhetoric, subjective rhetoric, and transactional rhetoric (Berlin, 1987). Objective rhetoric locates reality in the material world; subjective rhetoric argues it is a private and personal construct; and transactional rhetoric  10 discovers reality "in the interaction of the features of the rhetorical process itself - in the interaction of material reality, writer, audience, and language" (Berlin, p.155). Within these approaches, three composition pedagogies most clearly met my criteria: Current Traditional Rhetoric (objective rhetoric), Expressivism (subjective rhetoric), and Social Construction (transactional rhetoric). Current Traditional Rhetoric is the oldest of the three and the dominant composition pedagogy of the twentieth-century (Berlin, 1988; Burnham, 2001; Crowley, 1990, 1998). It traces its genesis to the work of rhetoricians identified with a philosophical orientation known as Scottish common sense realism. These rhetoricians - Campbell, Blair, and Whateley - were inspired by the tenets of the Enlightenment: reason as opposed to passion; the search of an objective truth; the pursuit of a concrete reality based on sensory perception. Good Current Traditional writing has to reflect this objective reality through the use of clear and correct syntax, unified and coherent structures, and precise vocabulary. Language is neutral, unambiguous and divorced from human emotion. Expressivism was mounted as a direct challenge to Current Traditional Rhetoric in the sixties and seventies. Once dismissed as a fringe movement, Expressivism now constitutes mainstream writing practice and is the dominant subjective rhetoric approach (Berlin, 1987). The term "expressive" and its reference to the expression of the self have roots in the linguistic research of Hymes, Jakobson and Sapir, and in the phenomenology of Merleau-Ponty, Husserl, Sapir and Cassirer (Kinneavy, 1980). In direct contrast to Current Traditional Rhetoric, Expressivism rejects the hierarchy of correct/incorrect that is its basis, and posits reality as highly subjective in that experiences and perceptions of social settings differ from individual to individual. By focusing on the "encoder," Expressivism encourages student writers to consider their interpretations of this subjective world as they develop writerly voices. Good Expressivist writing does not need to adhere to stylistic or grammatical conventions; instead, it is an avenue for the phenomenology of the self, reflecting the writer's emotions, thoughts, forms of expression and voice (Britton, et. al., 1975; Coles, 1978; Elbow, 1973; Macrorie, 1970; Murray, 1968). Expressivism's strongest criticisms have come from Social Construction (Babin & Harrison, 1999; Paley, 2001). Social Constructionists oppose Expressivism's emphasis on individual construction of subjective reality, arguing that the world be interpreted through social filters. They reject the idea of the self as a unified, knowable entity, and  11 critique Expressivist instructors for failing to acknowledge the impact of social context in their classrooms. Drawing upon Vygotsky, Rorty and Foucault, Social Construction stresses the social character of language and posits writing as a social act (Bruffee, 1983). This examination of the social nature of language has resulted in a social epistemic approach (Berlin, 1988) which, inspired by Freire, Giroux, Shor, provides for an even closer critique of the systems of power embedded in discourse. Consequently, good Social Constructionist writing must demonstrate critical awareness of the social construction of knowledge, and of the way language is used, abused, defined and defied in a discourse community. The three pedagogies - Current Traditional Rhetoric, Expressivism, and Social Construction - are discussed further in "Compositional Clashes" (Chapter Two).  World Views Berlin's (1987) contended that each of the three major rhetorical approaches objective, subjective and transactional - located reality in a different domain. Within the context of a study about student writing, Berlin's hypothesis took on deeper implications: namely, are perspectives on good writing informed by one's understanding of reality? At issue was the interplay between rhetoric and knowing. A s Berthoff (1972) noted, rhetoric "reminds us that the function of language is not only to name, but also to formulate and to transform - to give form to feeling, cogency to argument, shape to memory" (p.647). Without language, there is no reality, reinforcing the inextricable link between the two. One interprets one's world through the prism of language. Given that this study took place in an academic milieu dedicated to defining and questioning what it means to know, reality assumed an added dimension beyond that of interaction with one's immediate environment. Specifically, how does one understand the nature of knowledge itself, and how is this understanding related to assessments of language? Investigating the relationship between rhetoric and reality required a means of situating respondents that went beyond psychologically inspired personality tests to gauge a fundamentally deeper matter: participants' world views. In their landmark text Sociological Paradigms and Organizational Analysis, Burrell and Morgan (1979) argued that all major social theory can be conceived in terms of paradigms. Each paradigm is based upon a set of mutually exclusive metatheoretical  12 assumptions about ontology and social change. Each paradigm or generates theories and perspectives fundamentally different from those engendered by other paradigms. The four paradigms are Functionalist, Interpretive, Radical Humanist and Radical Structuralist. Functionalism presupposes an objective reality determined through positivist research: social structures have discernible functions and must be maintained through consensus and an adherence to order. In contrast, the Interpretive paradigm is characterized by its reliance on subjectivity and defines social reality as little more than a shared network of assumptions and subjective meanings. Radical Humanism, like the Interpretive paradigm, is based upon a subjective ontology that defines reality as socially created and sustained. However, while the former seeks to preserve this order, the latter aims to overthrow it by getting people to recreate their understanding of their social environment through critical thinking. Finally, Radical Structuralism views reality as objective and concrete, subscribing to the idea of overarching and dominating social structures that must be overthrown and replaced by another, fundamentally different set of social hierarchies and institutions. Apart from providing necessary scaffolding for reality as defined by the parameters of this study, the paradigms or world views suggest a conceptual overlap with the three pedadogies this study investigates. Like Current Traditional Rhetoric, Functionalism depends on an objective, fact-driven form of reality which can be proven or disproven. Reality or truth is posited as having a tangible presence in the external world and the individual is responsible for uncovering and subscribing to it. Expressivism and the Interpretive paradigm conceive reality as subjective and internally generated. The individual is the field of inquiry and the researcher is driven by the spirit of accommodating and reconciling disparate perceptions of reality so they become part of a shared social fabric. Finally, Social Construction, Radical Humanism, and Radical Structuralism draw heavily upon interactions between the individual and the social milieu. The social fabric is the focus of investigation: it must be scrutinized, challenged and ultimately overthrown. How these epistemological overlaps translate into actual assessing of student writing is the subject of this study. The four world views are discussed further in "Respondents' World Views" (Chapter Seven).  13 Socio-Demographic Characteristics Universities comprise hundreds of thousands of individuals with unique personal histories as well as academic situations and expectations. Research on institutions of higher learning reveals campuses are often divided along the lines of gender, ethnicity and status. While the percentage of women earning professional and graduate degrees has risen considerably, women are still severely underrepresented in fields like computer science, and engineering (Statistics Canada, 2004). Additionally, the percentage of women faculty members has only risen by a fraction since the early part of the twentieth century. Women are less likely to be tenured than their male counterparts (42 vs. 66%; N C E S , 2005) or be full professors (15 vs. 39 %, N C E S , 2005). Women faculty also earn consistently less than their male counterparts ( N C E S , 2005), even in cases where salary-related characteristics are the same for men and women, and are penalized for having children and starting families by being denied tenure (Acker & Armenti, 2004. They are more likely to be engaged in teaching and service activities than in research and administration ( N C E S , 2005). Women faculty have also reported feelings of marginalization and isolation despite their academic qualifications (Aisenberg & Harrington, 1988; Kirsch, 1993). Ethnicity is another lightning rod. Even though minority participation has increased significantly in the last forty years, the rate of participation in higher education among Hispanic and African-American students still lags behind the participation rates of Caucasian students (Harvey, 2002). Additionally, minority faculty are still underrepresented and, if hired, are more likely to be underpaid sessionals than tenuretrack professors. While universities have attempted to redress imbalances through the inception of affirmative action policies and departments of ethnic studies, diversityoriented policies are now under considerable fire (a recent suit against the University of Michigan's Law School's affirmative action policy is a case in point), and minority faculty and students have reported cases of marginalization and discrimination (Aguirre, 2000; Suzuki, 1994). Finally, there is status. Even though universities pride themselves on being democratic institutions, hiring practices distinguish sessionals from tenure-track or tenured professors. Forty-six percent of all faculty are considered "part-time" and nontenure track appointment account for 65% of all faculty appointments in American  14 universities (American Association of University Professors, 2006). Since budgetstrapped universities can now hire two sessionals for every tenured professor, the trend shows no sign of abating. Sessional or part-time instructors generally receive less pay, work full-time hours, and have little control over their curricula (Rajagopal, 2002). Additionally, because the hiring boom of tenured faculty ended in the early eighties, sessional instructors tend to be younger than their tenured counterparts. Respondents' personological and situational characteristics are discussed in greater detail in "Respondents' World Views" (Chapter Eight). Framework of the Study The purpose of this study was to document fractures concerning what is meant by good writing and explain where they came from. Fractures were calculated from disjunctions between respondent scores. Hence, a faculty/student fracture reflected the magnitude of the difference between the way faculty and students assessed various measures of good writing. The possible fractures of writing were measured in the following ways: •  Paragraph Assessment: Differences in respondents' ranking and grading of three contrasting samples of academic writing, each representing one of the three pedagogies used in the study;  •  Importance Assigned to Writing Attributes: Differences in respondents' scoring of the importance of 45 writing attributes derived from the three pedagogies;  •  Importance Assigned to Attribute Families: Respondents' attribute scores were aggregated into "attribute families," yielding differences in respondents' scoring of the families.  The subsequent investigative step was to determine the variables that best explained fractures. There were two sets of contesting predictor variables: situational and personological. In other words, to what extent did faculty (or student) assessments of good writing vary as a function of their "person" or "situation?" The predictor or explanation variables were as follows: •  Personological Variables: World views (Functionalism, Interpretivism, Radical Humanism and Radical Structuralism), gender, age, first language, and country of origin;  15 •  Situational Variables: University role, full/part-time employment status, qualification, departmental grouping, length of U B C employment, country of highest academic qualification, language of highest academic qualification, year of graduation, marking training.  Overview of the Study  Data were collected at the University of British Columbia in Vancouver, B.C. during the Fall of 2004 and the Spring and Summer of 2005. Respondents comprised 157 faculty (teaching assistants, sessional instructors, tenured and tenure-track professors, senior instructors, post-doctoral fellows, and emeritus professors) and 523 first-year students enrolled in either first-year English or Writing 098, a preparatory writing course. This study consists of eleven chapters. In this first chapter, I have outlined the central problem which prompted this research, the theoretical framework, purposes and methodology for the study.  "Compositional Clashes" (Chapter Two) presents an  historical overview of compositional research in North America and examines the theoretical underpinnings, weaknesses and strengths of Current-Traditional Rhetoric, Expressivism and Social Construction. "Procedures" (Chapter Three) describes steps taken to collect data while "Instrument Development" (Chapter Four) discusses those taken to develop, test and validate each section of the four-part questionnaire particularly respondents' views concerning the importance of various attributes of writing and how they subscribed to four world views. "Responses to Sample Student Writing" (Chapter Five) presents the first statistical findings for this study i.e. how each of the sample paragraphs were ranked and graded by faculty and students. "Responses to Attributes of Writing" (Chapter Six) examines how the more detailed attributes of writing derived from the three pedagogies were scored by faculty and students, and groups the attributes into families. "Respondents' World Views" (Chapter Seven) details how faculty and students responded to the four world view scales derived from Burrell and Morgan's paradigms while "Respondents' Socio-Demographic Characteristics" (Chapter Eight) categorizes respondents by socio-demographic characteristics. "Rival Hypotheses" (Chapter Nine) show which set of independent variables had the greatest impact on instructors' assessments of student writing: who they were personologically, or who they were situationally. "Fractures in Perspectives on Good Student Writing" (Chapter Ten) takes a closer look at the rival hypotheses by examining  16 the impact of world views and background on faculty and student ranking and grading of the three paragraphs, as well as their scoring of the attributes of writing. The chapter also indicates several distinct and alternative lenses through which to evaluate student writing. "The Revolution Begins with Writing: Conclusions, Discussions and Recommendations" (Chapter Eleven) highlights the implications of the study, and sets a course of action for what English departments and composition researchers could do to improve the teaching of writing and evolve a consensus on student writing quality. Beginning a study of this scope is akin to embarking on a journey filled with potential twists and turns. This uncertainty is oddly fitting, considering the ambiguity and contradiction of "good academic writing." Nevertheless, as any good explorer will attest, challenges can be overcome with the aid of a useful map. "Compositional Collisions" (Chapter Two), with its analysis of the contested terrain of composition pedagogy, aims to fulfill that orienting function.  17 CHAPTER TWO: COMPOSITIONAL COLLISIONS  Academic composition traces its intellectual lineage to the study of rhetoric in ancient Greece, and current incarnation to scholastic developments in the nineteenthcentury. Despite this history, composition remains relatively under-theorized when compared to sociology and psychology, disciplines that also became part of the academic landscape in the nineteenth century. The status of composition studies is problematic. First-year English is a mandatory course in universities and colleges across North America and plays a crucial gate-keeping role. People who teach it are generally regarded as the "poor step-children" of the academy: their presence must be endured, even as they are consistently ignored. Composition instructors are usually underpaid sessional instructors who must teach several hundred students each term. Few composition instructors are trained in composition and rhetoric; instead, most composition instructors have graduate degrees in English Literature but, trapped within a dismal job market, must teach writing for several years before they can even be considered for tenure-track positions. Through all these changes, one constant remains: writing pedagogy is always rooted in ideology (Berlin, 1988). The way writing is taught and assessed is invariably dependent upon a specific set of values and beliefs. The ideologies that have had considerable impact on the practices of writing instructors tend to reflect the epistemological, social and economic trends that have influenced the university as a whole. In this chapter, I examine three highly influential composition pedagogies as well as the epistemological underpinnings shaping them. The three composition pedagogies are Current Traditional Rhetoric, Expressivism and Social Construction. My selection of the three pedagogies was based on the following criteria: 1) each pedagogy had its own distinct philosophical orientation and intellectual lineage; 2) each pedagogy promoted an idiosyncratic form of good academic writing that could be clearly identified as belonging to that particular rhetorical approach; 3) each pedagogy advocated a unique set of writing attributes that were integral to its conception of good academic writing; 4) each pedagogy had a proven historical record of having had a significant impact on the teaching and assessing of first-year English. The pedagogies were categorized as follows: objective rhetoric, subjective rhetoric, and transactional rhetoric (Berlin, 1987). Objective rhetoric locates reality in the material  18 world; subjective rhetoric deems it is a private and personal construct; and transactional rhetoric discovers reality "in the interaction of the features of the rhetorical process itself - in the interaction of material reality, writer, audience, and language" (Berlin, p.155). Within these guidelines, there were three composition pedagogies that most clearly met my criteria: Current Traditional Rhetoric (objective rhetoric), Expressivism (subjective rhetoric), and Social Construction (transactional rhetoric). Before probing these three pedagogies further, I examine the historical progression of first-year English, paying close attention to socio-political developments that have influenced university-level composition courses and the evolution of the university.  Historical Foundations of Composition Studies The pursuit of good academic writing in North America is at least as old as the history of higher education in the United States and Canada. Composition instruction can be traced to seventeenth-century Harvard, and eighteenth-century Yale (established in 1636 and 1701 respectively), and, in Canadian schools, to King's College in Windsor, Nova Scotia (established in 1788). Composition and rhetoric instruction at this stage involved what Bourdieu would call "cultural capital": accuracy of expression, appreciation of great texts, awareness of civic virtues, and possession of logic and an appropriate moral code (Crowley, 1998; Johnson, 1988; Kimball, 1986). Such was the link between rhetorical ability and nobility that early American and Canadian schools exhorted the elegant qualities of the intellectual and cultural elite through an emphasis on classical theory and models, as well as English belletristic texts. It would have been unthinkable for any Canadian hoping to take up a profession or assume a position of social, political or religious influence to do so without a college education in rhetorical arts (Johnson, 1988). Rhetoric's exalted position began to suffer towards the mid-nineteenth century as the classical model experienced an ideological restructuring. North American universities were increasingly enthralled by the research-oriented German university (Rudy, 1951) which promoted academic curiosity, as well as empirically tested knowledge and discovery. Within this new model, rhetoric, with its classical heritage and dependence on morality and mental discipline seemed an awkward fit. There was another discipline waiting in the wings to take over rhetoric's previously privileged position: English language and literature.  19 The ascension of English studies, which drew upon philology in the German model, marked the beginning of a new, intense chapter in the pursuit of "good academic writing" because it required the lack of English language proficiency among undergraduates to justify its standing as an academic discipline. Consider the matter from this angle: why study English at university when everyone entering higher education already possessed (in theory) a sophisticated written and verbal command of the language? Demographic and social changes leading to an influx of non-traditional students (women, minorities and members of the new middle class) provided an unprecedented means of answering this question. In the 1873/74 session, Harvard's English department, under the leadership of Adams Sherman Hill, was the first school to institute a written examination for incoming students. Students were no longer tested on rhetorical skill but expression, mechanics, handwriting, and familiarity with literary texts. Scandalously, about one half of all incoming students failed the test, provoking an academic crisis: what to do with the large numbers of students who had failed the English entrance examination, yet had to be admitted nonetheless? The answer was to establish a mandatory writing course to help rectify mistakes students were making in their writing. Hill introduced this course at Harvard in the 1880s and named it "English A." Other universities across the United States followed suit and "English A" and its titular variants aimed to provide incoming students with literary skills and knowledge. The demographic shift which had such an impact on post-Civil War American universities only took shape several decades later in Canada. Enrollment levels spiked in Anglo- and French-Canadian institutions during the early part of the twentieth century and following the Second World War. In order to cope with the large numbers of incoming students, instructors had to revamp programs and curricula that had focused on moral development through the study of literary texts to concentrate on writing skills. Ultimately, the emphasis on correction, rather than generation of thought, prevailed (Graves, 1994). Over the course of approximately 100 years, good academic writing moved from a philosophical, rhetorical examination to the standardization of linguistic codes Although democratic in the sense it soon became a requirement for all students (except for a select, exempted few) and was the university's response to the democratization of its student body, first-year English soon became firmly implicated in  20 safeguarding a hierarchy of knowledge and expression. Between 1885 and 1910, composition instruction assumed many of the characteristics still associated with the field today, particularly its stress on correctness. This correctness took on both mechanical and formal shapes: four distinct modes of discourse (narration, description, exposition and argument); methods of exposition (process analysis, definition, comparison/contrast, etc.); three levels of discourse (diction, sentence and paragraph); the "narrow-select-develop-outline" invention structure; rhetorical and grammatical sentence types; and such abstract concepts as Unity, Coherence and Emphasis (Connors, 1997). Consequently, the aim of first-year English was (and, some could argue, still is) to promote writing that fits this narrow criteria. The weeding out of "incorrect" writing became more pronounced as universities increased their student populations. A s class sizes grew, instructors had fewer chances to work with students individually but larger marking piles. Consequently, first-year English became even more product-based, with students generally handing in only one draft of their work and instructors focusing primarily on the surface "correctness" of the text. Instructors began using "correction charts" to mark students' work, highlighting errors of style, cohesiveness and usage. The notion of the English teacher obsessively making sure every comma is in its place stems from this period. Thus, the stage was set for the next major phase of writing instruction: Current Traditional Rhetoric.  Current Traditional Rhetoric What does good first-year student writing look like at the start of the twenty-first century? A survey of key handbooks on writing indicates distinctive trends. Writing instruction revolves mechanical correctness n diction, grammar, punctuation and syntax. The management and organization of form and content also feature prominently. A consistent theme is handbooks arranged around discrete textual units: the word, the sentence, the paragraph, and the whole composition. Each element is presented individually and independently of the other. Readers are advised to manage these disparate elements with efficiency and a keen understanding of structure, unity and coherence and thesis statement (see Strunk & White, 2000, 4  t h  edition; Tickle,  1996; Troyka, 2002 for a sampling of writing handbooks used at UBC). Good writing is writing that upholds a set of established rules.  21 The terminology, themes and concerns of these glossy new writing handbooks can be traced back to a nineteenth-century composition pedagogy with epistemological roots in the eighteenth century: Current Traditional Rhetoric. Fogarty (1959) was the first to use the term to describe the present-day traditional form of Aristotelian rhetoric. While the term "Current Traditional Rhetoric" has been criticized for being over-arching and overly generalized (Connors, 1997), it continues to wield considerable influence in North American campuses.  Epistemological Underpinnings Current Traditional Rhetoric traces its genesis to rhetoricians identified with a philosophical orientation known as Scottish common sense realism. These rhetoricians were George Campbell, Hugh Blair, and Richard Whateley, men whose writings had tremendous impact on composition instruction in nineteenth-century US and Canadian colleges. Their work greatly influenced the "big four" of nineteenth-century writing instruction in North America (Adams Sherman Hill, Fred Newton Scott, Barrett Wendell, and John Franklin Genung) and reinforced the pedagogy's hold in North America. Philosophical underpinnings shaping Current Traditional Rhetoric presumed the world to be comprised of information that could be accessed through sensory perception. According to this world view, there was a concrete reality and indisputable truth to be verified through one's senses. This awareness had two significant implications. First, because facts could be proven or disproven through logical reasoning, there was no longer any incentive to focus on emotions, as had been central to the classical rhetorical model of Aristotle and Cicero. Second, knowledge was no longer considered generative but posited as already present in the external world. All scholars interested in pursuing this truth had to do was expose themselves to a wide range of experiences, then examine their memories and sense perceptions objectively and methodically. Any cultural or social context that might distract the scholar from this empirically verifiable truth was to be discarded. The human mind was posited as a passive receiver: reactive rather than formative (Knoblauch and Brannon, 1984). This value system inspired reason's position of prominence within compositional studies and had a profound impact on the field. The primacy of reason undermined language as generative of ideas and passions; language was identified as neutral, unambiguous, divorced from human emotion, and representative of an external reality.  22 The whole point of discourse was not to move human listeners or readers, but to communicate knowledge and make ideas known with ease and expedience to the reader. Primary Features Within the Current Traditional Rhetoric framework, it is imperative for writers to use correct and precise terminology in order to present the substance of their thoughts and ideas accurately. Words must be used correctly to refer to an external, objective world with precision and accuracy. Additionally, the way a person communicates is influenced by the way he or she thinks: a good mind produces good writing while a disorganized, clumsy thinker generates disorganized and clumsy writing. The characteristics of a Current Traditional Rhetoric classroom are as follows: students are instructed on four modes of discourse (narration, description, etc.) and generally expected to produce research or argumentative essays using some of the methods of exposition. They are taught to produce suitable paragraphs and sentences, with particular emphasis on generating appropriate thesis statements or topic sentences. Students are drilled on the importance of producing writing that is unified and clear, and on correct syntax and grammar. They are assigned such tools as writing handbooks and course reading packets (which generally contain essays subscribing to the virtues of clear and organized academic writing). Current Traditional Rhetoric was the dominant composition pedagogy of the twentieth century (Berlin, 1988; Burnham, 2001; Crowley, 1990, 1998). One reason for its continued success is its teachability, especially in crowded lecture halls. Instructors with limited formal training in teaching composition are given institution-sanctioned handbooks which purport to break down writing instruction into sets of simple and manageable formulae. These formulae, in theory, can be presented to students in neat chunks and regurgitated in the form of error-free five-paragraph essays. Nevertheless, the pedagogy has been the target of extensive criticisms.  Criticisms  While Current Traditional Rhetoric suggests an orderly, systematized approach to writing, its suggestion of a "one-size-fits-all" method to teaching writing can easily prove a source of frustration and confusion. Writers and writing come in all shapes and sizes;  23 consequently, the idea that the distinctly heterogeneous student bodies of North America's increasingly multicultural classrooms will subscribe to the narrow tenets of Current Traditional Rhetoric and appreciate the logic of its reasoning, (all within a single semester!) is naive at best. The pedagogy's insistence on surface correctness is problematic for several reasons. First, it implies one distinct and proper way to write, obscuring the slipperiness and rich fluidity of language and sidelining the idiosyncrasy of authorial voice and expression. Second, it suggests consensus over what is "correct" or "incorrect" in writing, overlooking the likelihood that composition instructors, who come from across the broad spectrum of English literary studies, may have varied and irreconcilable opinions about writing. Third, by focusing so closely on the correctness of surface features, the pedagogy overlooks the importance of invention and critical thought in writing (Crowley, 1986, 1998). Fourth, in its insistence on apolitical, neutral correctness, the pedagogy forestalls acknowledgement and investigation of its own dominant ideologies, ultimately producing students unaware of how ideologies have shaped their own use of rhetoric (Herndl, 1993). A common result is a host of frustrated students trying to reconcile their own perspectives on writing with those described in their Current Traditional Rhetoric textbooks and with what they perceive to be their instructors' demands - without even realizing the tangled, thorny issues that lurk beneath the discourse of superficial correctness.  Expressivism Criticisms against the limitations of Current Traditional Rhetoric were particularly pronounced in the nineteen sixties, eventually leading to an alternative way of teaching composition: Expressivism. Expressivism is arguably one of the more controversial of the composition pedagogies, inciting both heated criticism and fervent praise. Such is its divisiveness that even a key Expressivist proponent like Peter Elbow protested the use of the label. "I hate the term Expressivism...[it] tends to connote that I (or Expressivists) are more interested in writing about the self or expressing the self than writing that is trying to be accurate or valid about things outside the self" (quoted in Paley, 2001, p.10). Expressivist pedagogy values the personal and honest expression of the writer. Rather than muffle or disguise the voice of the writer, Expressivists encourage the inclusion of writers into the text they are creating. The traditional notion of the student  24 producing bland, author-vacated prose is displaced. In its stead are students as writers fully engaged in the production of their own writing, articulating and bringing to the academy their background and experiences.  Epistemological Underpinnings  The term "expressive" and its reference to the expression of the self has its roots in the speech studies of Hymes, Jakobson and Sapir, as well as the phenomenology of Merleau-Ponty, Husserl, Satre, and Cassirer (Kinneavy, 1980). While empiricists emphasized objective reality and avoided discussions of subjectivity, phenomenology emphasizes the self as well as the other, thus providing a systematic means for investigating perspectives on individuality and individuals methods of self-expression. Expressive discourse focuses on the "encoder," allowing the speaking self to express and achieve his or her own individuality (Kinneavy, 1980). This individuality, or understanding of the self, Kinneavy argued, is a composite of a Being-for-ltself (the encoder), a Being-for-Others (the decoder), and a Being-in-the World (reality or context). True self-expression can only be achieved when an individual has "an authentic Being-for-itself with an honest recognition and repudiation of his past, a vision of his future projects, an acceptance of his Being-for-Others, and a [sic] unillusioned picture of his Being-in-the-World" (Kinneavy, 1980, p.406). Britton (1970), a crucial proponent of Expressivist pedagogy, expanded the role of the personal in academic writing. Drawing upon phenomenologists, as well as such psychologists as Rogers, Vygotsky and Piaget, he challenged the Current Traditional Rhetoric view of language as a neutral, objective medium. Britton argued that human beings use language as a way of symbolizing, understanding and classifying reality as they have experienced it. The concept of individuals using language accurately and correctly to describe the external world is overly simplistic because it is predicated on the assumption that there is only one "right" interpretation. However, as Britton demonstrated, individuals have different representations of the world in which they inhabit since they all have differing experiences of it. Since human beings are not mechanistic cameras, their ways of representing the world tend to vary. The representations of a single individual may also vary over time, depending on context and changing attitudes.  25 Britton et. al. (1975) identified three different forms of writing: transactional writing, which is used when engaging in everyday affairs (e.g. a business letter, or a progress report); poetic writing, in which the text is an art medium in the form of words with attention paid to the arrangement and formal pattern of the words used (e.g. a poem, novel, or song lyric); and expressive writing. Expressive writing is akin to "thinking aloud on paper" (p.89) and intended for the writer's own use, in the form of diary entries, or personal letters to friends or relatives. In contrast to transactional and poetic writing, which form two extreme ends of the writing continuum, expressive writing mediates the two and provides the central support to their existence: without it, neither transactional or poetic writing could exist. If the demands of a writing task are taken far enough and the need to do something with language becomes more acute, the dominant function of the task changes from expressive to functional. In contrast, when the requirements of the text as construct increase, the writing becomes an immediate end in itself. The form of the language used is vital and the dominant function shifts from expressive to poetic.  Primary Features Expressive language has three general features: first, it is close to the self, revealing and verbalizing the writer's consciousness and displaying close writer-reader relationships; second, it is not necessarily explicit because the writer relies on the reader to interpret what is said based on a shared general context of the past; third, because it submits to the free flow of ideas and feelings, expressive language is relatively unstructured (Britton, et. al., 1975). Expressive writing pays little attention to formal grammatical or stylistic conventions; its focus is personal thoughts and emotions. Because of its fluid, exploratory and implicit nature, expressive writing imposes the least demands upon writers. Writers working within the expressive function can ignore reader expectations, overlook conventions of clarity and precision, and avoid the conventions of linguistic skill. They can focus on generating and ordering their own thoughts, evaluate long-held values, and assess perceptions about prior experiences or even unfamiliar ideas. This is not to suggest that expressive writing is little more than narcissistic solipsism, inciting a babble of words comprehensible only to the writer; instead, Expressivists have argued that writing produced within this function offers writers significant opportunities for ideological,  26 philosophical and logical discovery. More specifically, it enables writers to discern and develop their personal, writerly voices. Expressivists posit student writers as writers with important things to say, and not as students forced to produce bad prose (see, for instance, Elbow, 1973; 1987; Emig, 1971; Graves, 1982; Macrorie, 1970; Murray, 1968). Students are accorded the writerly privilege of expressing their personal and individual voices through their writing. Elbow (1994) likened "voice" to a resonance that "points to the relationship between discourse and the unconscious" (p.xvii). "Voice" acknowledges that while writing can never fully capture the many facets and complexities of an individual, it can resonate with or allude to the writer's unconscious and conscious meaning(s). Voice suggests "energy, humour, individuality...believability" (Murray, 1984, p.144) and enables readers to "feel the pulse" of a writer (Ruszkiewicz, 1981, p.67). It allows writers to subvert the supposed neutrality of language to reveal who they are. According to Expressivist teachers, "good writing" allows students to showcase their writerly voices and represent their own concerns, desires and ways of looking at the world. Such is the importance of individual voice that the focus shifts from the audience to the writer grappling wholeheartedly and enthusiastically with his or her own words and meaning (Elbow, 1987). Rather than produce rule-driven, formulaic essays characterized by lack of authorial voice, writers in an Expressivist classroom are encouraged to create writing reflecting their creativity and unique forms of expression. Recognizing the fluid subjectivity of the writer and writing, Expressivists urge students to incorporate themselves into their work, including points of view, experiences and beliefs. The point is to avoid stale, pretentious, impersonal "Engfish" (Macrorie, 1970) to produce writing that authentically captures who the writer is. The pursuit of this personal truth and its revelation in the text is the goal of Expressivist writing (Yancey, 1994). Criticisms One criticism frequently leveled against Expressivism is the pedagogy's sidelining of grammar and mechanics. The fear is that Expressivist instructors, fascinated as they are with the development of idiosyncratic voices, fail to provide students with an appropriate linguistic background, condemning them to a lifetime of grammatical error making. Yet another concern deals with what is perceived as Expressivism's lack of  27 intellectual rigour (Young, 1978). This criticism is rooted in Expressivism's concentration on the student-as-writer: the reading of students' work as literary texts is seen as problematic because it can prevent instructors from critiquing matters of organization and clarity. Some Expressivist instructors use only student essays, not prose models as required readings in writing courses. A s Gradin (1995) pointed out, the focus on student writing is linked to "romanticism," a system of thought and selfexpression that seems at odds with a rational, more logical-minded rhetoric. Romanticism's self-absorbed, overwrought and untrammeled self-expression is directly at odds with academia' dispassionate and objective reputation. Consequently, students trained in Expressivist classrooms might be poorly prepared to produce acceptable academic writing.  Social Construction Criticisms against Expressivism have come from Social Construction, the third and final composition pedagogy this study will examine. Their critique centres around two broad themes: first, the idea of the self as a unified, knowable entity; and, second, what they assume to be the failure of Expressivist instructors to acknowledge the impact of social context in their classrooms. According to Social Construction, the "personal" is a highly problematic topic of discourse particularly since, as postmodernists have repeatedly asserted, "selfhood" is constantly in a state of flux and cannot be presented as coherent and complete (Faigley, 1992). How can writing instructors expect students to write about themselves as distinct and unique when that self is a contradiction? This leads to the second criticism: that Expressivism focuses on the self to such extent it overlooks the dialectical relationship between writers, their communities, and socio-economic and political aspects of their existence (Berlin, 1988; Lefevre, 1987). Negating the social context in which writers operate can have thorny consequences when instructors fail to make students aware of the exacting and particular standards that shape academic discourse (Bartholomae, 1995; Bizzell, 1982). Expressivist pedagogy, from a Social Construction view, is in danger of positing individuals as separate from or above society's influences, rather than a product of social relations.  28 Epistemological Underpinnings Social Construction draws upon post-modern perspectives on discourse, language and communal interaction espoused by Bakhtin, Foucault, Kuhn and Vygotsky. It also draws heavily upon Rorty's anti-foundationalism, a philosophical orientation that advocates awareness of how understanding is shaped by context. Social Construction takes as its starting point an understanding of knowledge that diverges quite radically from the objectivist paradigm favoured by Current Traditional Rhetoric, as well as the phenomenology of Expressivism. Where the former argues that knowledge is the result of an unfiltered view of reality, and the latter situates truth in the individual, Social Construction champions viewing the world through a social filter. The prevailing ideologies and conceptions guiding human behaviour and understanding are social constructs which result from communal consensus. Social Construction rests on several principles. First, "reality" can only be ascertained through language and social interactions (Berthoff, 1981). The individual "never responds to things in themselves but to discursive formations of things in themselves" (Berlin, 1993, p. 108). Second, because language is the product of social interaction, it is never neutral or transparent but invested with meaning. Third, knowledge is not based on objective truth but generated out of language. "Concepts, ideas, theories, the world, reality, and facts are all language constructs generated by knowledge communities and used by them to maintain community coherence" (Bruffee, 1986, p.777). Without language, itself a product of communal interaction, there can be no knowledge. Fourth, knowledge is always situated in a particular context or situation. Thus, close attention must be paid to "local knowledge" (Geertz, 1983) since antifoundationalism asserts that fact, truth and correctness are "intelligible and debatable only within the precincts of the contexts of situations or paradigms or communities that give them their local and changeable shape (Fish, 1989, p.344). The politicized arm of Social Construction is Social Epistemic and argues for a close investigation into relationships among power, discourse and language (Berlin, 1988). Social Epistemic relies upon the radical pedagogy of Giroux, Shor and Freire, as well as cultural studies done by the Birmingham School. It perceives discourse as ideology which is understood as "transmitted through language practices that are always at the center of conflict and contest" (Berlin, 1988, p.478). Discourse and language are  29 inextricably linked to issues of power and cannot be understood as discrete entities. Social Epistemic, however, does not limit itself to theory but embraces a practical end. Thus, investigations into the social construction of language and knowledge must result in political action: students must learn to critique the power dynamics that have influenced the construction of significant texts (Fitts & France, 1995), and become "their own agents for social change" (Shor, 1980, p.48). Primary Features The Social Construction writing classroom is characterized by the following features: students must have mastery over academic discourse and participate effectively, or "work for change with some power" (Bizzell, 1992, p. 150), in their discourse communities. Poor student writing is produced by academic novices, i.e. individuals who have had little exposure to academic discourse. Thus, in order to facilitate the production of good writing, students should be made familiar with academic conventions. Social Constructionists conceive composing as a socialization process (Baardman, 1994; Shaughnessy, 1977). Student writers are least successful when most ignorant of academic discourse conventions. However, as beginning writers learn discourse conventions readers expect, they begin to adopt their readers' ways of thinking, values and world-view. The writing challenge beginning students face is posited as follows: each time students have to produce an academic paper, they have to "invent" the university, i.e. learn to speak the language of the academy and appropriate its discursive conventions (or pretend that they do) (Bartholomae, 1988). In order to write, all writers must imagine themselves as privileged "insiders"; paradoxically, they must assume privilege without actually having any. They must learn the "commonplaces, set phrases, rituals, gestures, habits of mind, tricks of persuasion, obligatory conclusions, and necessary connections that determine the 'what might be said' and constitute knowledge within the various branches of our academic community" (Bartholomae, 1988, p.278) in order to develop this necessary privilege. Social Constructionists claim students can develop this discursive knowledge through collaborative learning, particularly participation in peer-groups (Bruffee, 1983) which allow members to arrive at collective decisions about meaning through consensus or negotiation. Another strategy is providing students tools for understanding the thinking that goes behind academic discourse, and drawing attention  30 to how discourse conventions reveal how scholars define and interact with the world (Maimon et. al., 1989). The politicized form of Social Construction, however, takes this understanding a step further to urge critical analysis of how discourse is implicated in maintaining power structures. The goal is to make students aware of "the cultural codes - the various competing discourses - that attempt to influence who they are" (Berlin, 1991, p.50). The composition instructor's larger purpose is "to encourage our students to resist and negotiate these codes - these hegemonic discourses - in order to bring about more personally humane and socially equitable economic and political arrangements" (Berlin, 1991, p.50). Students must display awareness of the socio-political structures that shape their cultural and academic discourses, and be prepared to analyze these institutions with a criticality that will spur further action. Criticisms Critics of Social Construction allege the pedagogy's emphasis on consensus and social agreement obscures the fact this so-called consensus may negate personal agency through its promotion of social agreement over individual expression (Flower,1994; Stewart, 1988). The sidelining of the individual for the group is problematic when it results in the marginalizing of dissenting or minority voices. Rather than learn how to become active and critical participants in discourse communities, students might be pressured to abandon their forms of expression and adopt communicative strategies which privileged and powerful members of the academy have validated. Consequently, structures of power remain entrenched and unchanged. Another issue is the pedagogy's exclusive concern with discourse as the mediator of knowledge. This obsession fails to acknowledge the knowledge-making potential of entities other than the linguistic or symbolic (Petraglia, 1991). The pedagogy's fairly narrow focus, particularly on the relationship between discourse and power, is in danger of translating into a form of arrogance. Hairston (1992) denounced the practice of composition instructors who advocate politicized theory or social change in the classroom as "radical," "regressive," "silly," "simplistic" and basically coercive (p. 183). Additionally, teachers who insist on the moral superiority of radical theory threaten to impose new authority on their students (Gale, 1996) n ironic for a pedagogy that aims to challenge and dismantle power structures in aid of a new democracy. Finally, there is  31 the danger of radical pedagogy being co-opted as a do-or-die academic trend adopted by academics who champion arcane theory just so they can remain cutting-edge. Such practice risks disregarding or diSDaining students' welfare (Spellmeyer, 1996).  Conclusion Composition is a field rich with rival and ideologically contrasting pedagogies. Each pedagogy comes with its own epistemological foundation and firm convictions, revealing divisive opinions even within the field. Drawing upon the three pedagogies, I devised a questionnaire that would facilitate the documentation of fractures in perspectives on good writing among and between instructors and students. Chapter Three ("Procedures") describes steps taken to collect data from study respondents.  32 CHAPTER THREE: PROCEDURES The central aim of this study was to document fractures in instructors' and students' perspectives on good academic writing, and to determine if assessments of good writing were more powerfully shaped by respondents' personological or situational characteristics. Specific objectives were to: 1) ascertain, through the use of three sample texts, each representing one of the three commonly recognized composition pedagogies, what writing style respondents most commonly identified with good firstyear student writing; 2) investigate instructors' and students' assessments of various attributes of first-year student writing; 3) determine if there were fractures among instructors and between instructors and students in relation to assessments of first-year student writing; 4) examine if fractures were shaped by who instructors were personologically, versus who they were situationally; 5) identify points at which these fractures were most significant. In order to achieve these objectives, I created a questionnaire comprised of four sections: "Sample Student Writing," "Attributes of Writing," "Differing World Views" and personal and academic "Background." In this chapter, I explain the rationale for the questionnaire as a whole, and steps taken to collect data.  Rationale for the Questionnaire Because the study would necessitate going into classrooms and taking up valuable instruction time, it was vital to keep the questionnaire and its instructions concise. Both instructors and students would be invited to participate in this study. In order to gather comparable responses from both instructors and students, I had to make sure the instructor and student versions of the questionnaires were nearly identical. This meant a fine balance between making instructions and word choices simple enough for beginning students, yet not so simplistic instructors would find them puerile. I tested various versions of the questionnaire informally among first-year U B C students with fairly extensive E S L difficulties, as well as faculty members across the university willing to participate. After several drafts, I was able to arrive at a questionnaire that was understandable yet comprehensive.  33 Procedures for Data Collection I conducted the study at the University of British Columbia (UBC) for several reasons. First, from a logistical perspective, U B C is the Canadian university with which I am most familiar. Having spent several years here teaching first-year writing to hundreds of students in both U B C ' s English Department and Writing Centre, I am well versed in the nature of first-year English instruction offered at this university, and have access to students and instructors. Second, U B C is a metaphorical research gold mine. The university offers hundreds of academic programs, and caters to almost 40,000 students from 120 countries (University of British Columbia, Welcome, 2006). Thirty three thousand of the 40,000-strong student body are undergraduates, all of whom are expected to have at least six credits of first-year English. Third, U B C was ranked by Newsweek magazine as one of the world's 50 global universities (the highest ranking Canadian school on the list), i.e. institutions characterized by openness, diversity and distinction in research (Newsweek, 2006). I contacted potential respondents by going to the home pages of every department in the Faculty of Arts, the Department of Language and Literacy, and the Technical Communication program and acquiring the names and email addresses of faculty members. I sent emails informing instructors of my study (Appendix A). If they agreed to participate, I sent a copy of the questionnaire (Appendix B), either through campus mail or email attachment, or dropped off hard copies in their mailboxes or offices. I also sent all respondents my campus mailing address so they could return the questionnaire by campus mail. Additionally, with the permission of the Head of the English Department, I set up a box in the department for respondents to leave their completed questionnaires. Overall, I sent out 504 emails to the departments listed above. One hundred and sixty nine instructors contacted (33.5%) agreed to participate in the study. I received 160 filled-out questionnaires in return (94.7% of the total questionnaires distributed). Of these, three questionnaires could not be used because respondents had not completed at least two sections of the study, leaving 157 questionnaires (31.2% of the number of instructors who were initially emailed). In retrospect, I should have emailed a tillable pdf version of the questionnaire to all eligible instructors or left copies in every single faculty mail-box. This might have increased the rate of return among instructors. Nevertheless, one could argue instructors who consented to and eventually participated in the study  34 were interested in and thoughtful about problems student writers face. Thus, their responses were particularly informative. I also targeted first-year students enrolled in first-year English courses. In order to gain access to these students, I emailed instructors of first-year English asking if I could attend their classes to distribute questionnaires. I contacted all instructors teaching first-year English in the winter term of 2004/2005. Ten instructors allowed me into their classes, giving me access to 21 different sections of first-year English and several hundred students. The students were enrolled in one of two courses: English 112 ("Strategies for University Writing") which teaches students to apply principles of university-level discourse to their writing, and English 110 ("Approaches in Literature") which introduces students to the fundamentals of university-level literary analysis. I was also invited to attend three sections of Writing 098, "Introduction to Academic Writing", a course offered by U B C ' s Writing Centre. Writing 098 is designed for first-year students who have not yet passed the Language Proficiency Index, a gate-keeping examination all U B C students who have not scored an "A" grade in their provincial-level English examination must pass before they are allowed to enroll in first-year English courses. I attended 18 of the 42 first-year English 112 sections in person. In each class, I explained the study and distributed questionnaires. Students were given 15 to 20 minutes to complete the questionnaire and I was available throughout the time if they needed a word clarified or an instruction explained. I did not personally attend three sections whose instructors preferred to distribute the questionnaires. Two of these instructors returned the completed questionnaires to me via campus mail. I collected the third set of completed questionnaires from the instructor in question personally and was able to talk to her about how the distribution process went. All completed questionnaires were stored in a locked filing cabinet. I was the only person with access to these files. Data entry was conducted with the aid of S P S S version 11.1 transformed all data into numerical codes and used various statistical procedures including Analysis of Variance, mean scores, correlations, varimax rotations, and dendograms. The procedures discussed in this chapter enabled me to collect data documenting the fractures in perspectives on good writing among instructors and between instructors and students. I discuss these results elsewhere. "Instrument Development" (Chapter Four), however, tests the various fracture and predictor indices.  35 C H A P T E R FOUR: INSTRUMENT D E V E L O P M E N T  The instrument used in this study was a questionnaire consisting of four sections: "Sample Student Writing," "Attributes of Writing," "World Views," and "Background." The questionnaire's overall purpose was to document fractures in instructors' and students' perspectives on good academic writing, and identify the world views and sociodemographic characteristics of respondents who endorsed one view of good writing over another. It consisted of the following scales of measurement: 1) a scale to gauge how respondents ranked the three sample paragraphs; 2) a scale to gauge how respondents graded the three sample paragraphs; 3) a metric for measuring how respondents scored 45 attributes of writing; 4) four indices, one for each world view, that gauged how respondents perceived the world around them. Two sections in the study, "Attributes of Writing" and "World Views," were created especially for this study and culled from literature on composition pedagogies and Burrell and Morgan's paradigms respectively. Consequently, the scales used in those sections needed to undergo the following tests to ensure their validity: •  Attributes of Writing: 1) correlation between each attribute and the pedagogy from which it was derived; 2) convergent and discriminant validity of the attributes collectively as whole scales of measurement.  •  World Views: 1) convergent and discriminant validity of the four indices to determine the correlation between each adjective and the world view from which it had been derived; 2) factor validity of the adjectives within each of the four world view indices.  In this chapter I explain the rationale for the development of each section of the instrument used in this study. I tested each section of the study as part of a series of pilot studies. I describe the findings of these pilot tests and subsequent tweaks and reshaping of the various sections below.  36 Section A: Sample Student Writing One way of measuring fractures was to have respondents rank and grade writing samples. Consequently, "Sample Student Writing" comprised three sample opening paragraphs for a take-home paper for first-year English, each in the style of one of the three composition pedagogies used in this study. Sample paragraphs were used for several reasons. First, they made tangible three different styles of writing endorsed by the three pedagogies. Second, they provided a yardstick for measuring how respondents assessed these different writing styles. Respondents were invited to rank each paragraph either "top," "middle" or "top." They were instructed not to give any two paragraphs the same rank. Respondents were also asked to give each paragraph a grade from zero to ten. The challenge was writing three opening paragraphs that would capture the essential flavour of the three pedagogies - Current Traditional Rhetoric, Expressivism and Social Construction. There were several factors to consider. First, the paragraphs had to be relatively short, of about 100 words each. Second, because the study dealt with first-year student writing, the paragraphs had to be written in the manner of a middling to-slightly above average entry-level student. Thus, the paragraphs could not be overly polished in terms of thought or language. Third, in order to maintain consistency, it was necessary to pick one topic that could fit the three differing writing styles. The topic had to be innocuous, lest its substance detract readers from style, and needed to be relevant to a first-year English class. With these stipulations in mind, I settled on "The Importance of a University Education." Finally, there were matters of grammar, punctuation and syntax. While Current Traditional Rhetoric is the pedagogy most closely associated with "correct language", I did not want mangled syntax or unclear meaning to form unnecessarily misleading barriers for respondents or cause them to rank other paragraphs lower because of mispunctuation. Thus, I kept the syntax relatively error-free. The paragraphs underwent several drafts and were tested among members of my thesis committee. Words were changed or omitted if they were unclear or awkward. Each paragraph had a set of defining features. The Current Traditional Rhetoric paragraph adheres to the "formula" of an opening paragraph mandated by scores of writing textbooks. It begins with a brief nod to the issue at hand: students entering universities in Canada, and sacrifices such a decision entails. Each sentence flows  37 smoothly to the next with few transitional glitches. The paragraph wastes little time in setting up a frame for a thesis statement systematically presented in three parts. The paragraph ends with a clear thesis, indicating the writer will have little difficulty moving on the body paragraphs of the essay. The language is carefully neutral, without allusion to potentially problematic words or themes, and its writer's voice is factual, orderly and free of expression or emotion. In contrast, the author's persona is an integral part of the Expressivist paragraph and enters as early as the second word of the text. The paragraph establishes various details of this persona: the writer's geographical background, family economic situation, and historical connection (or lack thereof) with higher education. The author's insight stems from personal and emotionally driven reflection, and the paragraph makes clear the essay's frame of reference will be the author herself. Thus, despite the thesis-like statement at the end of the paragraph, the implication is that the importance of a university education can only be ascertained according to the writer's private understanding of the concept, and proven ordisproven based on personal experience. The Social Construction paragraph begins by demonstrating the author's familiarity with current social issues, claiming a university degree is no longer a passport to a comfortable middle-class life. It displays an understanding of the contradiction surrounding the pursuit of higher education and challenges accepted dogma about the value of a college diploma. The paragraph raises doubts about such generally accepted words as "importance" and "university education," implying that meaning is merely an arbitrary concept requiring careful analysis. The Social Construction sample ends, not with a straightforward, formulaic thesis statement, but with a question, indicating a certain comfort level with ambiguity and the negotiating of meaning. Finally, the tone of this paragraph is challenging and almost confrontational. (The three paragraphs are in Appendix B, "Questionnaire.") Once the wording of the paragraphs was settled, the paragraphs were tested informally among my thesis committee, two writing instructors, and ten Writing Centre students for whom English was a second language. Respondents of this pilot test were asked to grade the paragraphs anywhere from zero to ten. They were also invited to comment upon the directions provided to them. While respondents commented favourably on the clarity of the paragraphs and instructions, at least two respondents noted that even though they were generally inclined towards one writing style, they did  38 not necessarily grade that style highly. Grading practices and personal preferences did not always coexist harmoniously. Consequently, I instructed respondents to rank as well as grade the paragraphs to determine if respondents were generally consistent in their ranking and grading of the paragraphs. In other words, did respondents assign a correspondingly high grade to a paragraph that they liked? These findings are presented in "Responses to Sample Student Writing" (Chapter Five). Section B: Attributes of Writing The second way of measuring fractures was to have respondents endorse attributes of good writing. "Attributes of Writing" consisted of 45 writing attributes taken from scholarship on the composition pedagogies around which this study revolves. Each attribute has been associated with Current Traditional Rhetoric, Expressivism or Social Construction. Attributes were included in this study because they facilitate a vital jump in investigations of good writing from the pages of scholarly discourse to the business of marking student writing. They concretize and provide comprehensive terminology for nebulous or complex writing features, thus providing an accessible gauge for the qualities respondents associate with good writing. In this section of the study, respondents were asked to score the attributes from "not at all important" to "critically important." Thus, useful information could be gleaned about the importance respondents accorded the attributes, as well as whether they grouped them in clusters. A key challenge lay in selecting which attributes to use. Originally, I had a list of about 60 attributes, more than half related to Current Traditional Rhetoric. The preponderance of Current Traditional Rhetoric attributes was due to this pedagogy's enormous weight and influence in composition circles. Due to this extensive history, and the infinite number of Current Traditional-informed workbooks, there was a large number of attributes from which to choose. However, including all these attributes could lead to two immediate problems. First, a disproportionate number of Current Traditional Rhetoric items could overwhelm Social Construction and Expressivist-related attributes and damage the validity of measuring tools. The second problem dealt with varying levels of importance. While correct use of tenses or modifiers is undeniably significant within a Current Traditional Rhetoric milieu, it may not carry the same heft as effective use of transition words, or paragraph unity. In other words, the inclusion of many minor attributes might dilute the presence of key Current Traditional Rhetoric exemplars.  39 An informal first step involved parsing 60 attributes with members of my thesis committee, one of whom is a noted professor in the field of composition research. With their help, I trimmed the attributes down to an initial 30, ten from each pedagogy. The initial 30 pilot test attributes are listed in Table 1.  Table 1: Initial 30 Pilot Test Attributes According to Pedagogy Current Traditional Rhetoric Command of grammatical rules  Expressivism  Social Construction  Inclusion of author's voice  Critical investigation of issues  Command of punctuation and spelling rules  Acknowledgement of author's subjectivity  Ability to approximate academic discourse  Coherence of essay  Honesty of author's voice  Attention to reader  Adherence to syntax rules  Evidence of personal insight  Ability to conduct academic inquiries  Structure and development of thesis statement  Use of narrative techniques  Contribution to scholarship  Placement of main ideas  Inclusion of author's experiences  Awareness of academic standards  Sophistication of vocabulary  Incorporation of personal pronouns  Use of academic rhetorical strategies  Clarity of sentences and paragraphs  Ability to express personal truths  Familiarity with related scholarship  Structure and development of topic sentences  Acknowledgement of author's positionality  Ability to critique related scholarship  Unity of paragraphs  Authenticity of author's voice  Acknowledgement of the social construction of language  Pilot Testing the Attributes of Writing I conducted a first pilot test by randomly approaching 30 faculty members from such diverse programs as Educational Studies, Social Work, History, Law, Zoology, Film, Economics, Commerce, Chemistry, Language Education, Medicine, Educational Psychology, Philosophy, Rehabilitation Sciences, English and Nursing. Respondents were asked the following question: "When marking your students' writing, how important is each of the following attributes?" They were presented with five options for each  40 attribute: "Not at all Important," "Slightly Important," "Moderately Important," "Very Important," and "Critically Important." I was present while each respondent scored the attributes. I was guided by two questions. First, would respondents understand the attributes? Second, would respondents score all or most of the items "Very Important" or "Extremely Important," thus failing to show any differential selectivity to the 30-item construct under investigation? There were some useful comments. A few respondents objected to the yoking together of "Punctuation" and "Spelling" since these are two distinct items. Thus, I separated them in the later version. Others felt "Contribution to Scholarship" and "Ability to Critique Related Scholarship" would be beyond the reach of most first-year students. "Ability to Approximate Academic Discourse" and "Acknowledgement of the Social Construction of Language" drew blank stares. A s one respondent pointed out, "The writing is either academic discourse, or isn't. Why the need to 'approximate it?'" I removed the first three attributes from the final questionnaire and replaced the fourth with the more readable "Examination of Society's Influence on Language." "Acknowledgement of Author's Subjectivity" and "Acknowledgement of Author's Positionality" also evoked raised eyebrows, with respondents commenting on the abstractness of these phrases. I replaced them with the relatively more concrete "Acknowledgement of Author's Biases," "Author's Involvement with the Subject" and "Development of Author's Point-of-View." Finally, respondents objected to "Use of Narrative Techniques" since narration is rarely used in more expository-driven genres of student writing. Consequently, I replaced it with items that, though rooted in artistic expression, can appear in academic writing: "Use of Imagery" and "Use of Metaphors." Table 2 presents the 30 respondents' assessments of the importance of these 30 pilot attributes in descending order. These scores were determined after respondents' scores had been translated numerically.  "Not at all Important" was coded one point;  "Slightly Important" two; "Moderately Important" three; "Very Important" four; and "Critically Important" five points.  41 Table 2: Importance of Thirty Pilot Test Attributes in Descending Order  4.57 4.50 4.43 4.17 3.96 3.95 3.90 3.90 3.87 3.87 3.83 3.82 3.80 3.77 3.73 3.55 3.55 3.51 3.40 3.37 3.33 3.11 3.04 2.98 2.91  Standard Deviation .57 .57 .57 .87 .89 .90 .99 .91 1.04 .82 .91 .83 1.06 .77 .83 1.12 .97 1.13 1.06 .89 .76 1.24 1.22 1.32 1.37  Score Range 3-5 3-5 3-5 2-5 2-5 1-5 1-5 2-5 1-5 2-5 2-5 2-5 1-5 1-5 2-5 1-5 1-5 1-5 1-5 2-5 2-5 1-5 1-5 1-5 1-5  2.87 2.75 2.56 2.51 2.30  .90 1.03 1.22 1.26 1.26  1-5 1-5 1-5 1-5 1-5  Attribute  Mean  Critical investigation of issues Clarity of sentences and paragraphs Coherence of essay Structure and development of thesis statement Unity of paragraphs Ability to conduct academic inquiries Familiarity with related scholarship Awareness of academic standards Evidence of personal insight Command of grammatical rules Structure and development of topic sentences Placement of main ideas Ability to critique related scholarship Adherence to syntax rules Command of punctuation and spelling rules Honesty of author's voice Attention to reader Acknowledgement of author's subjectivity Contribution to scholarship Ability to approximate academic discourse Sophistication of vocabulary Inclusion of author's voice Authenticity of author's voice Acknowledgement of author's positionality Acknowledgement of the social construction of language Use of narrative techniques Use of academic rhetorical strategies Ability to express personal truths Inclusion of author's experiences Incorporation of personal pronouns  There was a considerable gap (4.57-2.30=2.27) between items with the highest and lowest mean importance estimates ("Critical Investigation of Issues" and "Incorporation of Personal Pronouns," respectively). Thus, there was a distinct gradation in respondents' assessment of the importance of these 30 different attributes. A s the score range and standard deviation (SD) columns demonstrate, there were diverse opinions about how important the attributes were with most items receiving scores that could range anywhere from "Not at all Important" to "Extremely Important." Finally, the S D s suggest greater consensus about the importance of some attributes than about others. In general, attributes with high importance means had low S D s , indicating greater consensus about their importance.  Conversely, attributes with low importance means  42 but higher S D s indicated disagreements among instructors about how important each attribute was in student writing. These findings suggested the 30 attributes spanned between half and two-thirds of the spectrum of possible importance (2.27/5-1=0.57) and could be used to discern conflicting opinions about student writing as instructors made differentially critical judgments about the importance of each one.  Final Version of Attributes of Writing I extended the number of attributes from 30 to 45 to balance the number of attributes representing each of the three pedagogies, do some justice to the extensive detail of Current Traditional Rhetoric attributes, and include items that dealt more specifically with grammar (e.g. "Correct Use of Articles" and "Correct Use of Prepositions") and coherence ("Effective Use of Transition Words and "Links between Paragraphs"). I also expanded on attributes dealing with the socially astute, occasionally radical nature of Social Construction by including items like "Acknowledgement of Social Context" and "Critical Analysis of Current Events." One concern was conveying the complex ideas behind Social Construction and Expressivism as succinctly worded writing attributes. The Social Construction attributes required several drafts because the pedagogy draws upon post-structuralist and postmodernist thought and often expresses itself in multifaceted discourses. Another consideration was the lack of familiarity respondents would have with the attributes. In the pilot-testing round, a senior instructor of composition suggested providing context to the attributes, i.e. categorizing them under separate headings. Classifying the attributes according to their respective pedagogies was problematic. Such headings as "Current Traditional Rhetoric" and "Social Construction" might have been off-putting for respondents who had never heard of these terms, and would reveal too much about the study. Second, while each of these attributes has been claimed by one of the three pedagogies, the objective was to determine if an empirical and logical case could be made for such categorizations. Subsequently, the attributes were grouped using headers that would make sense to anyone familiar with academic writing: "Content," Mechanics," and "Style." One useful consequence of this categorization was an unusual (for academic circles) attribute like "Inclusion of Author's Voice" could be listed under "Content" and "Style," offering an  43 additional way of determining how instructors responded to a potentially problematic variable. Finally, the attributes had to be discernable in text form. An early draft included attributes like "Awareness of the social nature of knowledge" and "Concern for social reform." However, instructors have no accurate way of determining whether firstyear students have this understanding just from reading their papers. Thus, these attributes were amended as "Investigation of Society's Impact on Knowledge" and "Call for Social Reform". The 45 attributes in the final version of the questionnaire are below.  Table 3: Final Version of the Writing Attributes Categorized According to Pedagogy Current Traditional Rhetoric (n=18) Adherence to syntax rules  Expressivist (n-12) Acknowledgement of author's biases  Clarity of sentences and paragraphs  Inclusion of author's background  Call for social reform  Coherence of essay  Inclusion of author's experiences  Critical analysis of current events  Command of grammatical rules  Inclusion of author's thoughts and feelings  Discussion of world events  Command of punctuation rules  Inclusion of author's voice  Examination of society's influence on language  Command of spelling rules  Inclusion of personal truths  Familiarity with related scholarship  Correct use of articles  Author's involvement with the subject  Investigation of power structures  Correct use of prepositions  Development of author's pointof-view  Investigation of society's impact on knowledge  Effective use of transition words  Honesty of author's voice  Links between paragraphs  Incorporation of personal pronouns  Placement of main ideas  Use of imagery  Precision of word choice  Use of metaphors  Social Construction (n=14) Acknowledgement of social context  Evidence of academic inquiry Evidence of critical thinking Attention to reader Faithfulness to academic standards  Proof of organization  Use of academic discourse conventions  Sophistication of vocabulary  Use of academic rhetorical strategies  Structure of thesis statement Structure of topic sentences Unity of paragraphs Variety of sentence structures  44 Validity of the Attributes of Writing In order to test the reliability of the attributes of writing as scales of measurement for the three pedagogies, the following questions were investigated. First, which attributes correlated most or least strongly in terms of importance with their respective pedagogies? Second, were there attributes whose importance correlated better with pedagogies that were not traditionally their own, and, if so, what were these attributes, and with which pedagogy did their importance exhibit a stronger quantitative relationship? Third, what did correlations reveal about the importance instructors and students assigned the 45 attributes individually compared to their respective pedagogies collectively? Table 4 provides answers. The table categorizes the attributes by the pedagogies with which they are most closely associated in composition studies in descending order of correlation of importance. The findings presented were obtained during the final phase of data collection and involved 157 faculty members and 523 students. Asterisks represent significant statistical correlations: ** at the .001 level and * at the .01 level. Finally, "Level" refers to whether respondents were instructors or students. Students were given the numerical code "1" and instructors were coded "2." Positive correlations indicate attributes to which instructors attached greater importance while negative ones indicate that students attached greater importance. The table's rightmost column shows how strongly each attribute' importance correlated with the respondent's level; either instructor or student. Both groups tended to assign similar importance ratings with the exception of several items. These dissimilarities will be discussed further in "Responses to the Attributes of Writing" (Chapter Six).  45 T a b l e 4: C o r r e l a t i o n s b e t w e e n I m p o r t a n c e R e s p o n d e n t s A s s i g n e d W r i t i n g A t t r i b u t e s a n d C o m p o s i t i o n P e d a g o g i e s  PEDAGOGY/ATTRIBUTE  CTR  Current Traditional Rhetoric  Command of Punctuation Rules Correct Use of Prepositions Command of Spelling Rules Correct Use of Articles Command of Grammatical Rules Links between Paragraphs Effective Use of Transition Words Adherence to Syntax Rules Unity of Paragraphs Clarity of Sentences and Paragraphs Coherence of Essay Structure of Topic Sentences Variety of Sentence Structures Precision of Word Choice Proof of Organization Placement of Main Ideas Sophistication of Vocabulary Structure of Thesis Statement  EXP SCON Level  .72** .68** .66** .65** .64** .63** .61** .60** .60** .55** .55** .53** .53** .52** .52** .50** .50** .48**  .03 .10 .07 .08 .08 .13* .10 .09 .17** .03 .03 .13** .24** .15** .07 .11* .16** .11*  13** 13** 12* 15** 19** 18** 11* 20** 24** 22** 21** '20** 16** 21** 18** 13* 19** 14**  .08 .03 .12* .03 .03 .01 .04 .14** -.02 .08 .13** .04 .20** .02 .17** -.05 -.09 -.02  .02 .05 .19** .04 .01 .13** -.04 .23** .19** .25* .11* .12** .13**  .71** .67** .66** .64** .62** .62** .61** .58** .55** .54** .54** .52** .39**  .25** .14** .24** .26** .21** .32** .20** .31** .28** .27** .25** .20** .32**  -.26** -.21** . 14** -.20** -.08 -.18** -.34** -.11* -.05 -.03 -.13** -.12* -.13**  .17** .02 .09 .08 .01 .05 .01 .21** .25** .15** .24** .22** .35** .24**  27** .24** .27** .31** .27** .27** .36** .17** .05 .11* .24** .03 .09 .37**  .47** .59** .59** .56** .53** .53** .48** .46** .45** 44** .44** .37** .34** .33**  -.09 -.11* .18** .16 .24** .09 .35** .08 .27** .24** .18** .26** .25** .20**  Expressivist  EXP/ Inclusion of Author's Experiences Inclusion of Author's Thoughts/Feelings Inclusion of Author's Voice (Style) Inclusion of Author's Background Inclusion of Author's Voice(Content) Incorporation of Personal Pronouns Inclusion of Personal Truths Author's Involvement with the Subject Honesty of Author's Voice Development of Author's Point-of View Use of Imagery Use of Metaphors Acknowledqement of Author's Biases Social Construction  Acknowledgement of Social Context Critical Analysis of Current Events Investgtn. of Society's Impact on Knowledge Examntn. of Society's Influence on Language Discussion of World Events Investigation of Power Structures Call for Social Reform Use of Academic Discourse Conventions Evidence of Academic Inquiry Familiarity with Related Scholarship Use of Academic Rhetorical Strategies Evidence of Critical Thinking Faithfulness to Academic Standards Attention to Reader T h e  i m p o r t a n c e  o f e a c h  v e r y s t r o n g l y w i t h t h e s c a l e d  o f t h e 1 8 C u r r e n t T r a d i t i o n a l R h e t o r i c a t t r i b u t e s c o r r e l a t e d i m p o r t a n c e  o f C u r r e n t T r a d i t i o n a l o v e r a l l .  T h e  h i g h e s t  46 correlating items were "Command of Punctuation Rules" (.72**), "Correct Use of Prepositions" (.68**), "Command of Spelling Rules" (.66**), "Correct Use of Articles" (.65**) and "Command of Grammatical Rules" (.64**). These five items are the most grammatically oriented of the Current Traditional Rhetoric attributes, dealing primarily with unambiguous word choice, punctuation and syntax accuracy. A student's punctuation or spelling is either correct or incorrect. Similarly, a student has either used or not used prepositions properly. These rule-based attributes require memorization and precise application rather than critical questioning. The five items with the lowest correlations of importance within Current Traditional were more subjective in nature. They were, in descending order: "Precision of Word Choice" (.52**), "Proof of Organization" (.52**), "Placement of Main Ideas" (.50**), "Sophistication of Vocabulary" (.50**) and "Structure of Thesis Statement" (.48**). Unlike their higher correlating counterparts, these attributes go beyond purely surfacelevel features to evaluate how writers have grappled with their material to communicate the substance of their opinions. The Current Traditional attributes all correlated much higher with their own pedagogy than with either Expressivism or Social Construction. While there were a few statistically significant correlations that crossed pedagogies (which is to be expected, given the large number of participants in this study), none of these scores exceeded .24. The implication is that the Current Traditional total, fictive though it is, agreed closely with each of the 18 elements which made it up better than with any of the 13 or 14 elements that are purported to make up Expressivism or Social Constructivism. Twelve Expressivist attributes were featured in this study, one of which appeared twice in two different portions of the questionnaire, making a total of 13. A presumed Expressivist importance total was calculated by summing each person's responses to those 13. The items with the five highest correlations of importance with the presumptive total all had scores of above .61. They were, in descending order, "Inclusion of Author's Experiences" (.71**), "Inclusion of Author's Thoughts and Feelings" (.67**), "Inclusion of Author's Voice (Style)" (.66**), "Inclusion of Author's Background" (.64**), "Inclusion of Author's Voice (Content") (.62**) and "Incorporation of Personal Pronouns" (.62**). In keeping with the pedagogy's author-centred focus, the high correlating attributes involved a tangible sense of the writer's presence in the writing, either in content or style.  47 None of the Expressivist attributes registered importance correlations of above .25 with Current Traditional Rhetoric. For instance, "Inclusion of Personal Truths" (which had an Expressivist correlation of .61**), correlated -.04 with the Current Traditional presumed total. While the personal occupies a front-and-centre position in the Expressivist tradition, there is little place for it in Current Traditional. In contrast, there were a couple of noteworthy Expressivist/Social Construction correlations: "Acknowledgement of Author's Biases" was the attribute with the lowest Expressivist (.39**) and highest Social Construction (.32**) correlations. The attribute's strong Social Construction correlation could arguably be because it is less about the author's idiosyncratic opinions front-and-centre in a text, and more about a rigorous examination of the validity of these beliefs, perfectly in keeping with the analytical nature of Social Construction. "Incorporation of Personal Pronouns" was another Expressivist item with a relatively high Social Construction correlation (.32**). These scores suggest a definite philosophical link between Expressivism and Social Construction, and demonstrate that individual expressivism and the way the individual constructs the social environment are not necessarily mutually incompatible entities. The 14 Social Construction attributes all generated statistically significant importance correlations with the presumptive Social Construction total. The five highest scores were: "Critical Analysis of Current Events" (.59**), "Investigation of Society's Impact on Knowledge" (.59**), "Examination of Society's Influence on Language" (.56**), "Discussion of World Events" (.53**) and "Investigation of Power Structures" (.53**). A common theme running through these higher correlating attributes is their emphasis on a critical analysis of the social context and a systematic questioning of accepted mores. The Social Construction attributes generated importance correlations of less than .25 with the presumed Current Traditional total. However, "Faithfulness to Academic Standards" correlated slightly better in importance with Current Traditional Rhetoric (.35**) than with its own pedagogy (.34**). This "cross-over" attribute may be more akin to the rule- and structure-oriented nature of Current Traditional Rhetoric than with the investigative, questioning nature of Social Construction. Reinforcing the opposing philosophical underpinnings of the two pedagogies, such attributes as "Call for Social Reform" and "Discussion of World Events" had Current Traditional correlation scores of a mere .01.  48  Social Construction attributes tended to correlate slightly better in importance with Expressivism than with Current Traditional Rhetoric, particularly around attributes like "Call for Social Reform" (.36**) and "Examination of Society's Influence on Language" (.31**). These correlations highlight the deep-rooted relationship between Expressivism and Social Construction while suggesting that Expressivism has an interest that extends beyond the personal to include the social environment in which the individual is placed. "Attention to Reader" had a better importance correlation with Expressivism (.37**) than with Social Construction (.33**), implying that instead of promoting self-indulgent writing, Expressivism is more attuned to the needs and concerns of readers than might have been previously assumed. Having established relationships between scored importance of the 45 individual attributes and scaled importance of the three pedagogies to which they were presumed to belong, the next step was gauging the validity of the pedagogies as whole scales of measurement. There were two general tests for the pedagogical scales to undergo. First, would respondents be able to discern differences among the three pedagogies? Second, would instructors and students demonstrate similar interpretations of the pedagogical groupings or would each respondent group read the composition pedagogies differently, rendering the scale unhelpfully idiosyncratic? Table 5 displays correlations of importance among the three composition pedagogies, as determined by instructors and students. The alpha reliabilities for each pedagogy are bolded in the diagonal. Table 5: Correlations between Pedagogies and Importance Assigned Attributes by Instructors and Students Current Traditional Instructors Students Current Traditional Expressivist Social Construction  Expressivist Instructors Students 19** .17*  Social Construction Instructors Students .37** .50**  .91  .89  .17*  .19**  .86  .84  .51**  .05  .37**  .51**  .40**  .75  .40**  .77  There was some correlation among the three pedagogies, which is to be expected since attributes of writing are likely to be used in sensible combinations with each other Instructors' scaled Current Traditional Rhetoric scores only registered a correlation of .17 with Expressivism but had a correlation of .50 with Social Construction. Students' scaled Current Traditional scores had a correlation of only .19 with Expressivism.  49 These correlations indicate respondents did register discernable differences among the three pedagogies, particularly in relation to the more rule-bound Current Traditional Rhetoric, as well as some conceptual overlap among the pedagogies. Italicized alpha reliability scores are useful for two reasons. First small differences between instructor and student alpha reliabilities indicate both respondent groups consistently converged on the pedagogies' overall meanings as constructed from the individual attributes that made them up. Instructors and students both appeared to interpret the attributes in similar ways, a necessary finding for this study given the jargon-heavy terminology used to describe the attributes. Second, the higher alpha reliabilities (in the diagonal) contrasting against the lower inter-pedagogical correlations indicate respondents were validly able to converge on the construct content of each pedagogy while diverging between one pedagogy and another (Campbell & Stanley, 1963). While attributes correlated highly in importance with calculated totals of their respective pedagogies, there was a blurring of traditional pedagogical lines. Some attributes might be more logically and accurately grouped with items that cross pedagogical boundaries. I discuss what these groupings might be and how they are to be formed next.  Section C: Differing World Views "Differing World Views" was based on Burrell and Morgan's (1979) conceptual mapping of four kinds of social theory: Functionalist, Interpretive, Radical Humanist, and Radical Structuralist. Each theory or paradigm is rooted in metatheoretical assumptions about the nature of social science and society. The paradigm generates theories and perspectives unique from and fundamentally opposed to those engendered by other paradigms, subsequently resulting in a fundamental and broad view that shapes how individuals comprehend their reality. World views influence how individuals relate to particular social issues. For instance, a capitalist and a socialist will have dramatically different views on state intervention in national economic policies. These ways of perceiving and understanding the world were of particular interest to this study since it sought to establish the link between rhetoric and reality, i.e. to determine if respondents' perspectives on good student writing were influenced by how they saw the world around them. What was at  50 stake went beyond the "Big Five Personality Traits" since the subject of investigation was the relationship between one's perceptions of the world with assessments of writing. Additionally, the world views scale offered a means for differentiating each individual respondent as well as an opportunity to map each respondent's world view profile in relation to other respondents. These world view profiles will be discussed further in "Respondents' World Views" (Chapter Seven). Burrell and Morgan focus on ontology, epistemology, human nature and methodology. Ontology concerns the extent social scientists regard "reality" as internal or external to the individual, i.e. if "reality" is objective and imposed from without, or the subjective result of individual consciousness. Epistemology concerns how an individual might understand the world and convey this knowledge to others. Is it possible to identify and communicate knowledge as hard, real and tangible, or as soft, experiential and essentially idiosyncratic? Can knowledge be acquired or must it be personally experienced? Next, there are assumptions about human nature, particularly the relationship between human beings and their environment. One extreme view conceptualizes human beings as products of environmental conditioning, responding to their surroundings in a deterministic fashion.  The opposing view assigns humanity a  much freer, more powerful role: individuals create their own environment and are in control of their fate. Assumptions concerning ontology, epistemology and human nature have direct bearing on methodology. If the social world is posited as a hard, objective and external reality, the methodology used in such an investigation would involve an analysis of the relationships among the various elements of which this reality consists. In contrast, if social reality is perceived as subjective experiences, the study will focus on the way individuals create, modify and comprehend the environment in.which they find themselves. The first, objective approach is "nomothetic" and emphasizes systematic protocols, techniques and the rigorous testing of hypotheses. The second, subjective approach is "ideographic." It stresses an insider, impressionistic focus that allows subjects to express their everyday nature during the course of the investigation. The ways in which ontology, epistemology, human nature and methodology relate to approaches to the social science are summarized in Figure 1.  51  The Subjective-Objective Dimension Subjectivist approaches to social science Nominalism  Objectivist approaches to social science ontology  •  Realism  Anti-positivism^  epistemology  ».  Positivism  Voluntarism  human nature  ^  Determinism  •  Nomethetic  Ideographic  <  *  methodology  Figure 1 : Burrell and Morgan's scheme for analysing assumptions about the nature of social science* *(Adapted from Burrell & Morgan, 1979, p.3).  A second layer of analysis deals with approaches to the nature of society. It has two opposing poles: the sociology of regulation and sociology of radical change. The former refers to social theory, i.e. why societies remain united and unified rather than fall apart, stressing the importance of regulating human behaviour within a cohesive social network. This sociology presumes it is possible to determine and satisfy human needs within the social system. The sociology of radical change, conversely, argues for radical change, deep-rooted structural conflict, and forms of domination and structural conflicts. This sociology is concerned with Utopia rather than what is, and with change not compliance. It focuses on the emancipation of the individual from social structures that stifle and stunt human potential. Finally, radical change argues that the social system prevents human fulfillment and that the status quo results in deprivation. Key concerns of the regulation-radical change dimension are presented below.  Table 6: The regulation-radical change dimension* REGULATION Sociology is defined by a RADICAL CHANGE sociology is defined concern for: by a concern for: The status quo Radical change Social order Structural conflict Consensus Modes of domination Social integration and cohesion Contradiction Solidarity Emancipation Need satisfaction Deprivation Actuality Potentiality *(Adapted from Burrell and Morgan, 1979, p. 18)  52 The subjective-objective and regulation-radical dimensions are subsequently placed together as a pair of axes to form the four paradigms. The relationship between the paradigms is illustrated in Figure 2.  RADICAL C H A N G E Radical Humanism  Radical Structuralism  SUBJECTIVE  OBJECTIVE  Interpretivism  Functionalism  REGULATION Figure 2: Four paradigms for the analysis of social theory* *(Adapted from Burrell & Morgan, 1979, p.22) According to Burrell and Morgan, the paradigms are defined by "very basic metatheoretical assumptions which underwrite the frame of reference, mode of theorizing and modus operandi of the social theorists who operate within them" (p.23). The term "paradigm" is used to emphasize the commonality of perspectives which unite a group of theorists whose work can be regarded as approaching social theory within a shared boundary. Each paradigm shares one axis with one other paradigm but is differentiated from that paradigm through its positioning along the other axis. For instance, "Radical Humanist" and "Radical Structuralist" share the horizontal radical/regulation axis but occupy different positions along the subjective/objective axis. Thus, the paradigms should be viewed as contiguous but separate: contiguous because of their shared characteristics, but separate because the differentiation among them is distinct and important enough to justify four unique entities. F u n c t i o n a l i s m : The Functionalist paradigm posits society as ontologically prior to the individual and seeks to place that individual within the wider social context. It is committed to providing an explanation of what is, particularly the reason for the regulated, rational and integrated nature of the social fabric, and assumes a continuing order and pattern. The underlying theme of the paradigm is its reliance on scientific  53 enquiry, particularly the possibility of objective investigation adhering to rules and regulations which govern the external world. Interpretivism: The primary concern of the Interpretive paradigm is the individual's subjective experience. Where the Functionalist paradigm conceives the investigator as an observer of action, the Interpretive places the individual as the central actor, one who creates the social environment through subjective experiences. This paradigm is antipositivist, rejecting the natural sciences for an attempt to get inside and to understand the phenomenon under investigation from within. Explanation is to be found within the realm of individual consciousness and subjectivity. Radical H u m a n i s m : Like the Interpretive paradigm, Radical Humanism posits that individuals create the world in which they live. However, the Radical Humanist believes the individual to be alienated and trapped within the social organization which he or she has both created and sustained through everyday life. Much Radical Humanist theory is focused squarely on the Functionalist paradigm and its reinforcement of the status quo. Radical Humanists have devoted considerable effort to critiquing the role of science, logic, rationality, technology, language and other aspects of the capitalist superstructure as vehicles of cognitive domination and barriers to the full achievement of human potential. The primary focus of this paradigm is the emancipation of individuals through the dismantling of these barriers. Radical S t r u c t u r a l i s m : Radical Structuralism is based on a materialist view of the natural and social world and on an ontology which emphasizes the hard and concrete nature of a social reality which exists independently from the minds of human beings. Like Radical Humanism, Radical Structuralism focuses primarily on critiquing and ultimately changing the status quo. Both share the goal of releasing humankind from the domination which is seen as characteristic of contemporary industrial society. Nevertheless, unlike Radical Humanism, Radical Structuralism cares little about the role and nature of individual human beings. Instead, the paradigm pays considerable attention to the way society is composed of contradictory structural elements. This deep-seated structural conflict will allow humankind emancipation from the social world which they inhabit. The objective is to use the crisis that arises from such conflict to completely replace one set of structures with another of a fundamentally different kind. Burrell and Morgan's four paradigms have had a significant impact on organization science (Deetz, 1996; Goles & Hirschheim, 2000), particularly because they challenge  54 the hold functionalism has had on the social sciences (Willmott, 1993). The paradigms have also been used to analyze a vast array of contexts including different approaches to AIDS education (Boshier, 1989), the cause and prevention of fishing accidents (Boshier, 1996), higher education journal literature (Milam, Jr., 1991), knowledge management (Schultze & Stabell, 2004), and community psychology and disability studies (Goodley & Lawthom, 2005). The 2x2 matrix provides alternative ways of investigating social phenomena as well as usefully complex methods for studying massive, multifaceted social organizations like a modern-day university. The paradigms acquire additional resonance within the context of this study: they not only demonstrate the diversity of perceptions about the world that reside within a globalized, elaborate entity like the University of British Columbia but also provide insight into the practical implications these differences bear upon the narrower reality of perspectives on good first-year writing. Operationalizing the Four World Views The key challenge was translating Burrell and Morgan's four paradigms into four reliable and valid indices that could be used in a user-friendly questionnaire form. Selfdescriptive adjectives appeared to be a useful point of entry and a widely-used strategy (Gough & Heilbrun, 1985). Burrell and Morgan's analysis of the world views of the various sociological theorists situated in each paradigm lent itself to adjectives. The grist and complexity of each paradigm would not be lost and a list of relatively familiar adjectives would be easily understandable to respondents. The next step was to go over Burrell and Morgan's four paradigms with a figurative fine tooth-comb and extrapolate relevant and accurate adjectives. I worked on transforming detailed sociological analyses into markers that could describe individuals. Once I had a core set of words, I used a thesaurus to find suitable synonyms. To understand how I operated, consider part of Burrell and Morgan's description of Radical Structuralism: Radical Structuralism is aimed, first and foremost, at providing a critique of the status quo in social affairs. It is a perspective which is concerned not just to understand the world, but to change it. The underlying focus of interest tends to be upon the structures within society, and particularly the way in which they interrelate. Writers within the paradigm tend to view society as composed of elements which stand in contradiction to each other. They are interested in the effects of these contradictions, particularly with regard to the role which they play  55 in creating economic and political crises. Radical structuralism is a view which focuses upon the essentially conflictual nature of social affairs and the fundamental process of change which this generates. Deep-seated conflict is viewed as the means by which man achieves emancipation from the structures of the social word in which he lives. It is a sociology of radical change but, in contrast to that of the radical humanist paradigm, one which tends to place relatively little direct emphasis upon the role and nature of man as an individual human being. However, common to both is the underlying aim of man's release from the various forms of domination which are seen as characterizing contemporary industrial society. (1979, pp.326-327) From the paragraph, I extrapolated these adjectives: "revolutionary," "radical," "conflict driven," "anarchist" and "aggressive." Adjectives like "conflict driven" and "aggressive" spawned synonyms like "adversarial" and "assertive." Using this method, I amassed a total of 120 adjectives, 30 words for each paradigm, to test through item analysis. I pilot tested the adjectives by asking respondents the question "How Do You Generally Think and Act?" Respondents were asked to score the 120 adjectives by describing the way they generally thought and acted. They were given five options for each adjective: "Never," "Rarely," "Sometimes," "Often" and "Always." Respondents were also asked their gender and highest level of education. Upon gathering a disparate pool of respondents, I began by contacting a convenience sample of friends and relations and asking them to contact a few of their friends, colleagues or relations (snowball sampling). I received 107 completed questionnaires. Respondents ranged in age from their late teens to sixties and varied in terms of occupation: teachers, students, secretarial staff, medical professionals, pastors and computer programmers were among the respondent pool. Of these individuals, 58 were male and 45 female; one respondent did not divulge gender. One had a high school diploma, six were technical school graduates, eight were Community College graduates, 25 had Bachelor's degrees, 25 had Professional degrees, 29 had Master's degrees, seven possessed doctorates, and one had a post-doctorate diploma. Two respondents did not divulge their educational achievements. The 120 words I began with are listed below in Table 7.  56 Table 7: Pilot Test Adjectives According to World View FUNCTIONALIST  INTERPRETIVE  Balanced Compliant Conciliatory Conformist Consensus Seeking Conservative Conventional Deferential Efficient Empirical Factual Law Abiding Logical Neutral Objective Orderly Organized Practical Pragmatic Problem Solving Proof Seeking Rational Realistic Reasonable Regulated Scientific Stable Status Concerned Structured Traditional  Accepting Agreeable Appeasing Artistic Caring Compassionate Concerned Considerate Creative Discerning Empathetic Emotional Humanistic Idealistic Individualistic Instinctive Introspective Intuitive Nurturing Obliging Open Minded Perceptive Reflective Sensitive Solicitous Spiritual Subjective Sympathetic Tolerant Understanding  RADICAL HUMANIST Action Oriented Activist Advanced Alienated Analytical Avant Garde Aware Autonomous Challenging Change Driven Critical Thinking Emancipated Forward Looking Freedom Seeking Ideological Independent Intellectual Investigative Liberal Liberated . Libertarian Political Potential Seeking Progressive Provocative Questioning Reformist Self Governing Socially Conscious Thought Provoking  RADICAL STRUCTURALIST Adversarial Aggressive Anarchist Antagonistic Assertive Class Conscious Communal Community Oriented Concrete Conflict Driven Confrontational Contentious Controversial Deterministic Disruptive Egalitarian Extreme Forceful Insurgent Intense Militant Powerful Radical Rebellious Revolutionary Self Sacrificing Single Minded Strong Structural Minded Violent  The aim was to pare down 120 adjectives to no more than 80 items, 20 for each paradigm. The task was to determine the veracity of the potential 80 adjectives as exemplars of their respective paradigms. Consequently, I determined the convergent and discriminant validity (Campbell & Fiske, 1959; Campbell & Stanley, 1963) of the adjectives to the four paradigms. Convergent and discriminant validity are inter-locking propositions that gauge the differential specificity of conceptual constructs. Convergent validity depends on observed correlations between multiple theoretically related concepts while discriminant validity tests or the absence of correlations among construct measures that ought not to be theoretically linked. Special attention was paid to the following: 1) how well each adjective correlated with the paradigm to which it was nominally assigned; 2) if and where adjectives correlated better with paradigms other than their own; 3) the internal reliability  57 (coefficient alpha) of a scale comprised of these 30 words. Table 8 lists the Functionalist adjectives correlations of importance in descending order with Functionalism.  Table 8: Correlations of 30 Functionalist Adjectives with four World View Indices in Descending Order of Functionalism ADJECTIVE Practical Orderly Structured Realistic Organized Rational Reasonable Balanced Stable Factual Objective Problem Solving Logical Efficient Pragmatic Compliant Deferential Empirical Proof Seeking Conformist Law Abiding Scientific Conventional Traditional Regulated Conservative Consensus Seeking Status Concerned Conciliatory Neutral Average Inter-Item Correlation Standardized Alpha  RADICAL STRUCTURALISM  INTERPRETIVISM  RADICAL HUMANISM  .69 .69 .66 .61 .61 .59 .57 .55 .56 .54 .53 .53 .52 .52 .52 .50 .50 .50 .47 .40 .40 .40 .36 .28 .32 .30 .29  .30 .31 .17 .19 .30 .24 .43 .41 .28 .13 .09 .11 .06 .41 .31 .51 .46 .06 .08 .21 .33 -.04 .22 .14 .18 .06 .15  .38 .20 .28 .34 .31 .47 .32 .33 .22 .51 .48 .41 .48 .34 .33 .07 .21 .45 .30 -.08 -.14 .45 -.22 -.18 -.13 -.10 .14  .06 .00 .17 .04 .06 .22 -.01 -.06 .07 .23 .32 .23 .19 .11 .09 -.11 .04 .21 .10 -.13 -.38 .34 -.27 -.11 -.20 -.14 -.02  .29  .06  .17  .31  .23 .12 .21  .51 .01  -.12 -.16  -.24 -.21  FUNCTIONALISM  .89  A few adjectives correlated better with paradigms other than Functionalism. Both Compliant and Conciliatory had their highest correlations with the Interpretive paradigm and "Scientific" with Radical Humanism. These items were removed, as were seven adjectives that had correlation scores of less than .40 with Functionalism: Conventional,  Traditional, Regulated, Conservative, Consensus Seeking, Status Concerned, and Neutral. The mean inter-item correlation of Functionalist adjectives scale was .21, resulting in a standardized item alpha of .89 - well above the test-constructionists' benchmark of .70 for inclusion/retention (Nunnally & Bernstein, 1994). When combined together, the 30 Functionalist adjectives served as reliable indicators of the Functionalist paradigm.  Table 9: Correlations of 30 Interpretive Adjectives with Four World View Indices in Descending Order of Interpretivism ADJECTIVE  FUNCTIONALISM  INTERPRETIVISM  Concerned Nurturing Compassionate Considerate Caring Understanding Sensitive Sympathetic Empathetic Reflective Instinctive Creative Intuitive Perceptive Appeasing Humanistic Solicitous Tolerant Agreeable Artistic Accepting Obliging Discerning Open Minded Idealistic Spiritual Emotional Individualistic Introspective Subjective Average Inter-Item Correlation Standardized Alpha  .36 .28 .28 .45 .34 .45 .27 .23 .12 .25 .26 .20 .34 .36 .27 .18 .24 .34 .19 .01 .18 .37 .22 .44 -.01 .17 -.10 .14 .01 -.03  .74 .69 .69 .68 .67 .64 .62 .59 .59 .56 .54 .50 .50 .49 .47 .44 .42 .42 .41 .40 .38 .37 .37 .37 .33 .37 .32 .26 .25 .08 .20 .88  RADICAL HUMANISM .31 .15 .21 .32 .26 .29 .07 .02 .04 .32 .17 .36 .44 .34 -.02 .35 .15 .21 .00 .13 .05 .06 .26 .65 .13 .17 -.11 .48 .09 .09  RADICAL STRUCTURALISM .11 -.06 -.05 .33 .06 .00 -.16 -.09 -.14 .05 .04 .11 .21 .13 -.09 .07 .00 -.03 -.10 -.06 .07 -.12 -.19 .33 .13 -.18 .01 .35 -.13 .06  59 Table 9 lists the Interpretive adjectives in descending correlation order. Two items correlated more strongly with Radical Humanism than with Interpretivism: Open Minded and Individualistic.  Eight adjectives correlated poorly with the Interpretive paradigm,  with correlation scores of less than .40: (Obliging, Accepting, Discerning, Open Minded, Idealistic, Spiritual, Emotional, Introspective and Subjective). These adjectives were removed. Mean inter-item correlation was .20 and standardized item alpha for the scale was .88 - again, well above the benchmark. Table 10 lists the Radical Humanist adjective correlations in descending order.  Table 10: Correlations of 30 Radical Humanist Adjectives with four World View Indices in descending order of Radical Humanism ADJECTIVE Thought Provoking Progressive Activist Liberated Independent Libertarian Political Critical Thinking Investigative Self Governing Autonomous Intellectual Provocative Questioning Advanced Avant Garde Freedom Seeking Emancipated Reformist Aware Forward Looking Potential Seeking Analytical Change Driven Political Socially Conscious Action Oriented Challenging Liberal Ideology Alienated Average Inter-Item Correlation Standardized Alpha  FUNCTIONALISM .23 .43 .22 .28 .31 .21 -.03 .24 .35 .37 .24 .18 .01 .14 .37 .21 .31 .20 .02 .42 .38 .23 .27 .24 -.03 -.05 .27 .24 .07 -.03 .03  INTERPRETIVISM .19 .41 .20 .30 .25 .08 .06 .14 .21 .18 .14 .15 -.13 .12 .15 .34 .43 .21 .09 .30 .39 .28 .09 .18 .06 .25 .18 .13 .04 .23 -.01  RADICAL HUMANISM .70 .67 .65 .63 .59 .56 .56 .55 .55 .55 .54 .53 .52 .52 .51 .50 .50 .49 .51 .46 .46 .45 .40 .43 .56 .36 .32 .34 .22 .24 .14 .20 .88  RADICAL STRUCTURALISM .56 .45 .54 .22 .30 .28 .41 .28 .26 .16 .23 .28 .63 .42 .23 .31 .15 .09 .41 .16 .13 .16 .13 .47 .41 .13 .22 .50 -.08 .27 .14  60 Provocative correlated better with Radical Structuralism than with Radical Humanism, and there were nine words with relatively poor Radical Humanist correlation scores, i.e. below .40 (Forward Looking, Analytical, Change Driven, Socially Conscious, Action Oriented, Challenging, Liberal, Ideological, and Alienated). The standardized item alpha for the scale was once again relatively high at .88. Table 11 lists the Radical Structuralist adjectives presented in descending order of Radical Structuralist correlation scores. Table 11: Correlations of 30 Radical Structuralism Adjectives with four World View Indices in Descending Order of Radical Structuralism ADJECTIVE  FUNCTIONALISM  Confrontational Radical Forceful Controversial Conflict Driven Disruptive Extreme Rebellious Powerful Revolutionary Adversarial Contentious Militant Anarchist Intense Insurgent Structural Minded Antagonistic Aggressive Strong Assertive Concrete Class Conscious Deterministic Violent Community Oriented Communal Self Sacrificing Egalitarian Single Minded Average Inter-Item Correlation Standardized Alpha  -.10 -.09 .26 -.12 -.12 -.29 -.04 -.26 .19 .03 .04 .03 -.10 -.19 .08 -.21 .51 -.25 -.19 .30 .16 .44 .13 .31 -.17 .21 .17 .07 .09 .07  INTERPRETIVISM  -.20 -.06 .00 -.19 -.20 -.25 .00 -.19 .12 .14 -.17 -.11 -.32 -.07 . .05 -.21 .16 -.30 -.32 .22 .17 .20 .12 .36 -.15 .29 .27 .44 .25 -.06  RADICAL HUMANISM  .29 .46 .45 .26 .24 .18 .36 .33 .44 .63 .31 .35 .16 .25 .36 .22 .38 .03 .13 .38 .42 .22 .12 .40 .05 .30 .10 -.02 .00 -.03  RADICAL STRUCTURALISM  .69 .69 .67 .64 .61 .61 .61 .59 .58 .58 .57 .52 .52 .51 .51 .50 .45 .43 .42 .40 .37 .34 .33 .32 .30 .27 .25 .02 -.01 -.02  61 They were five items that correlated more with paradigms other than Radical Structuralism.: Revolutionary, Assertive, Deterministic and Community Oriented scored their highest correlations with Radical Humanism while Concrete had its highest correlation with Functionalism. There were also five adjectives with poor, and in some cases negative, correlations with Radical Structuralism. They were Communal, Self Sacrificing, Egalitarian, and Single Minded. Correlations for these words were all under .30. All these words were subsequently dropped. With regard to reliability, the standardized item alpha for the Radical Structuralist scale was a little lower than the previous scales but still reasonably high at .86. Through this systematic whittling, I refined and streamlined the four world view indices from 120 to 80 adjectives. The 80 adjectives were used in the final phase of data collection involving 523 students and 157 instructors, and are listed alphabetically in Table 12.  Table 12: Eighty Adjectives Listed Alphabetically According to Paradigm Functionalist Balanced Conformist Deferential Efficient Empirical Factual Law Abiding Logical Objective Orderly Organized Practical Pragmatic Problem Solving Proof Seeking Rational Realistic Reasonable Stable Structured  Interpretive Agreeable Appeasing Artistic Caring Compassionate Concerned Considerate Creative Empathetic Humanistic Instinctive Intuitive Nurturing Perceptive Reflective Sensitive Solicitous Sympathetic Tolerant Understanding  Radical Humanism Activist Advanced Autonomous Avant-Garde Aware Critical Thinking Emancipated Freedom Seeking Independent Intellectual Investigative Liberated Libertarian Political Potential Seeking Progressive Questioning Reformist Self Governing Thought Provoking  Radical Structuralism Adversarial Aggressive Anarchist Antagonistic Class Conscious Conflict Driven Confrontational Contentious Controversial Disruptive Extreme Forceful Insurgent Intense Militant Powerful Radical Rebellious Strong Violent  Testing the Validity of the Eighty Adjectives The next phase of investigation tested the validity of the 80 adjectives as instruments of measurement. Findings are summarized in tables 13 to 16.  62 Table 13: Correlations of 20 Functionalist Adjectives with World View Indices in Descending Order of Functionalism ADJECTIVE  Functionalism  Interpretivism  Radical Humanism  Radical Structuralism  Problem Solving Orderly Structured Logical Rational Organized Realistic Efficient Practical Stable Factual Reasonable Balanced Pragmatic Law Abiding Proof Seeking Empirical Objective Deferential Conformist Average Inter-Item Correlation Standardized Alpha  .62** .60** .60** .59** .57** .56** .55** .54** .53** .53** .52** .52** .50** .48** .45** .44** .42** .42** .20** .16** .20 .83  .28** .27** .23** .21** .31** .28** .21** .25** .29** .31** .20** .30** .31** .31** .33** .20** 14** .19** .15** .06  .50** .27** .28** .34** .32**  .11* -.01 -.00 .02 -.04 -.02 .00 .01 -.03 -.13* .05 -.12* -.04 .08 -.26** .15** .13** .04 .17** -.01  24**  .18** .34** .23** .29** .30** .23** .33** .46** .24** .40** .38** .24** .09 .10  Table 14: Correlations of 20 Interpretive Adjectives with Four World View Indices in Descending Order of Interpretivism ADJECTIVE Compassionate Sympathetic Caring Empathetic Nurturing Considerate Concerned Sensitive Humanistic Reflective Understanding Tolerant Perceptive Intuitive Agreeable Artistic Instinctive Solicitous Appeasing Creative Average Inter-Item Correlation Standardized Alpha  Functionalism .29** .26** .31** .30** .27** .33** .31** "14** .25** .28** .33 .33** .40** .25** .27** -.01 .15** .27** .16** .15**  Interpretivism .71** .68** .67** .65** .63** .62** .59** .59** .57** .57** .55** .54** .53** .47** .43** .42** .40** .39** .38** .38** .25 .87  Radical Humanism .33** .28** .31** .36** .27** .26** .31** .21** .53** .43** .24** .31** .45** .35** .15** .18** .26** .39** .19** .31**  Radical Structuralism .04 -.04 -.07 -.05 -.09 -.16** .02 -.07 .07 .05 -.11* -.14** .06 .11* -.18** .10 .20** .19** .01 .17**  63 Table 15: Correlations of 20 Radical Humanism Adjectives with four World View Indices in Descending Order of Radical Humanism Functionalism ADJECTIVE 44** Critical Thinking .28** Liberated .43** Intellectual .29** Questioning .32** Thought Provoking .30** Advanced .40** Investigative .34** Autonomous .37** Potential Seeking .39** Progressive .23** Reformist .28** Self Governing .24** Emancipated .13** Political .30** Independent .37** Aware .19** Freedom Seeking .11* Activist .05 Avant Garde •|4** Libertarian . Average Inter-Item Correlation Standardized Alpha  Interpretivism .30** .43** .37** .32** .43** .25** .33** .27** .45** .34** .21** .29** .28** .11* .30** .42** .38** .17** .16** .23**  Radical Humanism .63** .57** .56** .56** .56** .55** .55** .54** .54** .54** .54** .54** .52** .52** .50** .47** .47** .46** .39** .32**  Radical Structuralism Radic 14**  .12* .07 .19** .07 .21** .13** .08 .20** .17** .26** .15** .12* .27** .19** .04 .21** .34** .31** .16**  .23  .86  Table 16: Correlations of 20 Radical Structuralist Adjectives with four World View Indices in descending order of Radical Structuralism ADJECTIVE  Functionalism  Confrontational Controversial Aggressive Radical Rebellious Adversarial Extreme Antagonistic Disruptive Forceful Insurgent Militant Powerful Conflict Driven Anarchist Violent Intense Contentious Strong Class Conscious Average Inter-Item Correlation Standardized Alpha  -.04 -.01 -.03 -.01 -.15** .05 -.17** -.07 -.19** .12* .01 .01 .23** -.01 -.16** -.22** .10 .17** .34** .21**  Interpretivism -.10 .01 -.14** .15** .05 -.03 -.01 -.10* -.16** .03 .04 -.02 .16** -.05 -.04 -.25** .25** .11* .29** .23**  Radica .09 .35** .06 .38** .24** .21** .12* .08 -.02 .29** 14** .13** .39** .04 .08 -.11* .40** .22** .38** .19**  Radical Structuralism .64** .62** .60** .59** .57** .55** .55** .54** .52** .51** .51** .50** .48** .48** .45** .45** .44** .41** .29** .24** .21  .84  Testing the validity of the 80 adjectives involved two separate though related steps. The first determined how well each adjective correlated with the paradigm to which it  64 was assigned; the second factored the adjectives into alternate groupings and tested correlations between these groupings and the paradigms. With regard to adjective/paradigm correlations, the objective was to see how well each adjective correlated with its associated paradigm, or if there were indeed cases in which adjectives correlated better with paradigms that were not their own. As the tables indicate, each adjective's assessment of importance correlated better with its own world view than with others, with the exception of two items. First, Solicitous generated identical correlations of .39 with both its Interpretive paradigm as well as Radical Humanism. Strong had a stronger correlation with Functionalism (.34**) than with its "home" world view, Radical Structuralism (.29**). The adjectives registered statistically significant correlations of importance with their own paradigms. A s further evidence of the distinctiveness of each world view, Confrontational, a Radical Structuralist adjective, had a negative correlation of-.10 with its philosophical opposite, the Interpretive world view. Similarly, an Interpretive adjective like Considerate had a statistically significant correlation of-.16** with Radical Structuralism. Finally, alpha reliabilities for each world view were relatively strong: .83 for Functionalism, .87 for Interpretivism, .86 for Radical Humanism, and .84 for Radical Structuralism. The reliability of the four world view scales was further tested by factoring the 80 adjectives. In order to determine potential adjective families, the 80 adjectives were run through several factor analytic procedures, including Varimax, Equimax, and Quartimax rotations. After several rounds of testing, a Varimax rotation based on 12 families which, viewed collectively, accounted for 49% of variance, was found to make the most sense. Varimax rotation provided groupings for all but three of the 80 adjectives. In total, there were 12 adjective families. Adjective Family #1 is Interpretive and made up of the following Interpretive items: Compassionate, Sympathetic, Caring, Nurturing, Concerned, Considerate, Sensitive, Empathetic, Understanding, Tolerant, Humanistic and Agreeable. Each item deals with relating to others and demonstrating understanding and concern. Adjective Family #2 consists entirely of Functionalist adjectives that revolve around logical and factual cognition: Realistic, Reasonable, Logical, Rational, Practical, Stable, Balanced, Problem Solving and Factual. The next factor, Adjective Family #3, is derived from Radical Structuralism and comprises nine items that involve discord and disunity: Antagonistic, Adversarial, Anarchistic, Insurgent, Conflict Driven, Confrontational, Contentious, Militant and Disruptive.  65 Adjective Family #4, however, draws heavily upon Radical Humanism and consists of seven adjectives, four of which are Radical Humanist in nature and deal with liberation: Emancipated, Autonomous, Reformist and Self Governing. The remaining adjectives are either Interpretive (Solicitous) or Functionalist (Empirical and Pragmatic). Adjective Family #5 is also a hybrid factor, though primarily Radical Humanist, and revolving specifically around thought. It contains five Radical Humanist adjectives (Questioning, Proof Seeking, Critical Thinking, Investigative, and Thought Provoking), one Interpretive item (Reflective) and one Functionalist descriptor (Objective). Adjective Family #6, on the other hand, is purely Radical Structuralist and consists of items that denote aggression and force: Extreme, Forceful, Powerful, Aggressive, Violent, Rebellious, Intense, Strong, Controversial, and Radical. Adjective Family #7 is made up entirely of four Functionalist items that represent structure and method: Organized, Orderly, Structured and Efficient. Adjective Family #8 is largely Interpretive and consists of Interpretive items that suggest subjective insight: Instinctive, Intuitive, and Perceptive. The family also contains one Radical Humanist descriptor, Independent. Next, Adjective Family #9 is predominantly Radical Humanist, consisting of Radical Humanist adjectives the individual actively challenging the status quo: Activist, Political, Progressive, Libertarian, Liberated, Avant Garde, and Potential Seeking. This family also contains one Radical Structuralist item, Class Conscious. Adjective Family #10 consists of three items, all of them from the Radical Humanist paradigm and dealing with enlightened thought. They are Advanced, Intellectual and Aware. Subsequent families are made up of two items each. The first, Adjective Family #11, contains Conformist and Appeasing, which belong to the Functionalist and Interpretive paradigms respectively, though they share a theme of obeisance. Finally, Adjective Family #12 comprises two Interpretive descriptors that suggest imagination and originality: Creative and Artistic. The adjective families generally duplicate the four paradigms. In other words, the 80 adjectives, when rotated, come out conceptually as described by Burrell and Morgan. Thus, they succeed as an instrument for gauging the world views of respondents at U B C .  Testing the Validity of the Twelve Adjective Families  Another question concerned the extent each adjective family correlated with the world view from which it was derived. It was not enough for adjective families to capture  66 the conceptual  thrust of aworld  view;  instead,  afurther requirement  r e g i s t e r as i g n i f i c a n tc o r r e l a t i o n w i t h t h e w o r l d v i e w w h i c h s u m m a r i z e s  c o r r e l a t i o n s b e t w e e n  i td r e w  t h e 1 2 a d j e c t i v e f a m i l i e s a n d  was  each  f r o m .  family  T a b l e 17  t h e f o u r w o r l d  v i e w s .  T a b l e 17: C o r r e l a t i o n s b e t w e e n A d j e c t i v e F a m i l i e s a n d W o r l d V i e w s World View  #2 #7 #11  Functionalist Functionalist Functionalist/ Interpretive Interpretive Interpretive Interpretive Radical Humanism Radical Humanism Radical Humanist Radical Humanist Radical Structuralist Radical Structuralist  #1 #8 # 12 #4 #5 #9 #10 #3 #6  I n a l l c a s e s ,  Radical Structuralist Correlation  -.02 -.04 -.03  .64** .51** .16**  .18** .07  .13**  .19** .10* -.13**  .18** .04 -.10* .30**  .80** .31** .28**  .23** .26** .13**  .23**  .55**  -.04 .07 .06 .11*  .39**  .22**  .49**  .09*  .04  .10*  .43**  .20** -.03  -.03  .18**  .01  -.06  .08  .71**  -.08  -.01  .15**  .62**  a d j e c t i v e f a m i l i e s r e g i s t e r e d t h e i r s t r o n g e s t  c o r r e l a t i o n s o f  w h i c h  c o n s i s t e d  t h e y m o s t  o f 1 2  I n t e r p r e t i v ei t e m s r e g i s t e r e d a n  S t r u c t u r a l i s td e s c r i p t o r s , c o r r e l a t e d m o s t a correlation of .71**. Thus,  i m p o r t a n c e  c l o s e l y r e p l i c a t e d . F o r i n s t a n c e , A d j e c t i v e  . 8 0 . * * S i m i l a r l y , A d j e c t i v e F a m i l y # 3 , w h i c h w a s  g r o u p i n g s  Radical Humanism Correlation  .15**  w i t h t h e p a r a d i g m # 1 , w h i c h  Interpretive Correlation  Functional Correlation  Family  m a d e  I n t e r p r e t i v e c o r r e l a t i o n o f  u p e n t i r e l yo f  s t r o n g l y w i t h R a d i c a l  the adjective families consist  o f t h e a d j e c t i v e s w h i l e r e i f y i n gt h e c o n c e p t u a l  F a m i l y  R a d i c a l  S t r u c t u r a l i s m ,  of interpretable and f r a m e w o r k  g e n e r a t i n g logical  o f t h e f o u r  w o r l d  v i e w s .  Section D: Background T h e a g r e e d  q u e s t i o n t h a t i n f o r m e d " B a c k g r o u n d " w a s t o p a r t i c i p a t e i n t h i s s t u d y a n d  R e s e a r c h  a r e t h e r e s p o n d e n t s w h o  b e s t s u m m a r i z e s  acquisition, and  of great sensitivity in an  socio-demographics  institutionas  stratified as  h a v e  t h e i r c h a r a c t e r i s t i c s ?  o n t h e d i v i s i o n s w i t h i n p r e s e n t - d a y u n i v e r s i t i e ss u g g e s t e d  position, language matter  w h a t  w h o  t h e s e v a r i a b l e s :  or background. a modern-day  Position  is a  university.  67 There is the obvious distinction between student and teacher, as well as entrenched divisions among faculty members. At the top of the metaphorical pole are tenured professors, followed, in descending order, by tenure-track professors, sessional instructors, and teaching assistants. Salaries, perks, privileges and occupational security all differ depending on the individual's position within the university (Mysyk, 2001; Rajagopal, 2002; Puplampu, 2004). These divisions are likely to continue since contingent faculty (both part-time and full-time non-tenure-track) now makes up 65% of all faculty in degree-granting institutions in the United States, up from 58% in 1995 (Curtis, 2005). The situation is not much different in Canadian universities, with the number of full-time university instructors falling from 32,669 in 1991 to 30,484 in 2001 (Canadian Association of University Teachers, 2006) despite rising student enrollments. Thus, occupational status plays a significant role in distinguishing among faculty members. Another variable involves language acquisition, particularly if students and faculty members are not native speakers of English. While universities across North America are becoming more international, attracting students from a multitude of nations and linguistic backgrounds, English is still the lingua franca of the academy. Problems with student writing are no longer limited to the stylistic or grammatical carelessness of native speakers, but must include students expressing themselves in the sophisticated discourse of an unfamiliar language and grappling with the obstacles such expression entails (see Angelova & Riazantseva, 1999; Canarajah, 2001; Fishman & McCarthy, 2002; Hamp-Lyons & Zhang, 2001; Leki, 1995; 2003; Schneider & Fujishima, 1995 for more extensive analyses on the challenges E S L students encounter in academic environments). A student's language history and ease or lack thereof with academic English can have a significant impact on the role he or she is encouraged or invited to assume in a university environment. Does a respondent not being a native speaker of English substantially influence his or her understanding of "good" student writing in English? Li (1996) demonstrated differences in responses between U S and Chinese high-school teachers to sample student writing from both countries; similarly, rhetoricians have investigated how cultural differences in written discourse patterns or rhetorical conventions might negatively affect writing in a second language (Connor, 1996; Grabe & Kaplan, 1989). The challenge for this study was to avoid simplistic  68 conclusions of difference, as Kubota and Lehner (2004) have argued, but to shape a nuanced pictured of respondents' involvement with academic writing in English. The final variable was socio-demographic characteristics. While the advancement of women instructors and students in academia has reduced the gender gap somewhat, female faculty only make up 27% of tenured professors (Canadian Association of University Teachers, Section 2, p. 12, 2006), and, on average, earn $12,000 less than their male counterparts (Canadian Association of University Teachers, Section 2, p.2, 2006). Female student enrollment has increased steadily in the few decades with women making up 58% of the undergraduate student body in Canadian universities in 2003/2004 (Canadian Association of University Teachers, Section 3, p.4, 2006). Women, however, are still severely underrepresented in engineering, architecture and the computer sciences, while men make up less than 4 0 % of humanities and education undergraduates (Canadian Association of University Teachers, Section 3, p.4, 2006). Ethnicity and age are two other significant markers. In 2001, 11.1 % of Canada university instructors were visible minorities (Canadian Association of University Teachers, Section 2, p.7, 2006). A s for age, only 23.1% of full-time Canadian university teachers in 2003/2004 were below the age of 40 (Canadian Association of University Teachers, Section 2, p.9, 2006), largely because of budget cuts that have significantly curtailed the hiring of younger faculty to full-time positions. How markers of gender, country of origin and age influence perspectives on writing will be discussed in chapters to come. A breakdown of the four areas investigated in this section appears below:  •  Biography:  Respondents' Gender; Year in which Respondents were Born; Country of Origin; Year Respondents Arrived in Canada.  •  Role:  Whether Respondents were Instructors or Students; Instructors' Current Position at U B C (Teaching Assistant/Sessional Instructor/Tenure-Track Professor/Tenured Professor/Other); Department with which Instructors are Affiliated; Instructors' Academic Specialty;  69  Faculty in which Students are Enrolled; Students' Intended Academic Major; Number of Years Respondents have been at U B C .  • , Language:  Respondents' First Language; Respondents'Highest Academic Language; Years Respondents have written Academic Papers in English; Whether Instructors have received Formal Training in Marking Student Writing.  •  Academics:  Respondents' Highest Academic Qualification; Year of Graduation; Country in which Respondents Graduated with Their Highest Academic Qualification.  Because "Background" is the only section in the questionnaire that asks slightly different questions of instructors and students, the challenge was to ensure that respondents were asked the same types of questions, with near identical wording, albeit with divergences owing to their different roles within the university. I invited three Writing Centre students to comment on the questionnaire. Responses were generally favourable. Students had no difficulty understanding the questions they were posed. The final version of "Background" is provided in Appendix B, "Questionnaire." Having developed and tested the instrument to be used in the study, I contacted respondents and distributed the questionnaire. I discuss the results derived from the study in subsequent chapters, beginning with the fractures of writing that were uncovered in "Responses to Sample Student Writing" (Chapter Five).  70 CHAPTER FIVE: RESPONSES TO SAMPLE STUDENT WRITING  The overall purpose of this study was to document fractures in perspectives on good academic writing among instructors and between faculty and students at the University of British Columbia. "Sample Student Writing," the first section of the questionnaire," narrowed the focus of investigation by investigating fractures in respondents' ranking and grading of three sample opening paragraphs that might have been written for a take-home paper for first-year English. Each paragraph was written in the style of a specific composition pedagogy: Current Traditional Rhetoric, Expressivism and Social Construction (please refer to "Compositional Collisions," Chapter Two, for an expanded discussion of these pedagogies). Three questions lay behind this section of the questionnaire: 1) How would the paragraphs, each representing a different writing style and, by extension, a distinct composition pedagogical preference, be first ranked and then graded? 2) Would there be general consensus or disagreement among instructors about how these paragraphs were individually ranked and graded, and if so, to what extent? 3) Would instructors' ranking of the paragraphs also correlate with grades they assigned? If so, how strong and valid would such links be? Respondents were presented with three sample paragraphs and asked to do several things: 1) Read all three paragraphs before making any responses. 2) Rank the paragraphs according to the writing style they preferred most, middle, and least. 3) Assign each paragraph a numerical grade for writing quality out of 10 points. Half points were allowed. Respondents had the choice of giving a top-ranked paragraph the lowest grade of the three or grade a bottom-ranked paragraph highly. Respondents were presented this option so I could gauge their assessment of academic writing from two angles - ranks and grades. A marker may gravitate towards a particular writing style but believe that preferred style to be academically deficient. Conversely, markers might grade a certain style highly even though they are averse to it. This phase of the study sought to determine whether and to what extent such disjunctures occurred.  71 A total of 157 instructors across the Faculty of Arts, the Department of Language and Literacy, the Writing Centre, and the Technical Writing Program at the Faculty of Applied Science participated in the study. The questionnaire was also completed by 523 students enrolled in either first-year writing offered by the University of British Columbia's English Department, or a course on introduction to academic writing at the Writing Centre. (See "Procedures," Chapter Three, for further details). The total number of respondents was 680. Ranks Respondents Assigned the Paragraphs Each of the three paragraphs was ranked top, middle or top by at least one of the 680 respondents. Additionally, from a possible range of 11 marks (0-10), there was an observed range of marks awarded for each paragraph (four to ten marks for the Current Traditional and Social Construction paragraphs, and two to ten marks for the Expressivist sample). Table 18 summarizes the rankings for each paragraph. Percentages are in parentheses.  Table 18: Instructors' and Students' Preference Ranks for Paragraphs in Three Pedagogical Styles Rank  Top Middle Bottom Missing Total  Current Traditional Rhetoric (%)  Expressivism (%)  Social Construction (%)  Instructors  Students  Instructors  Students  Instructors  73 (46) 53 (34) 29 (19) 2(1) 157(100)  149 224 140 10 523  23 (15) 51 (32) 81 (52) 2 (1) 157 (100)  82 (16) 197 (38) 234 (44) 10(2) 523(100)  61 51 43 2 157  (28) (43) (27) (2) (100)  (39) (33) (27) (1) (100)  Students 284 (55) 117(22) 112 (21) 10(2) 523(100)  Instructors' Ranking of the Paragraphs  No one paragraph earned a majority in terms of how best to write. Of the 157 instructors, 46% assigned the Current Traditional paragraph top ranking, 39% ranked the Social Construction paragraph top, and 15% ranked the Expressivist paragraph top. Fifty two percent of instructors ranked the Expressivist paragraph bottom, the only majority position in this phase of respondent decision making. Nevertheless, fifteen  72 percent of faculty gave it the top rank while 32% ranked it middle, suggesting nearly half of all instructors surveyed were not prepared to dismiss it entirely. There was also lack of consistency surrounding the Social Construction paragraph. Thirty nine percent of instructors ranked the Social Construction paragraph top; 33% middle; and 27% bottom. Thus, this writing style had nearly as many faculty members who were not prepared to support it, as instructors who were. Students' Ranking of the Paragraphs A total of 523 students participated in the study. The greatest amount of consistency among student respondents involved the 55% who ranked the Social Construction paragraph top, with only 28% and 16% assigning the top rank to the Current Traditional and Expressivist paragraphs respectively. Nevertheless, the number of students who thought the Social Construction paragraph worth only a middle position (22%) nearly equaled the amount who ranked it bottom (21%). Students were also divided about the Expressivist writing sample. While 44% of students gave this generally unconventional style the bottom rank, 38% ranked it middle. The Current Traditional paragraph occupied a solidly middle position in the student ranking scheme with 4 3 % of students ranking it middle. There was an almost equal match-up in the number of students who thought this paragraph deserved the top rank (29%) with those who ranked it bottom (27%). When placing instructor and student responses side-by-side, several patterns emerge. First, fractures were observed between instructors' and students ranking of the Current Traditional (F=14.86; p<.001) and Social Construction (F=8.65; p<.001) paragraphs. Students preferred the more open-ended Social Construction paragraph though their instructors' endorsement of this paragraph was tepid. Second, there were various perspectives on good academic writing among instructors, suggesting lack of consensus even among faculty. The debate surrounding good student writing also surfaced in the grades awarded to the writing samples, as I demonstrate below. Grades Respondents Assigned to the Paragraphs Apart from being asked to rank the three paragraphs top, middle or bottom, respondents were asked to assign points to each writing sample. They were requested  73 not to give any two paragraphs the same number of points. I examined the assigned grades from three angles: first, looking at each paragraph individually; second, comparing grade ranges for all three paragraphs; third, investigating possible grade margins to demarcate the boundaries of "good" or "poor" writing. The first step was determining grade means, standard deviations and observed ranges for each paragraph. Table 19 summarizes this information. A total of 155 instructors and 513 students participated in this phase of data collection.  Table 19: Instructors' and Students' Assessments of the Paragraphs (Grades and Standard Deviations Paragraph  Grade Means  Standard Deviation  Observed Range  Instructors  Students  Instructors  Students  Instructors  Students  Current Traditional  7.87  7.34  1.11  1.39  4-10  1-10  Expressivist  7.08  6.92  1.28  1.51  2-10  0.5-10  Social Construction  7.76  8.00  1.25  1.48^  4-10  1-10  Instructors assigned the Current Traditional paragraph a grade mean of 7.87 (the highest grade mean among the three paragraphs), the Expressivist sample a grade mean of 7.08, and the Social Construction paragraph 7.76. Each of the three paragraphs earned a grade mean of above 7.00, despite the disparity in styles and handling of essay topic. There was a difference of only .11 between the Current Traditional and Social Construction paragraphs, and a gap of .79 between the Current Traditional and Expressivist paragraphs. Thus, while instructors appeared most comfortable assigning a score of about 70 percent to first-year writing, there was little agreement about what kinds of academic writing deserved such a grade. Standard deviations (SDs) among instructors for the three writing samples were 1.11 for the Current Traditional, 1.28 for the Expressivist, and 1.25 for the Social Construction paragraphs. The paragraph that earned the highest grade mean, Current Traditional, recorded the greatest consensus among instructors. The sample with the lowest grade mean, Expressivist, generated the greatest spread of opinion. The Expressivist writing style apparently incited a love-hate reaction: some instructors favoured it strongly while others were vociferous about its unsuitability.  74 Finally, instructors assigned the Current Traditional paragraph a grade range of four out often to ten out of ten; the Expressivist paragraph a range of two to ten; and the Social Construction paragraph four to ten. I combined grades respondents gave to each paragraph to determine if some instructors were naturally "tough" (i.e. distributing only a small portion of the total of 30 points that were theirs to award) or lax (i.e. dispensing a larger fraction amount of their possible 30 points across the various paragraphs). Of 30 points an instructor could have assigned over three paragraphs, their cumulative grades (out of 10 for each paragraph) ranged from 12 to 28.5 with a mean of 22.7 and a standard deviation of 2.46. Two-thirds of instructors assigned grades between about 20 and 27 although there was some skew toward the lower end. In percentage terms, grades they assigned ranged from 4 0 % to 9 5 % with an average of 75.5%. Two-thirds of instructor-assigned grades fell between 67.3% and 83.7%, which falls somewhere between a C+ and an A in U B C ' s grading scale. Of total grades assigned, the Current Traditional paragraph earned a mean of 35% the Expressivist a mean of 3 1 % , and the Social Construction sample 34%. The Current Traditional sample had a lead of only one percent over the middle-ranked Social Construction paragraph, and four percent over the bottom-ranked Expressivist text. Thus, ranking exaggerated differences that may not have been substantive. Despite its respective overall ranking position, none of the three paragraphs could claim a majority in terms of grades collectively awarded. There was a fracture between instructors' and students' grading of the Current Traditional paragraph (F=18.87; p<.001). Students assigned the Current Traditional paragraph a mean of 7.34 (.53 lower than the grade mean instructors assigned the paragraph). Students assigned the Expressivist sample a grade mean of 6.92. The highest grade mean, 8.0, went to the Social Construction paragraph. The grade mean differential between paragraphs with the highest and lowest grade means was 1.08, as opposed to a spread of .79 between paragraphs instructors graded highest and lowest. This suggests students had a narrower idea than instructors did of what scores should be assigned writing samples of differing quality. Student S D s were higher than instructors': 1.39 for the Current Traditional, 1.51 for the Expressivist, and 1.48 for the Social Construction paragraphs respectively. The S D for the Social Construction paragraph was greater than that of the Current Traditional  75 sample, despite the former earning the highest grade mean of the three paragraphs. Thus, despite its generally high grade standing among students, there was still considerable disagreement among this group of respondents about the presence of a Social Construction style in first-year writing. Furthermore, the high S D which the Expressivist paragraph generated demonstrates that students, like instructors, were strongly divided about this discursive style. Students assigned the Current Traditional and Social Construction paragraphs a range of grades from one out of ten to a total of ten points while the Expressivist sample had a slightly more extensive scope from half a point out of ten to ten out of ten. Of the 30 points students could have assigned over three paragraphs, grades dispensed ranged from 4.50 to 28.50 with an average or mean of 22.2 (SD=2.87) with two-thirds of student respondents assigning grades of between 20.50 to 24.50. Put in percentage terms, the grades they awarded the paragraphs collectively ranged from 15% to 95% with an average of 74%. Two-thirds of respondents assigned grades of between 68.3% and 82%, between a B- and an A- in U B C ' s grading scheme. The Current Traditional paragraph obtained a mean of 3 3 % of total grades assigned, the Expressivist paragraph 3 1 % , and the Social Construction sample 36%. Thus, the Social Construction paragraph, which received the most top ranking positions from students, had a grade lead of only three percent over the middle-ranked Social Construction sample and five percent over the bottom-ranked Expressivist text. None of the paragraphs could claim a majority in terms of grades collectively awarded. This led to another layer of data analysis. What would grade ranges for the three paragraphs be if extreme positions were highlighted and removed? I answer this question below.  Range of Grades Assigned to the Paragraphs The range of grades assigned by respondents offer immediate points of comparison of how instructors and students assessed the three samples. Figures 3 and 4 illustrate grade ranges for the three paragraphs based on grade scores provided by instructors and students.  76 12 10 VS  CZ»40  -X611  155 C T R Paragraph Grade  Expressivist Parag G  S o c Constvst Parag G  Figure 3: Instructors' Grade Ranges for the Three Paragraphs  12 10  -X«B -X297  -X*oe -XI77  o  O "  -X23S  H  513 C T R Paragraph Grade  Expressivist Parag G  S o c Constvst Parag G  Figure 4: Students' Grade Ranges for the Three Paragraphs The figures represent grade ranges for the Current Traditional Rhetoric, Expressivist and Social Construction samples, in that order. Boxes depict the grades given by 50% of respondents while black lines ("whiskers") on both sides of the boxes show the highest-to-lowest range remaining respondents assigned. Oblong-shaped items above and below the whiskers indicate extreme grades given by at least one  77 respondent. Numbers on the right of each oblong represent the numerical code assigned that respondent during data entry. Figure 3 depicts grades assigned by 155 instructors. The semi-interquartile range (SIQR) for the Current Traditional paragraph was between seven to 8.5 with a median of eight. The Expressivist paragraph had a similar SIQR from 6.5 to eight out often points, with a median score of seven out often. The numbers of the Social Construction paragraph match that of the Current Traditional sample exactly. Over 50% of respondents gave it between seven to 8.5 points with eight out of ten being the median score. Thus, the three paragraphs had identical grade ranges and near-duplicate scores from over 50% of all respondents: there was no writing style instructors could agree on as the definitive representative of good academic writing. Figure 4 depicts grades assigned by 513 students. The range of grades assigned by students overlapped. Over 50% of student respondents assigned the Current Traditional paragraph between seven to eight points out of ten with a median of 7.5. Over 50% of students assigned the Expressivist paragraph between six and eight points with a median of seven. The Social Construction paragraph followed a familiar pattern with most students assigning this sample between seven and nine out of ten with a median of eight. A s with instructors, no one writing style emerged favourite. While the Social Construction paragraph had the highest median grade: eight out often, it was only half a point higher than that of the Current Traditional paragraph, and a point more than that of the Expressivist sample. I also tested the relationship between paragraph rank and grade, paying close attention to whether there were discernible grade range margins when a paragraph was ranked bottom, middle or top. Table 20 summarizes these findings. Table 20: Grades Instructors and Students Assigned Current Traditional, Expressivist and Social Construction Paragraphs by Rank Majority Grade Range  Social Construction Rank  Expressivist Rank  Current Traditional Rank Bottom  Middle  Top  Bottom  Middle  Top  Bottom  Middle  Top  Instructors  7-8  6-9  7-10  5-8  6-9  6-10  5-9  7-8.5  7-10  Students  5-8  6-9  7-10  5-9  6-9  7-10  4-9  5-10  7-10  78 Grades listed in the table represent assessments assigned by at least 50% of respondents to the three paragraphs. Grades overlapped for all three rankings, indicating a general lack of agreement about "margins" differentiating "good" or "weak" paragraphs. Grades for the Current Traditional paragraph when in the bottom rank overlapped with grades it received in the middle and top ranks. This pattern was repeated for the Expressivist and Social Construction writing samples as well, resulting in a blurring of grade expectations across the pedagogically inspired paragraphs. The Current Traditional paragraph could earn a grade of 80% despite being ranked bottom, or a grade of 70% when assigned the top rank. A Social Construction paragraph could be awarded a grade of 90% with a bottom-place ranking, or a grade of 70% despite being ranked top. No matter the writing style, instructors did not display a distinct system for grading writing they found either "commendable" or "poor." Student grading exhibited the same fuzziness of margins. When ranked bottom, the grade range for the Current Traditional paragraph coincided somewhat with the grade range received in the top position. Similarly, the Expressivist paragraph could be assigned a grade of 90% and be ranked bottom, or 70% and occupy the top ranking. The Social Construction paragraph could be ranked bottom by students but still receive a grade of 90%, or be ranked top and be assigned only 70% of the grades allotted. Neither students nor instructors exhibited a standardized system of marking, no matter the writing style, suggesting alternate and viable perspectives on good student writing among respondents.  Relationships between Paragraph Ranks and Paragraph Grades Having discussed grades allocated by instructors and students, as well as the implications of such practices, I investigated the relationship between paragraph grades and ranks. Grade/rank relationships offer insight into how instructors and students assessed writing they perceived to be of differing quality. There were three ways of proceeding. The first was through grade/rank correlations. The second was internally, i.e. determining how grades varied depending on the position each piece of writing occupied in the minds of respondents. The third was through analysis of variance, i.e. identifying grades each paragraph received in relation to the other paragraphs in order to establish the significance of overall different group grade means.  79 Correlations between Paragraph Ranks and Paragraph Grades Table 21 summarizes statistical correlations between ranks and grades. Data in this phase was collected from 155 instructors and 513 students. Table 21: Grade/Rank Correlations for Current Traditional, Expressivist and Social Construction Paragraphs C u r r e n t Traditional Rank Student Instructor Current Traditional Grade Expressivist Grade Social Construction Grade  .56**  .59**  -.17*  -.25**  -.34**  -.34**  Expressivist R a n k Instructor -.16 .48** -.32**  Student -.23** .55** -.33**  Social Construction Rank Instructor Student -.33** -.30** -.20**  -.20**  .64**  .68**  Instructor grade/rank correlations were strongest within each pedagogically inspired writing style. The correlation between the Current Traditional paragraph grade and rank was a statistically significant .56. Correlations between the Expressivist and Social Construction grades and ranks were also statistically significant at .48 and .64 respectively. The Social Construction grade/rank correlation was the highest of all three paragraphs, suggesting respondents who ranked this paragraph favorably were most likely to grade this writing style highly as well. There were several statistically significant negative correlations, the largest involving the Social Construction paragraph. The Social Construction grade/Current Traditional rank correlation was -.34; the Social Construction grade/Expressivist rank correlation was -.32; the Current Traditional grade/Social Construction rank correlation was -.30. Respondents who ranked the Social Construction paragraph highly were least likely to rank other writing styles highly as well, and vice versa. This might be because the Social Construction writing style has a set of distinct characteristics that are highly valued by a specific group of instructors. This hypothesis will be tested further in "Results of Attributes of Writing" (Chapter Six). Student grade/rank correlations display many of the same patterns found in the instructor findings. Grade/rank correlations were strongest within each pedagogically inspired writing style. The grade/rank correlation for the Current Traditional paragraph was a statistically significant .59; it was .55 for the Expressivist and .63 for the Social  Construction samples, respectively. Once again, correlation was highest for the Social Construction paragraph. Students who preferred this writing style were most likely to grade it highly as well. While student grade/rank correlations were within .05 points of instructor grade/rank correlations, the student grade/rank correlation for the Expressivist paragraph was .07 higher than it was for instructors. An Expressivist writing style thus had a stronger hold among students than it did instructors. There were several statistically significant negative correlations in the data provided by students. The Expressivist paragraph had statistically negative grade/rank correlations with both the Current Traditional Rhetoric and Social Construction paragraphs (r=-.25 for each correlation) while the Current Traditional Rhetoric paragraph correlated negatively with the Social Construction (r=-.33) and Expressivist (r=-.23) samples. Among students, the highest negative correlations involved the Social Construction paragraph grade. It correlated with almost equal negative significance with both the Current Traditional (r=-.34) and Expressivist (r=-.33) paragraph ranks, indicating that advocates of a more critical, open-ended form of writing were unlikely to favour formulaic or deeply personal discursive styles. Respondents thus appeared to perceive a clear distinction between the three styles of writing investigated. Current Traditional Rhetoric, Social Construction, and Expressivism are pedagogies with different philosophical orientations and foci so pronounced they can be detected even in short opening paragraphs and result in statistically significant fractures of opinion. Would patterns of positive intra-pedagogy grade/rank correlation hold up when the paragraphs were broken down by the three available ranking positions? I discuss these answers below.  Relationships between Paragraph Grade and Bottom, Middle and Top Paragraph Ranks In order to understand how instructors and students graded pieces of writing they perceived to be of varying quality, I determined the grade means which the paragraphs received in each of their three potential ranking positions. Table 22 summarizes this information; number of respondents is provided in parentheses.  81 Table 22: Instructors' and Students' Paragraph Assessments by Rank Paragraph Rank  Instructors' Grade Mean  Grade Mean Differential between Ranks  Grade Mean Differential between Ranks  SD  SD  Students' Grade Mean  6.30 (140) 7.27 (223) 8.47 (148)  + .97 +1,20  1.26 1.21 .79  Current Traditional Bottom Middle Top  7.05 (29) 7.42 (53) 8.53 (73)  + .37 +1.11  1.26 .85 .74  6.52 (81) 7.54 (51) 8.07 (23)  +1.02 + .53  1.18 1.10 .93  6.09 (234) 7.34 (196) 8.32 (81)  +1.25 + .98  1.38 1.08 1.29  + .79 +1.65  1.15 .92 .62  6.37 (111) 7.68(117) 8.70 (283)  +1.31 +1.12  1.56 1.07 .96  Expressivist Bottom Middle Top Social Construction Bottom Middle Top  6.52 (43) 7.80 (51) 8.59 (61)  A total of 155 instructors provided both paragraph grade and rank information. There was a direct relationship between rank and grade: grades consistently increased as paragraphs were assigned higher ranks. When the Current Traditional paragraph was ranked top, it earned its highest grade mean, 8.53; when it was ranked middle, it received its second highest grade mean, 7.42; and its lowest grade mean, 7.05, when given the bottom position. Standard deviation was lowest among instructors (.79) when this sample was ranked top, and highest (1.26) was ranked bottom. Thus, there was strongest consensus when respondents agreed that that this was the best sample of the three. Conversely, when respondents felt that this was the weakest paragraph, the grades they gave the text had the widest range. While instructors had a shared conception of how to respond to pieces of writing they liked, they lacked a collective understanding of how to evaluate writing for which they had less regard. The table presents a parallel grade/rank relationship for the Expressivist paragraph. When ranked top, the Expressivist paragraph had its highest grade mean, 8.07, and its lowest grade mean, 6.52, when ranked bottom. Standard deviation was lowest (.93)  82 among instructors who gave the writing sample its highest rank, and largest (1.18) when respondents ranked it bottom. There was a similar grade/rank relationship for the Social Construction paragraph. When in the top rank, the Social Construction paragraph earned its highest grade mean, 8.59. When ranked bottom, this paragraph earned its lowest grade mean, 6.52. Standard deviation was smallest among respondents (.62) when the Social Construction text was ranked highest and largest (1.15) when the paragraph was ranked lowest. (Reinforcing the relationship between paragraph rank and grade, additional data analysis revealed that each paragraph received its highest grades when it was also ranked top. Please refer to Appendix C for further details.) A total of 511 students provided both paragraph grade and rank information. There was a direct relationship between the ranks and grades students assigned to each paragraph. When ranked top, the Current Traditional paragraph received a grade mean of 6.30; this number increased to 7.27 when the paragraph was ranked middle, and 8.47 when ranked top. Standard deviations of 1.26 and .79 when the sample occupied the bottom and top positions respectively indicate that consensus was greatest the higher the paragraph was ranked. A s was the case with instructors, students appeared to disagree about the value of a piece of writing if they believed it to be of weaker quality. There was considerably more agreement about writing perceived as "good." Conforming to a familiar pattern, the Expressivist paragraph was assigned its lowest grade mean (6.09) when in the bottom rank, its middle grade mean (7.34) when ranked middle, and it highest grade mean (8.32) when in the top position. The standard deviation for the paragraph was greatest (1.38) when it was ranked highest but lowest when it was in the bottom rank (1.08) A s if to reinforce Expressivism's problematic role in academic discourse, even students who preferred this writing style seemed conflicted about it. The Social Construction paragraph received its lowest grade mean (6.37) when ranked bottom, its next highest grade mean (7.68) when placed middle, and its highest grade mean (8.70) when ranked top. Like the Current Traditional example, standard deviations dropped by .60 from 1.56 to .96 when the paragraph moved up the ladder from the bottom to the top rung.  83 In short, there was a definite relationship between a paragraph's internal rank and grade. Additionally, no matter the writing style, instructors and students generally assessed writing the same way. They tended to agree on how a text was to be graded when they thought highly of it, but were more divergent in their opinions when they rated it poorly. Nevertheless, there were dissenting opinions over an unconventional style, even among the people who favoured it.  Effects of Paragraph Ranking on Grading Did ranks assigned by 157 instructors and 523 students have any statistically verifiable effect on the grades given to the paragraphs? Answers are presented below.  Table 23: Variance Accounted for in Instructors' and Students' Paragraph Grades by Paragraph Rank FTest Student Instructor Grade Mean Grade Mean  Significance Instructor Student  Current Traditional Rank  39.01  135.59  .00  .00  Expressivism Rank  21.11  111.35  .00  .00  Social Construction Rank  62.10  66.51  .00  .00  Paragraph rank had statistically significant impact on grades instructors assigned the Current Traditional (F=39.01; p<.001), Expressivist (F=21.11; p<001) and Social Construction (F=62.01; p<.001) paragraphs. Similarly, ranks students assigned the paragraphs had statistically significant impact on their grading of the Current Traditional (F=135.59; p<.001), Expressivist (F=111.35; p < 0 0 1 ) a n d Social Construction (F=66.51; p<.001) samples. These F scores suggest the following. First, both instructors and students observed differences among the three pedagogies. Second, one of the concerns of a study of this nature, which depends on the participation of strangers, is for respondents to take the endeavor seriously. The table suggests respondents were committed to evaluating and identifying differences in the three opening paragraphs. Third, the largest F-test  84 scores for both instructors and students among the three pedagogies were derived from Social Construction. This would indicate that while each of the three writing styles were distinct in the minds of both sets of respondents, the Social Construction paragraph stood out among respondents.  Analysis and Implications Findings derived from "Sample Student Writing" reveal fractures between instructors' and students' ranking and grading of the three pedagogically inspired sample paragraphs. Additionally, instructors failed to demonstrate consensus as to what constituted good student writing, hinting at fractures among faculty. There were dissenting opinions among respondents on how to rank and grade the paragraphs, varying perspectives on writing quality, overlapping grade ranges and consistency in ranking and grading relationships. Dissenting Opinions of How to Rank and Grade the Paragraphs Instructors' and students' perspectives on good writing were characterized by dissent rather than accord. Fractures were observed between instructors' and students' ranking of the Current Traditional and Social Construction paragraphs, and in their grading of the Current Traditional sample. Students were more likely to rank the Social Construction sample higher than their instructors did. Instructors, however, tended to rank and grade the Current Traditional paragraph higher. Fractures between instructors and students were further complicated by the lack of consensus among instructors. Each of the three paragraphs was ranked top, middle or bottom by a sizable number of instructors. Despite Current Traditional's entrenched presence in writing handbooks and courses, there was no "majority" for any of the pedagogically representative sample paragraphs (Current Traditional included) in terms of "how to write"; the clear majority for the Expressivist text functioned along the lines of "how not to write." Still, while Expressivist writing is not generally presented as a model for academic writing, it did receive some faculty support. The overall lack of consensus about three writing styles was also revealed in grade means each paragraph received from instructors. The difference between paragraphs with the highest and lowest grade means was less than a point. Instructors had such varied perspectives on good academic writing that three dissimilar writing styles could  85 earn fairly close grades. Possible fractures among faculty are investigated in "Rival Hypotheses" (Chapter Nine). These findings illustrate the difficulties student writers face. Fractures in perspectives on good writing have such tangible presence in university classrooms they result in a splintering of opinions between instructors and students. While students might have their own understanding of good writing, there is little guarantee the instructor assessing their writing shares these views. Worse, the splintering extends among faculty. Thus, even if a student manages to figure out what a particular instructor wants, it is unlikely to be what other instructors will value as well.  Varying Opinions of Writing Quality If there was no definite agreement among instructors about which writing style best represented good academic writing, there were also diverging views on how first-year writing should be graded. This variety is demonstrated in grade ranges received both singly and collectively. Each of the three paragraphs received a fairly wide range of grades from both instructors and students. While this grade range also reflects extreme positions (which can sometimes occur in writing assessment), it suggests grade ranges are fluid and flexible. A s if to underscore this broad spectrum of opinion, several respondents informed me, either through emails, notes on the questionnaire, or in personal conversation, that they felt the paragraphs were of a higher standard than writing they regularly received from first-year students. Conversely, a few other instructors argued that the paragraphs were of inferior quality and did not adequately represent their expectations of first-year writing. Anecdotal evidence aside, the paragraphs' collective grade ranges highlight a lack of agreement among instructors and students on how to assess first-year writing. If, for instance, points assigned to all three paragraphs were combined to make up a total of 30 (10 points for each paragraph), the grade range instructors awarded the paragraphs collectively was from 12 to 28.5 with a mean of 22.7. Two-thirds of instructors assigned grades between about 20 and 27, or from 67.3% to 83.7%. Students dispensed a collective grade range of anywhere from 4.50 for 28.50 with a mean of 22.2. Two-thirds of student respondents awarded the paragraphs grades between 20.50 to 24.50 or from 68.3% to 82%. At the University of British Columbia, a grade of 67% is equivalent to a  86 C+ while 84% earns a student a strong A-, highlighting the range of responses elicited from two-thirds of both respondent groups.  Overlapping Grade Ranges Grade ranges for all three paragraphs generally overlapped. Consequently, no paragraph had a distinct and definitive grade range that placed it well and above the other two samples in this study. All three paragraphs had identical grade ranges and near-duplicate scores from over 50% of all instructors. No one writing style trumped all others in terms of grade ranges and can legitimately claim to best represent good academic writing, foreshadowing fracture potential. No one writing style emerged as the student favourite either. Findings from over 50% of student respondents indicate grade ranges for all three paragraphs overlapped, with no sample receiving grades that put it head-and-shoulders above the rest. Grade ranges also overlapped when a paragraph was assigned a bottom, middle or top ranking, signifying general lack of agreement about where the "margins" that distinguish "good" or "weak" paragraphs should be. Fuzzy grade ranges are problematic because they reduce legitimate expectations instructors and students might have about what good academic writing should look like. If writing is the coin of the realm, what does it say about the efficacy of this commodity if instructors and students cannot agree on the value of its worth? Consistency in Ranking and Grading Relationships One reason for the fractures of writing documented in this chapter could be the definite sense exhibited among respondents of the distinctiveness surrounding each writing style. There was a positive rank/grade correlation within each pedagogy but negative rank/grade inter-pedagogy correlations. This was the pattern for both instructors and students. Additionally, grading had a statistically significant impact on paragraph ranking in that respondents registered the differences among the three writing styles in their scores. Findings revealed a positive correlation between the grade a paragraph earned and the rank it was awarded. If a paragraph was ranked highly, it was more likely to earn a high grade as well. Additionally, when a paragraph occupied the top rank, it received a higher grade mean than was it was ranked bottom. Respondents, in other words, knew  87 what they liked and were prepared to grade it accordingly. The relatively high standard deviations when the paragraph occupied the bottom slot indicate disagreement was generally greatest when a piece of writing was deemed weak. The contrast between the fuzziness of the grade ranges and the crisp linearity of internal grade/rankings might seem paradoxical: if respondents are generally clear about the relationship between grade and rankings why are grade ranges so inconsistent? There are at least two answers to this question. The first is that the university must expect a certain level of structure and consistency. Thus, some conformity of thought and behaviour is necessary. Nevertheless, the ivory tower also houses thousands of individuals, each with a unique set of assumptions and beliefs about academic standards. This clashing of structure and idiosyncrasy shapes perspectives on good academic writing. The various upstanding individuals that make up a university realize that good writing must be rewarded with the appropriate rank or grade; the dilemma lies in ambiguous and tangled definitions of "good" or "poor." Respondents' adeptness at picking up on variations in writing focus and style makes sense. A central tenet of composition studies is it reliance on traditional dichotomies: self vs. other; internal vs. external; subjective vs. objective (see Bawarshi, 1997 for more on this point). Deep-rooted beliefs in the fallacies of pedagogies that are not one's own are energetically represented ad infinitum in journal articles, conference proceedings and published books in the field (for examples of such key debates see Bartholomae, 1995; Elbow, 1995). The sound and fury do not signify nothingness; rather, characteristics underpinning each pedagogically inspired writing style are so pronounced they can be picked up on by instructors and first-year students who have probably had little exposure to composition scholarship. None of the three pedagogies were dismissed out-of-hand by either instructors or students; each pedagogy had a sizable support base at this university, and one could make a compelling case for its inclusion in academic discourse. The findings dispel comforting myths about agreement on good writing, revealing gashes hoping to approximate a whole. In "Responses to Attributes of Writing" (Chapter Six), I extend the investigation by examining fractures between instructors' and students' assessments of the importance accorded 45 attributes of good writing.  88 CHAPTER SIX: RESPONSES TO ATTRIBUTES OF WRITING "Responses to Sample Student Writing" documented fractures between instructors' and students' ranking and grading of three sample paragraphs. In this chapter, I pursue the investigation of fractures in perspectives on good academic writing from a different angle, namely instructors' and students' assessments of 45 attributes of good writing. Academic discourse is a genre students must master if they are to succeed academically. Unfortunately, the attributes that define this genre and distinguish it from other types of writing can be a mystery to students. The purpose of the section of the questionnaire entitled "Attributes of Writing" was to crack the code of unspoken expectations by ascertaining the importance instructors and students assigned writing attributes derived from three composition pedagogies: Current Traditional Rhetoric, Expressivism and Social Construction. Specifically, the questions I sought to answer in this phase of the study were as follows: 1) Which attributes were assessed most or least important, and just how important were they? 2) How well did attributes correlate with their pedagogies? 3) Would the findings reveal other ways of grouping the importance of the attributes beyond that of the three established pedagogies? Respondents were posed the following question: "How important is each of the following attributes in a take-home student essay for first-year English?" The 45 attributes were listed over two pages. To the right of each attribute were five options for respondents: "Not at all"; "Slightly"; "Moderately"; "Highly"; and "Critically." Respondents were requested not to leave any blanks and told to circle their responses. Their scores were subsequently translated numerically into scores of one out of five ("Not at all") to five out of five ("Critically"). On average, respondents provided information for 17.96 out of a total of 18 Current Traditional attributes, 12.93 of 13 Expressivist items, and 13.93 of 14 Social Construction attributes, indicating respondents were generally conscientious about evaluating each attribute. Next, respondent scores were divided by the number of attributes within its respective pedagogy. In order to determine how important Respondent X found an attribute like "Correct Use of Prepositions," I divided the  8 9  numerical score he or she gave to the attribute (between 1 and 5 points) by the total number of Current Traditional attributes he or she had scored resulting in a mean score. Table 24 categorizes the attributes according to the three pedagogies used in this study: Current Traditional Rhetoric, Expressivism and Social Construction.  Table 24: Writing Attributes According to Pedagogy Expressivism Current Traditional Adherence to syntax rules Clarity of sentences and paragraphs Coherence of essay Command of grammatical rules Command of punctuation rules Command of spelling rules Correct use of articles Correct use of prepositions Effective use of transition words Links between paragraphs Placement of main ideas Precision of word choice  Acknowledgement of author's biases Inclusion of author's background Inclusion of author's experiences Inclusion of author's thoughts and feelings Inclusion of author's voice Inclusion of personal truths Author's involvement with the subject Development of author's point-of-view Honesty of author's voice Incorporation of personal pronouns Use of imagery Use of metaphors  Proof of organization Sophistication of vocabulary  Social Construction Acknowledgement of social context Call for social reform Critical analysis of current events Discussion of world events Examination of society's influence on language Familiarity with related scholarship Investigation of power structures Investigation of society's impact on knowledge Evidence of academic inquiry Evidence of critical thinking Attention to reader Faithfulness to academic standards Use of academic discourse conventions Use of academic rhetorical strategies  Structure of thesis statement Structure of topic sentences Unity of paragraphs Variety of Sentence Structures  Importance Assigned the Forty Five Writing Attributes Assessments of importance were gauged through mean scores. Instructor mean scores are listed in descending order in Table 25, together with the pedagogy to which the attribute traditionally belonged. Student mean scores, overall ranks and standard  9 0  deviations (SDs) for each attribute are listed in parentheses. "CTR" represents Current Traditional Rhetoric, " E X P " Expressivism, and " S C O N " Social Construction. Table 25: Instructors' Assessments of the 45 Writing Attributes in Descending Order of Importance PEDGY SCON CTR CTR SCON CTR CTR SCON CTR CTR CTR  RANK 1(3) 2(1) 3(2) 4(19) 5(11) 6(4) 7 (22) 7 (tied) 9 (5) 10(6)  ATTRIBUTE Evidence of Critical Thinkina Clarity of Sentences and Paragraphs Coherence of Essay Evidence of Academic Inquiry Proof of Organization Placement of Main Ideas Faithfulness to Academic Standards Adherence to Syntax Rules Command of Grammatical Rules Structure of Thesis Statement  MEAN 4.75 (4.25) 4.61 (4.49) 4.60 (4.38) 4.38 (3.77) 4.23(3.86) 4.12 (4.21) 4.07 (3.51) 4.07 (3.79) 4.06 (4.11) 4.05(4.10)  11 (8) 12(9) 13(14) 14(9) 15(7) 16(12) 17(17) 18(16) 19 (12) 20(20)  Unity of Paragraphs Links between Paragraphs Precision of Word Choice Correct Use of Articles Command of Spelling Rules Structure of Topic Sentence Correct Use of Prepositions Effective Use of Transition Words Command of Punctuation Rules Development of Author's POV  3.94 (3.97) 3.91 (3.89) 3.85(3.82) 3.83(3.89) 3.78 (4.05) 3.75(3.84) 3.74(3.80) 3.73(3.81) 3.67(3.84) 3.58 (3.60)  .82 (.79) .79 (.87) .75 (.81) .89 (.86) .98 (.89) .93 (.88) .94 (.83) .85 (.84) .85 (.90) .96 (.89)  21(38) 22(14) 23 (23) 24 (25) 25(21) 26 (24) 27 (27) 28 (27) 29 (29) 30 (29)  Familiarity with Related Scholarship Attention to Reader Acknowledgement of Social Context Honesty of Author's Voice Variety of Sentence Structures Acknowledgement of Author's Biases Sophistication of Vocabulary Author's Involvement with Subject Inclusion of Author's Voice (Content) Use of Academic Discourse  3.48 (2.82) 3.38 (3.82) 3.21(3.41) 3.17 (3.30) 3.15 (3.56) 3.11 (3.36) 3.07 (3.26) 3.00 (3.26) 2.99 (3.20) 2.89 (3.20)  1.30(1.05) 1.10 (.85) 1.08 (.81) 1.17 (1.04) .83 (.81) 1.07 (.94) .85 (.86) 1.18 (.90) 1.18 (1.09) 1.07 (.91)  SCON SCON SCON EXP CTR EXP CTR EXP EXP SCON  31(26) 32 (32) 33 (29) 34 (35) 35(34) 36 (37) 37 (32) 38(41) 39 (39) 40 (35)  Investigation of Society's Impact on Critical Analysis of Current Events Use of Academic Rhetorical Investigation of Power Structures Inclusion of Author's Voice (Style) Use of Imagery Inclusion of Author's Thought/Feelings Use of Metaphors Examination of Society's Influence Discussion of World Events  2.85 (3.28) 2.82 (3.11) 2.79 (3.20) 2.76 (2.95) 2.75(3.09) 2.56 (2.88) 2.52 (3.11) 2.42 (2.69) 2.36 (2.74) 2.35 (2.95)  1.16 (.93) ' 1.02 (1.02) 1.01 (.90) 1.19 (.93) 1.17 (.98) .94 (.98) 1.23 (1.12) .97 (.97) 1.11 (.99) 1.07(1.01)  SCON SCON SCON SCON EXP EXP EXP EXP SCON EXP  41 (43) 42(42) 43 (45) 44(39) 45 (44)  Incorporation of Personal Pronouns Inclusion of Author's Experiences Inclusion of Author's Background Inclusion of Personal Truths Call for Social Reform  2.08 (2.49) 1.94 (2.60) 1.90 (2.38) 1.75(2.74) 1.65 (2.46)  .60 (.81) .54 (.68) .59 (.72) .79 (.97) .81 (.89) .98(77) .96 (.89) .74 (.88) .73 (.83) 1.00 (.88)  1.02 (.87) .97 (1.01) .98 (1.07) .91 (1.14) .88 (.87)  CTR CTR CTR CTR CTR CTR CTR CTR CTR EXP  EXP EXP EXP EXP SCON  91 On average, 673 respondents (155 of instructors and 518 of students) participated in this phase of data gathering. Instructors assigned Evidence of Critical Thinking the highest mean score, 4.75, and Call for Social Reform lowest, 1.65, with a steady decrease in importance of about 0.07 from one item to the next. The ten writing attributes with the highest mean scores, listed in descending order, were: Evidence of Critical Thinking, Clarity of Sentences and Paragraphs, Coherence of Essay, Evidence of Academic Inquiry, Proof of Organization, Placement of Main Ideas, Faithfulness to Academic Standards, Structure of Thesis Statement, Adherence to Syntax Rules, and Command of Grammatical Rules. With the exception of Critical Thinking, Academic Inquiry and Faithfulness to Academic Standards, these attributes were related to Current Traditional Rhetoric and are staples of most writing handbooks and writing classrooms. Two central, intertwined themes were clarity and structure: sentences and the essay must be clear and lucid; this clarity is dependant on correct syntax, organization, positioning of primary arguments, and grammar. Seven of the items with the highest mean scores appeared under the heading "Mechanics," emphasizing the importance of clear, precise structure. The overall goal is standardization, language, content, and thought. Attributes with the highest and fourth highest mean importance scores among instructors were Evidence of Critical Thinking and Evidence of Academic Inquiry. Critical thinking is about inciting students to grapple with conflicts and contradictions, and freeing themselves from simplistic and confining ways of understanding the world and writing (Bruffee, 1988; Gale, 1996; Graff, 1992). It draws directly upon the Culture of Thought described in "The Myth of Good Academic Writing" (Chapter One). Six of the attributes with the lowest mean scores were Expressivist (Imagery, Thoughts and Feelings, Metaphors, Personal Pronouns, Author's Experiences, Author's Background and Personal Truths). The others (Society's Influence on Language; World Events; Social Reform) were from Social Construction. Two themes emerged from these items: inclusion of the personal and social awareness. Low scores assigned to attributes dealing with personal or social analysis point to a practical problem. If instructors do not encourage students to discuss their social environment or personal discoveries in writing classrooms what are they to write about? To compound the challenge, seven of the ten attributes with the lowest importance  92 scores are under "Content" (Call for Social Reform, Discussion of World Events, Discussion of World Event, Examination of Society's Influence on Language, Inclusion of Author's Background, Inclusion of Author's Experiences, Inclusion of Author's Thoughts and Feelings," and Inclusion of Personal Truths.) This consensus about the low importance of content-related items highlights a central characteristic of first-year English: it is not a subject-specific course with a set curriculum and established body-of-knowledge like History 101. Course content and objectives can be diffuse, depending on the interests and expertise of the instructor. For instance, "Strategies for University Writing" (English 112), a mandatory first-year English course offered at the University of British Columbia, is vaguely described as focusing on "the rhetorical principles and strategies central to university-level discourse" (University of British Columbia, Department of English, 2005, para.1). Nevertheless, students are expected to generate at least four pieces of writing, though it is unclear what they must write about in a course without a set epistemological foci. Finally, standard deviations suggest respondents generally agreed on the importance of individual items - with low standard deviations indicating high consensus. "Evidence of Critical Thinking" earned the highest mean score among instructors and also had one of the lowest standard deviations (.60), indicating consensus about its high importance. Items that had the twenty highest mean scores generally had standard deviations of below a point, indicating some degree of agreement. Items with the five lowest mean scores had standard deviations of below a point, suggesting relatively little disagreement among instructors about the comparative low importance of these attributes. The largest standard deviations involved attributes placed middle (numbered #21 to #40) with importance scores ranging from 3.48 to 1.65 out of 5. Among attributes with the lowest consensus about their importance were Familiarity with Related Scholarship, (#21/ SD=1.30), Inclusion of Author's Voice ("Content") (#29/SD=1.18), Critical Analysis of Current Events (#32/SD=1.24), Investigation of Power Structures (#34/SD=1.19), and Inclusion of Author's Thoughts and Feelings (#37/SD= 1.23). These five items were derived from either Social Construction or Expressivism. Instructors who may have never associated these attributes with writing before might have scored them as "not at all" important. Instructors aware of developments in composition scholarship might have felt passionately about these attributes and scored them "critically" important.  The relative unimportance of these attributes in comparison  93 to the mechanically inspired items is not to suggest instructors were necessarily against critical social analysis or personal expressions, but that opinion over their importance was divided. Students assigned Clarity of Sentences and Paragraphs the highest importance among all attributes (mean=4.49) and Inclusion of Personal Truths the lowest (2.38). The range between the two ends of the spectrum was 2.11, in contrast to instructors' range of 3.10. The difference in spread of importance might have stemmed from firstyear students' general unfamiliarity with academic writing and its attributes; rather than fiercely espousing a handful of attributes over others, students assigned the attributes closer levels of importance (averaging .05). The ten writing attributes with the highest student importance scores, listed in descending order, were: Clarity of Sentences and Paragraphs, Coherence of Essay, Evidence of Critical Thinking, Placement of Main Ideas, Command of Grammatical Rules, Structure of Thesis Statement, Command of Spelling Rules, Unity of Paragraphs, Links between Paragraphs, and Correct Use of Articles. The first six of these items also appeared in the instructor top-ten list with the four remaining attributes generating toptwenty instructor mean scores. With the exception of Evidence of Critical Thinking, a historically Social Construction attribute, the top attributes all belonged to Current Traditional Rhetoric, indicating a student bias toward mechanic aspects of writing. Instructors' and students' rankings of importance generally matched each other with Current Traditional and a few Social Construction attributes gaining relatively high placement, and Expressivist and the more provocative Social Construction items doing less well. Spearman's rho revealed a strong correlation of .92 between the order of instructor/student rankings of the 45 attributes and a Pearson correlation of .93. Rankings by themselves are limited sources of information because they do not convey the width of the gap between the attributes' assessments of importance. Consequently, I paid close attention to differences in the mean scores assigned by respondents. There were three occasions in which instructor mean scores exceeded student means by at least .50 out of 5. They involved Familiarity with Related Scholarship (difference= .66), Faithfulness to Academic Standards (.56), Evidence of Academic Inquiry (.61), and Evidence of Critical Thinking (.50), which was also the item which earned the highest mean score from instructors. All four attributes are Social Constructionist and deal with adherence to academic discourse conventions. There  94 were also several instances when student assignations of importance exceeded that of instructors: Inclusion of Personal Truths (difference= .99), Inclusion of Author's Thoughts and Feelings (.59), Inclusion of Author's Experiences (.66), and Discussion of World Events, (.60). The first three items are Expressivist, implying that while students did not find personal disclosure crucial to first-year writing, they did value its presence more than their instructors might. Finally, Discussion of World Events, a Social Construction attribute, suggests students were more likely than instructors to value some reference to the world around them in their academic writing. Overall, there were numerous fractures between instructors' and students' assessments of the importance of the 45 attributes. The largest fractures involved such student-favoured attributes like Call for Social Reform (F=97.40; p<.001), Inclusion of Personal Truths (F=95.51; p<.001), and Inclusion of Author's Experiences (F=53.51; p<.001). Other notable fractures involved instructor-endorsed attributes like Evidence of Academic Inquiry (F=55.12; p<.001), Evidence of Critical Thinking (F=50.75; p<.001), and Faithfulness to Academic Standards (F=47.18; p<.001). Thus, there was lack of agreement between instructors and students as to which attributes constituted good writing. While instructors emphasized thought, contextspecific standards and inquiry, students expressed fondness for social analysis and personal disclosure.  Levels of Critical Importance Among the Forty-Five Attributes Attributes were tested for criticality. Of the possible five levels of importance that respondents could assign to each attribute (from one for "Not at all" to five for "Critically"), how often was an attribute deemed critical and how were Critically Important points distributed? W a s criticality fairly evenly spread out among the 45 attributes, or did particular items receive a larger share of critical points than did others? Additionally, did instructors and students generally agree on which attributes were crucial, or was opinion divided? Answers are in Table 26.  95 Table 26: Instructors' and Students' Distributions of Critically Important Points Among the 45 Writing Attributes Attribute  Percentaoe Instructors Stud  Evidence of Critical Thinkina Coherence of Essay Clarity of Sentences and Paragraphs Evidence of Academic Inquiry Proof of Organization Placement of Main Ideas Structure of Thesis Statement Faithfulness to Academic Standards Command of Grammatical Rules Adherence to Syntax Rules Command of Spelling Rules Correct Use of Articles Familiarity with Related Scholarship Links between Paragraphs Structure of Topic Sentences Unity of Paragraphs Correct Use of Prepositions Precision of Word Choices Command of Punctuation Rules Effective use of Transition Words Development of Author's Point-of-View Attention to Reader Acknowledgement of Author's Biases Inclusion of Author's Voice (Content) Acknowledgement of Social Context Honesty of Author's Voice Critical Analysis of Current Events Use of Academic Discourse Conventions Author's Involvement with the Subject Investigation of Power Structures Investigation of Society's Impact on Knowledge Inclusion of Author's Thoughts and Feelings Inclusion of Author's Voice (Style) Use of Academic Rhetorical Strategies Variety of Sentence Structures Sophistication of Vocabulary Discussion of World Events Inclusion of Author's Experiences Inclusion of Author's Background Use of Imagery Use of Metaphors Examination of Society's Influence on Language Call for Social Reform Inclusion of Personal Truths Incorporation of Personal Pronouns  79 66 64 51 41 40 38 35 29 29 27 25 25 23 23 23 21 19 17 17 16 13 9 9 9 9 8 8 7 7 5 5 5 5 4 3 2 1 1 1 1 1 1 0 0  45 49 57 24 24 39 38 13 37 21 36 24 5 24 22 26 20 20 25 20 15 21 9 11 6 11 7 6 8 4 6 9 5 6 11 8 4 3 3 4 3 4 1 6 1  Not all attributes were rated "critically" important by instructors. On one end of the spectrum was Evidence of Critical Thinking, deemed critically important by 79% of the 157 instructors; on the other was Inclusion of Personal Truths and Incorporation of  96 Personal Pronouns which failed to be labeled critically important by even one faculty respondent. The table makes apparent hierachies of importance in the minds of faculty about the value of the 45 items. Evidence of Critical Thinking was deemed a crucial element of academic writing by 79% of instructors, the largest endorsement for any of the attributes. Coherence of Essay, and Clarity of Sentences were considered critically important by 66% and 64% of instructors respectively. To put the value accorded to Evidence of Critical Thinking in some perspective, consider that it was termed critically important by 13% more faculty than the second highest item, Coherence of Essay. If Coherence and Clarity occupy a solid, second-tier level, Evidence of Academic Inquiry places a notch below with the support of 5 1 % of faculty - some 15 percentage points below the endorsement generated by Coherence of Essay. Academic Inquiry is followed by Proof of Organization, Placement of Main Ideas, Structure of Thesis Statement and Faithfulness to Academic Standards - items which were awarded "critically important" points from between 4 1 % to 35% of instructors. The level below contains Command of Grammatical Rules and Adherence to Syntax Rules, both of which received critically important points from 29% of instructors. The ten items with the largest instructor endorsements varied considerably in perspectives on importance. The difference in instructor support between Evidence of Critical Thinking and Adherence to Syntax Rules was a hefty 50%. In contrast, there was a drop of only 27% in instructor endorsement between Command of Spelling Rules (the item which earned the 1 1 highest number of critically important points) and th  Incorporation of Personal Pronouns, the attribute that received the lowest number of critically important points overall. These items were considered substantially less important than items with faculty support of upwards of 30%, and were on a relatively even keel in lack of importance in first-year writing. The various levels of criticality of importance within the higher scoring attributes can be interpreted as an extension of the jostling between thought and mechanics in firstyear English. First-year English occupies an anachronistic position in the academy. While its function is to introduce students to the rigours of academic life and the intricacies of academic analysis and critical thinking, first-year English has devolved into a skills-based program that redresses the linguistic difficulties of beginning students. Yet, while instructors bemoan the sad state of student writing, the teaching of academic  97 writing is often looked upon as an unnecessary evil, a service any self-respecting university should not be forced to perform (see Connors, 1997, for more on this point). The attributes deemed critically important by less than 2 % of faculty respondents all dealt either with a promotion of highly subjective language and content or social examination. Instructors did not brand any of these writing features essential to firstyear academic writing. There was general overall agreement between instructors and students on the criticality assigned the 45 attributes. Spearman's rho revealed a rank/order correlation of .89 and a Pearson correlation of .89 between the two groups. Nevertheless, there were several noticeable fractures between the two respondent groups about the essentiality of certain attributes in academic writing. Evidence of Critical Thinking received the endorsement of 79% of instructors, yet was deemed critically important by only 4 5 % of students. Evidence of Academic Inquiry was assigned critically important points by 5 1 % of instructors, as opposed to only 24% of students; Familiarity with Related Scholarship was scored critically important by 25% of instructors and only 5% of students; thirty five percent of instructors found Faithfulness to Academic Standards critically important, in contrast to 13% of students. With the exception of Critical Thinking, attributes which at least 30% of all student respondents deemed critically important were derived from Current Traditional Rhetoric. In contrast, a similar percentage of instructors labeled at least three Social Construction items critically important, suggesting some sort of a balance between Current Traditional Rhetoric and certain elements of Social Construction in the minds of instructors. Thirty seven percent of students labeled Command of Grammatical Rules critically important, in contrast to 29% of instructors. Thirty six percent of students found Command of Spelling Rules critically important, as opposed to only 27% of instructors. Students were less likely than instructors to establish discernable levels of importance among the attributes. Coherence of Essay, the attribute which generated the second largest critically important student endorsement, was judged as such by 49% of students. This was followed closely by Evidence of Critical Thinking (45%), Placement of Main Ideas, (39%), Structure of Thesis Statement (38%), Command of Grammatical Rules (37%), and Command of Spelling Rules (36%). To put the relatively narrow range of student endorsement scores in clearer terms, Evidence of Academic Inquiry, which earned the tenth highest number of critically important points was  98 deemed as such by 24% of students, only 33 percentage points lower than Clarity of Sentences and Paragraphs, the item deemed critically important by the largest number of student respondents. In contrast, there was a drop of 50% in a similar range between the items with the highest and tenth highest "critically important" points as distributed by instructors. Thus the absence of certain high scoring items may have more serious ramifications for instructors than for students. Good academic writing as conceived by instructors seems to rely on intellectual inquiry, critical reasoning, as well as knowledge of and familiarity with discipline-specific discourse. Students equated good writing with syntactical and structural precision. Thus, despite overall agreement between the two groups, the variances are noteworthy because they deal with scholarship and thought - precisely the attributes which distinguish academic writing from other genres.  Attribute Families Composition studies revolve around the notion of distinct pedagogies. Berlin's (1987) influential taxonomy conceptualized composition pedagogies as discrete entities, each defined by adherence to a central theme (i.e. objective, subjective or transactional rhetoric). Overlap is rarely discussed or acknowledged. Each pedagogy is associated with a set of features or writing attributes that further or express its pedagogical goals. My objective was to disentangle attributes from their respective pedagogies and consider the following. Here are 45 writing attributes, any of which can be used to form factors or families of writing attributes. When asked to score these individual attributes, how do instructors group them together, and in what combinations? In order to determine potential attribute families, the 45 attributes were run through several factor analytic procedures, including Varimax, Equimax, and Quartimax rotations. A Varimax rotation based on seven families which, when viewed collectively, accounted for 60% of the variance, was found to make the most sense. Only instructors' scores of importance were used in this step as they possess in-depth knowledge about academic writing that students do not and are the gate-keepers of student progress. A s such, instructors can provide informed choices about how to group attributes in ways students are unable to do at this stage of their academic  9 9  careers. The attribute families gleaned from principal component factor analysis with Varimax rotation are listed in the table below.  Table 27: Rotated Component Matrix for the Seven Attribute Families Attributes Correct Use of Prepositions Command of Punctuation Rules Correct U s e of Articles Command of Spelling Rules Adherence to Syntax Rules Command of Grammatical Rules Effective Use of Transition Words Precision of Word Choice Sophistication of Vocabulary Variety of Sentence Structures Inclusion of Author's Voice (Content) Inclusion of Author's Voice (Style) Inclusion of Author's Experiences Author's Involvement with the Inclusion of Author's Inclusion of Author's Background Incorporation of Personal Pronouns Development of Author's Point-ofHonesty of Author's Voice Attention to Reader Investigation of Power Structures Investlgtion of Society's Impact on Acknowledgement of Social Context Discussion of World Events Critical Analysis of Current Events Exam, of Society's Influence on Call for Social Reform Acknowledgement of Author's Biases Inclusion of Personal Truths Structure of Topic Sentences Unity of Paragraphs Placement of Main Ideas Proof of Organization Structure of Thesis Statement Links between Paragraphs Clarity of Sentences and Paragraphs Evidence of Academic Inquiry Familiarity with Related Scholarship Evidence of Critical Thinking Coherence of Essay Faithfulness to Academic Standards Use of Metaphor Use of Imagery Use of Academic Rhetorical Academic Discourse Conventions % of Variance Explained % of Rotated Variance  1 .89 .85 .85 .83 .75 .75 .74 .55 .51 .49  2  3  4  5  6  7  .32 .34 .31 .32 .81 .80 .62 .61 .59 .56 .54 .53 .52 .51  .39  .30  .39  .48 .37  .76 .76 .70 .63 .66 .62 .60 .59 .44  .41 .39  .31  .49 .32  .75 .74 .73 .73 .72 .61 .50  .32  .47  .34  .36 .71 .65 .59 .47 .46  .33 .39  18.62 13.58  16.01 11.04  .78 .75  6.65 10.62  6.00 8.91  4.65 5.69  3.99 4.79  .86 .83 3.42 4.70  100 The 45 writing attributes are listed in the far-left column and the names assigned to the attribute families are presented horizontally across the columns at the top of the table. The first (leftmost) factor accounts for the largest percentage of variance, the second for the second largest percentage, and so forth - both unrotated and rotated as seen at the bottom the table. The first attribute family was Mechanics and consists of ten attributes: Correct U s e of Prepositions Command of Punctuation Rules, Correct Use of Articles, Command of Spelling Rules, Adherence to Syntax Rules, Command of Grammatical Rules, Effective Use of Transition Words, Precision of Word Choice, Sophistication of Vocabulary and Variety of Sentence Structures. These attributes revolve around objective standards of correctness. Six of ten attributes contain the words "rules" or "correct," suggesting opposition to any deviation from established conventions. The positioning of Effective Use of Transition Words, Precision of Word Choice and Sophistication of Vocabulary within this family indicates word selection must conform to rigorous rules of application. Finally, Variety of Sentence Structures reifies the sentence as an objective, orderly and correct structures. While "variety" might hint at some subjectivity, a quick perusal of any grammar handbook reveals a range of pre-determined sentence structures, each of which must conform to established structural patterns to be considered readable. Mechanics emerged as the first attribute family, with the largest percentage of variance both unrotated (18.62%) and rotated (13.58%). This finding reinforces the hold mechanics has on writing classrooms in academia, and goes some way towards explaining why Expressivism and some elements of the Social Construction pedagogy are less represented in writing textbooks and curricula. The second attribute family is Author's Voice (accounting for 11.04% of rotated variance) and is made up often attributes primarily author-centred and internally driven. They are: Inclusion of Author's Voice (Content), Inclusion of Author's Voice (Style), Inclusion of Author's Experiences, Author's Involvement with the Subject, Inclusion of Author's Thoughts and Feelings, Inclusion of Author's Background, Incorporation of Personal Pronouns, Development of Author's Point-of-View, Honesty of Author's Voice, and Attention to Reader. With the exception of Attention to Reader, these attributes insist upon a tangible sense of authorial voice. The attributes in this family demand the legitimate insertion of the author into the text while requiring the author to be directly  101 and honestly involved with both subject matter and reader. Author's Voice reveals the traditional binary of author-centred/reader-centred writing to be ultimately false. Personal, author-driven academic discourse is not meant to be solipsistic, but written to be read and geared towards a reader to whom careful attention must be paid. The writer must pay attention to the insertion of voice into the text, and consider how the reader might interpret it. Social Analysis, the third family (accounting for 10.62% of rotated variance), is made up of nine attributes: Investigation of Power Structures, Investigation of Society's Impact on Knowledge, Acknowledgement of Social Context, Critical Analysis of Current Events, Discussion of World Events, Examination of Society's Influence on Language, Call for Social Reform, Acknowledgement of Author's Biases, and Inclusion of Personal Truths. With the exception of the final two attributes, this family champions awareness of the social milieu and need for change. The canvas of first-year writing is broadened from the grammatical details of Mechanics.  Instead of obsessing over the minutiae of  linguistic and syntactical rules, students must demonstrate an up-to-date and informed perception of the world beyond the university. They must be aware of inherent social flaws and rectify them. The presence of Acknowledgement of Author's Biases and Inclusion of Personal Truths challenges the arbitrary boundary between the personal and the social. The family as a whole makes clear students must apply a critical eye to their social environment, prejudices and forces that have shaped their own thinking and positionality vis-a-vis society. Paragraph Structure, the fourth attribute family (accounting for 8.91% of rotated variance), is made up of seven items: Structure of Topic Sentence, Unity of Paragraphs, Placement of Main Ideas, Proof of Organization, Structure of Thesis Statement, Links between Paragraphs, and Clarity of Sentences and Paragraphs. The overriding motif is form, with specific emphasis on the appropriate method of presenting key ideas within a paragraph and situating them next to other pivotal points or sentences. This family stresses a clear, logical and highly organized system of writing. There is little ambivalence or creative chaos: students do not have the luxury of being purposefully obscure or opaque. Instead, their writing must be structured and orderly. The fifth attribute family is Academic Inquiry (accounting for 5.69% of rotated variance) and is made up of five items: Evidence of Academic Inquiry, Familiarity with Related Scholarship, Evidence of Critical Thinking, Coherence of Essay and  102 Faithfulness to Academic Standards. These are items which instructors scored very favourably, with four of the five appearing among the attributes with the ten highest mean scores. The family posits that in order to produce effective writing, students must accomplish three tasks: display extensive knowledge about the field and its standards; demonstrate an aptitude for critical thinking; and express itself through coherence of argument. Thus, academic analysis is ineffective if couched in muddy, meandering text and consistency of argument is essentially meaningless unless used with substantial content and thought. Zinsser has argued that clear thinking becomes clear writing: one cannot exist without the other (1998, p.9). This attribute family takes the stipulation a vital step further. Clear thinking and clear writing must be used with authoritative academic scholarship. It is precisely this family that makes academic writing unique from other forms of discourse. Figurative Language, the sixth attribute family (accounting for 4.79% of rotated variance), consists of two closely related items, Use of Imagery and Use of Metaphors. Both attributes use language in a creative, subjective manner and their positioning within the same family is unsurprising. Finally, Academic Context, the seventh attribute family (accounting for 4.70% of the rotated variance), is made up of two items: Use of Academic Rhetorical Strategies and Use of Academic Discourse Conventions/This family establishes academia as a highly specialized, unique environment to which sufficient respect must be paid. Good academic writing must pay heed to the milieu (in this case, the university) in which writing is created, and to the rhetorical strategies the setting demands. Instructors' and Students' Scoring of the Seven Attribute Families Using the seven attribute families instructors had established, scale scores (based on the seven factors and their factor scores) for the importance assigned each family were calculated for the entire 674 respondent pool. Scale scores were based on the 15 importance scales used for the individual attributes (from "not at all important"=1 to "critically important"=5). Scale scores were first calculated across all attributes within a family, then divided by that number to yield a score which reflected the extent each respondent considered the attribute family important. Correlations between these scale scores and the original factor scores from which they derived ranged from .81 to .97; hence, scale scores maintained the construct validity of their factorial constituents.  103 The table below displays mean scale scores in the following ways. Overall means as assigned by the 674 respondents collectively are listed in the far-left column. On average, 663 respondents provided scores for each family. Subsequent columns demonstrate how instructors and students scored the families with standard deviations provided in parentheses. The differences between instructor and student scores are also listed, followed by an analysis of variance.  Table 28: Overall, Instructors' and Students' Assessments of Importance of Seven Attribute Families. Differences in Instructors' and Students' Assessments of Importance. Variances Accounted for in Assessments of Importance by Level Students' Means (SD) 3.80 (.55)  Mean Difference  F  •Sig  3.78  Instructors' Means (SD) 3.69 (.63)  -.11  3.19  .08  Author's Voice  3.02  2.79 (.68)  3.09 (.61)  -.30  31.32  .00  Social Analysis  2.91  2.58 (.71)  3.01 (.56)  -.43  50.47  .00  Paragraph Structure  4.07  4.11 (.60)  4.06 (.55)  +.05  .70  .40  Academic Inquiry  3.87  4.29 (.51)  3.75 (.56)  +.54  116.76  .00  Figurative Language  2.72  2.52 (.86)  2.78 (.91)  -.26  9.57  .00  Academic Conventions  3.10  2.92 (.93)  3A5(.79)  -.23  9.31  .00  Combined Attributes  3.40  3.28 (.41)  3.44 (.37)  -.12  8.34  .00  Attribute Family  Overall Means  Mechanics  Respondents overall assigned the seven families different assessments of importance. Paragraph Structure was the family with the highest overall mean score (4.07), followed by Academic Inquiry (3.87), Mechanics (3.78), Academic Conventions (3.10), Author's Voice (3.02), Social Analysis (2.91), and Figurative Language (2.72). There was a spread of 1.35 between Paragraph Structure and Figurative Language. Respondents overall assigned the Combined Attributes (importance of the seven attribute families collectively) a mean of 3.40. When instructor and student assessments were viewed separately, a different order was exposed. Instructors assigned Academic Inquiry the highest mean score (4.29), followed by Paragraph Structure (4.11), Mechanics (3.69), Academic Conventions (2.92), Author's Voice (2.79), Social Analysis (2.58), and Figurative Language (2.52). There was a spread of 1.77 between Academic Inquiry and Figurative Language.  104 Instructors assigned the Combined Attributes a mean of 3.28. Students assigned Paragraph Structure the highest mean score (4.06), followed by Mechanics (3.80), Academic Inquiry (3.75), Academic Conventions (3.15), Author's Voice (3.09), Social Analysis (3.01), and, finally, Figurative Language (2.78). There was a spread of 1.28 between Paragraph Structure and Figurative Language. Students also assigned the Combined Attributes a mean of 3.44, indicating that students generally found the attributes more important to academic writing than their instructors did. The larger spread between the families to which faculty assigned highest and lowest means of importance indicates hierarchies in the minds of instructors about the value of certain attributes over others. Students, in contrast, did not seem to have preferences that were as entrenched. Another point of dissimilarity between instructors and students lay in their assessments of the importance of such families as Academic Inquiry and Paragraph Structure. Where instructors considered Academic Inquiry to be the most important of the families, students espoused Paragraph Structure instead, demonstrating divergent opinions about the main thrust of good first-year student writing. Instructors and students most differed around the importance of Academic Inquiry with a mean differential of .54 between them. Another, albeit smaller, divergence centred around Social Analysis which received a differential of -.43 in favour of students, hence the minus sign. Students' endorsements of the following families also outweighed instructors': Author's Voice (-.30), Figurative Language (-.26), Academic Conventions (-.23) and Mechanics (-.11). Paragraph Structure on the other hand, only had a mean differential of .05, with instructors giving this family a slight edge. Analysis of variance between instructors' and students' responses to the attribute families revealed fractures in six of the eight measures (seven families and the Combined Attributes): Author's Voice, Social Analysis, Academic Inquiry, Figurative Language, Academic Conventions and Combined Attributes. The largest fracture involved the importance assigned Academic Inquiry (difference=.54; F=116.76; p<.001), a family to which instructors assigned more importance, followed by Social Analysis (difference=-.43; F=50.46; p<.001), a family to which students attached greater importance. Other fractures, in descending order of size, involved Author's Voice (difference= -.30; F=31.32; p<.001), Figurative Language (difference= -.26; F=9.57; p<.001), and Academic Conventions (difference= -.23; F=9.31; p<.001). Students  .105 assigned greater importance to all three families. In short, instructors and students were generally divided on which attributes were important to good writing. While the two groups shared similar views about the importance of Paragraph Structure, students believed first-year writing to be mechanically oriented and veered towards socially aware and writer-centred attributes. Instructors expressed a preference for thought, coherence and scholarship. Thus, where students might obsess about grammar and correct syntax, instructors are looking for critical thinking and fieldspecific research. Instructor standard deviations were greater than student standard deviations in almost every family, indicating possible fractures in instructors' assessments of the attribute families. One hypothesis is instructors have likely acquired strong preferences in the course of writing and publishing their research. Thus, divergences in assessments of the relative importances of the families are to be expected. These potential fractures are examined in "Rival Hypotheses" (Chapter Nine).  Analysis and Implications A cursory look at study findings suggest instructors and students generally perceived good writing as grammatically and structurally correct, clear, precise, and devoid of intimate disclosures or societal rebellions. A more critical scrutiny reveals fractures between instructors' and students' assessments of the importance of the attribute families. Divisions were especially palpable around attributes that distinguish academic writing from other genres, items involving critical thinking, academic inquiry and academic standards. Instructors tended to endorse these attributes strongly whereas students were more likely to consider good writing in terms of mechanics. First-year students expecting to get good grades for grammatically or syntactically perfect writing might be in for a rude shock when they discover instructors also expect them to demonstrate thought and knowledge of discourse-specific conventions. Fractures between instructors and students are better gauged when attributes are grouped as families. The attribute families are arguably more precise scales of measurement than the pedagogies for several reasons. First, while each pedagogy has a historical heritage (i.e. objective, subjective or transactional rhetoric), each nonetheless contains a range of closely linked motifs. The attribute families draw attention to these themes, instead of obscuring one or another within the larger  106 pedagogical umbrella. Current Traditional Rhetoric, for instance, contains several different foci (grammar and syntax on one hand; paragraph structure the other) while Expressivism contains at least two families within its branches: insertion of author's voice, and the figurative and creative use of language. Social Construction blurs three distinct branches: the world beyond academia; the academy as a common social milieu, and the thinking skills used in either environment. Families like Author's Voice, Social Analysis and Scholarly Inquiry improve upon the pedagogies because they reveal lines between them to be seriously blurred. Author's Voice, for example, offers a tighter, clearer focusing on elements that were previously (and incorrectly) thought to come from either Expressivism and Social Construction. Instead, the family exposes barriers between Self and Social Environment as arbitrary, highlighting the vital role each plays in the shaping of the other and the necessity for the Self to position itself knowledgeably within the Social Context. Paley (2001), in an ethnographic study of two composition instructors, discovered a similar blurring of such binary conceptions: the Expressivist instructor in her study was just as likely to prompt students to write about their social milieu as she was about their personal reflections. Similarly, Gradin (1995) has argued for a deep epistemological connection between expressivism and social construction, drawing upon romanticism to make her case. Attribute families like Author's Voice and Social Analysis provide statistical verification for a possible link between the two pedagogies, confirming Paley's findings and Gradin's thesis. Finally, attribute families provide insight into how individual attributes are grouped together in the minds of instructors. No one attribute can function independently. Instead, they must work in tandem to demonstrate an overall understanding of features instructors hold dear: grammar, structure, academic conventions and the like. The pedagogies act as a broader, blurrier categorization of a variety of attributes and cannot offer a similarly precise depiction of how faculty expect attributes to co-exist. Academic Inquiry is an excellent example of attributes functioning together. Instructors require thought, coherence and scholarship to be linked in their students' writing.  Each feature alone is an incomplete part of the equation.  Fractures between instructors and students documented thus far can be perceived in two ways. First, faculty and students generally disagreed on how to rank and score the three sample paragraphs, as "Responses to Sample Student Writing" (Chapter  107 Five") revealed. Second, as I have discussed in this chapter, instructors and students generally had differing views on the assessments of importance assigned to the 45 writing attributes and, by extension, the seven attribute families. These fissures exert powerful influence on how faculty and students conceptualize good academic writing and go some way towards explaining why so many anecdotes abound about disagreements between instructors and students. With these findings in mind, the next task was to examine which variables best predicted or explained fractures of writing. I began by documenting respondents' world views.  108 CHAPTER SEVEN: RESPONDENTS' WORLD VIEWS In addition to documenting fractures in perspectives on good academic writing, my objective was to identify whether assessments of good academic writing were more powerfully shaped by respondents' world views and personal characteristics, or their academic situations and expectations. This chapter deals specifically with respondents' world views. I designed the section of the questionnaire entitled "World Views" to determine if there was a relationship between various perspectives on good student writing uncovered in previous chapters and respondents' world views. Four world views were tested: Functionalism, Interpretivism, Radical Humanism and Radical Structuralism (see "Instrument Development," Chapter Four, for further information on the validity of the world views as a scale for measurement). A total of 80 self-descriptive adjectives were used in the questionnaire, 20 per world view (see Chapter Four for the list of adjectives). Respondents were asked: Using the following alphabetical list of words, how would you describe ways you generally think and act? They were presented with the following options: Never; Rarely; Sometimes; Often; and Always. These were translated numerically into scores of one out of five (Never) to five out of five (Always). Respondents' endorsements of the adjectives and, by extension, world views are presented in this chapter.  How Respondents Scored the Eighty Adjectives Respondent scores were divided by the number of adjectives within its respective paradigm which they had scored. Occasionally a respondent would skip an item or two, but most scored all 20. Across the 80 adjectives in the "World Views" section, respondents scored an average of 19.63 of the Functionalist adjectives, 19.69 of the Interpretive, 19.67 of the Radical Humanist, and 19.71 of the Radical Structuralist items. The Interpretive adjectives received the highest endorsement of the world views (mean=71.05). This world view was followed by Functionalism, Radical Humanism and Radical Structuralism with collective mean scores of 69.48, 66.16 and 53.18 respectively. Thus, there was strong endorsement of both the Interpretive and Functionalist paradigms, enthusiasm for Radical Humanism and tepid support for Radical Structuralism.  109 Mean scores for individual descriptors are listed in descending order in the table below together with the world view to which the adjective belonged. Standard deviations are also provided. An average of 665 respondents participated in this phase of the study. Table 29: Respondents' Endorsements of the 80 Adjectives in Descending Order Adjective  World View  Mean  SD  Caring Understanding Concerned Independent Reasonable Aware Considerate Logical Questioning Realistic Compassionate Critical Thinking Law Abiding Tolerant Intellectual Balanced Sensitive Agreeable Sympathetic Rational Perceptive Practical Problem Solving Organized Reflective Stable Proof Seeking Efficient Creative Humanistic Structured Factual Strong Instinctive Progressive Investigative Freedom Seeking Intuitive Orderly Potential Seeking Objective Thought Provoking Empathetic Self Governing Advanced Class Conscious  Interpretive Interpretive Interpretive Radical Humanist Functionalist Radical Humanist Interpretive Functionalist Radical Humanist Functionalist Interpretive Radical Humanist Functionalist Interpretive Radical Humanism Functionalist Interpretive Interpretive Interpretive Functionalist Interpretive Functionalist Functionalist Functionalist Interpretive Functionalist Functionalist Functionalist Interpretive Interpretive Functionalist Functionalist Radical Structuralist Interpretive Radical Humanist Radical Humanist Radical Humanist Interpretive Functionalist Radical Humanist Radical Humanist Radical Humanist Interpretive Radical Humanist Radical Humanist Radical Structuralist  4.02 3.98 3.95 3.93 3.92 3.90 3.90 3.90 3.88 3.87 3.86 3.81 3.81 3.80 3.76 3.76 3.75 3.75 3.74 3.71 3.69 3.69 3.69 3.67 3.66 3.66 3.63 3.58 3.58 3.58 3.58 3.57 3.56 3.54 3.53 3.53 3.53 3.52 3.51 3.51 3.51  .80 .65 .76 .75 .66 .77 .74 .73 .82 .70 .80 .80 .99 .81 .78 .79 .83 .70 .78 .76 .70 .79 .75 .94 .85 .77 .86 .81 .85 .93 .80 .80 .77 .83 .72 .81 .94 .79 .90 .89 .78 .81 .87 .90 .75 .98  3.49 3.48 3.46 3.41 3.40  110 Adjective Liberated Nurturing Intense Autonomous Artistic Pragmatic Powerful Empirical Appeasing Contentious Deferential Conflict Driven Emancipated Adversarial Libertarian Activist Controversial Solicitous Political Forceful Avant Garde Aggressive Reformist Conformist Rebellious Confrontational Radical Extreme Insurgent Antagonistic Disruptive Militant Anarchist Violent  World View Radical Humanist Interpretive Radical Humanist Radical Humanist Interpretive Functionalist Radical Structuralist Functionalist Interpretive Radical Structuralist' Functionalist Radical Structuralist Radical Humanist Radical Humanist Radical Humanist Radical Humanist Radical Humanist Radical Humanist Radical Humanist Radical Humanist Radical Humanist Radical Structuralist Radical Humanist Radical Humanist Radical Humanist Radical Structuralist Radical Structuralist Radical Structuralist Radical Structuralist Radical Structuralist Radical Structuralist Radical Structuralist Radical Structuralist Radical Structuralist  Mean 3.37 3.31 3.26 3.23 3.16 3.10 3.07 3.00 2.98 2.97 2.96 2.91 2.89 2.86 2.86 2.85 2.83 2.82 2.78 2.74 2.72 2.72 2.71 2.68 2.65 2.62 2.62 2.57 2.57 2.45 2.22 2.16 2.10 1.71  SD .82 .92 .88 .91 1.05 .81 .82 .88 .81 .84 .74 .96 .94 .81 .35 .87 .91 .84 1.07 .90 .87 .90 .87 .84 .92 .92 .92 .89 .79 .83 .82 .87 .90 .81  Of descriptors with the twenty highest mean scores, nine were Interpretive: Caring, Understanding, Concerned, Considerate, Compassionate, Tolerant, Sensitive, Agreeable, and Sympathetic. A shared theme among these nine is the emphasis on what Burrell and Morgan (1979) referred to as intersubjectively shared meaning (p.260); the implicit message is that individuals must strive towards acknowledging and understanding the diversity of opinions that make up the larger social network. The objective is not to force one's point of view upon the larger group, but to be perceptive of and charitable towards others' beliefs. Functionalist and Radical Humanist adjectives also scored well, with several items each among the descriptors with the top 20 mean scores. The Functionalist items were Reasonable, Logical, Realistic, Law Abiding and Balanced, and Rational - all adjectives that display a keen awareness of the necessity of unemotional common sense. The Radical Humanist adjectives were Independent, Aware, Questioning, Critical Thinking  111 and Intellectual, items that involve a subverting of accepted boundaries. Thus, there are three distinct themes among items with the highest mean scores: conciliatory subjectivity; objective pragmatism; and intellectual defiance. This mixture of Interpretive, Functionalist and Radical Humanist adjectives also features among items with 2 1 to 4 0 highest mean scores. There were nine s t  th  Functionalist adjectives in this portion of the table (Practical, Problem Solving, Organized, Stable, Proof Seeking, Efficient, Structured, Factual, and Orderly), six Interpretive adjectives (Perceptive, Reflective, Creative, Humanistic, Instinctive, and Intuitive), four Radical Humanist descriptors (Progressive, Investigative, Freedom Seeking, Potential Seeking) and one Radical Structuralist item (Strong). Radical Structuralist adjectives were not widely endorsed. Fourteen of the adjectives with the 20 lowest mean scores belonged to this world view. They were Adversarial, Controversial, Forceful, Aggressive, Rebellious, Confrontational, Radical, Extreme, Insurgent, Antagonistic, Disruptive, Militant, Anarchist and Violent. These adjectives push the challenge implied in the Radical Humanist descriptors beyond the realm of abstract thought towards the physical realities of revolution. While the university has a long lineage as an agent of social change (Newman, 1852/1976; Pelikan, 1992) and is frequently exhorted to play a more politically active role (see, for instance, Aronowitz, 2000; Giroux, 1994, 2001; Readings, 1996; Shor, 1992), it is fundamentally an institution that depends on the cooperative participation of members to survive. Thus, while faculty and students may encourage social analysis and critical thought, they are unlikely to advocate brutal insurrection. Finally, standard deviations generally hovered around .80 of a point with only two adjectives (Artistic and Political) achieving standard deviations of 1.0 or more. While there was some discrepancy over how much respondents identified with adjectives, these divergences were generally subdued. In general the Functionalist, Interpretive and Radical Humanist world views generated support from respondents. This range of opinion is depicted pictorially in the box plot graphs below. Figure 5 is made up of four box plots, each representing the range of scores accorded adjectives within the Functionalist, Interpretive, Radical Humanist and Radical Structuralist world views respectively. The box plots provide a quick and useful way of comparing the endorsement each world view received from respondents.  112  Figure 5: Respondents' Endorsements of the Functionalist, Interpretive, Radical Humanist and Radical Structuralist World Views  From left to right, the box plots represent endorsement ranges for the Functionalist, Interpretive, Radical Humanist and Radical Structuralist world views, in that order. Boxes depict the endorsement given by at least 50% of the respondents while lines (whiskers) on both sides of the boxes demonstrate the range that remaining respondents assigned. Oblong-shaped items above and below the whiskers indicate extreme positions taken by at least one respondent. Numbers on the right of each oblong represent the numerical code assigned respondents during data entry. The endorsement range for the Functionalist world view was between 50 to 90. At least 50% of respondents assigned it somewhere between 65 to 75 points with a median of 70. The Interpretive world view endorsement very closely matched that of the Functionalist world view: a general range of between 51 to 93 points. At least 50% of respondents assigned it between 67 and 77 points with a median of 72. The Radical Humanist world view endorsement range was between 45 to 89. At least 50% of respondents placed it somewhere between 62 and 67 with a median of 67. Finally, at least 50% of respondents assigned Radical Structuralism between 45 and 60 points with a median of 51.  113 In general, Figure 5 reifies what earlier findings implied. While the Interpretive world view received the highest majority range and median score of all four paragraphs, it was followed closely by Functionalist and Radical Humanist world views. The difference in median scores between the Interpretive and Radical Humanist world views was only five. No one world view truly dominated this respondent group. Instead, three world views exerted considerable influence on faculty and students.  Respondents' World View Profiles No two individuals see the world the same way. Each respondent's profile is unique and derived from an idiosyncratic combination of the four world views. Figure 6 illustrates these differences by charting scores for two respondents: one with a high Radical Structuralist endorsement, the other with a high Interpretive endorsement. The figure also illustrates the differences between faculty and students respondent groups. As a point of reference, the average respondent profile is also provided. On average, respondents assigned a mean of 69 to the Functionalist world view, 71 to Interpretivism, 66 to Radical Humanism, and 53 to Radical Structuralism. Instructors tended to subscribe to the Functionalist, Interpretive and Radical Humanist world views more strongly than the average respondent did. They gave the Functionalist world view a mean of 74, Interpretivism a mean of 75, and Radical Humanism a mean of 73. On the other hand, they assigned Radical Structuralism a mean of 51, a little lower than the collective average of 53. Students' world view endorsements tended to be lower than average when Functionalism (mean=68), Interpretivism (mean=70), and Radical Humanism (mean= 64) were involved. However, they were more likely to subscribe to Radical Structuralism (mean=54) than the average respondent did. Thus, where instructors were more likely to be structured, orderly and intellectually aware in their thinking, students had a tendency to be more challenging and provocative. Respondent #39 and Respondent #295 represent two individuals with very different world views. Respondent #39 strongly endorsed Radical Structuralism (72 points), followed by Radical Humanism (67), then Interpretivism (59) and, finally, Functionalism (55). In contrast, Respondent #295 strongly endorsed Radical Humanism (92) and Interpretivism (91), followed by Functionalism (76) and Radical Structuralism (63 points). To put these numbers in some perspective, Respondent #295 ascribed the Interpretive world view 32 points more than Respondent #39 did.  1 1 4  Respondents' World View Scores 1= 157 Instructors S=523 Students; Respondent # 39; Respondent #295 A=Average Profile 665 Respondents  88 85 95  83  90  80 79  / /  /' /  8 5  78 77  / /  #295 IV 75%  75  ""  88V 86 \ 83 81 79  76 ,72 \  X /  76 75 74  / 69  95  65  90 85 80  79 78  WSBSKBBKSBSm y^  7 2  \  \  75%  /HHBBHHBHHi 58  70  57 56  65  73 72  55  70 65  60  60  55  55 50%  '54 50%  71  -53 52  45  45 40  40 35  %ile 98%  \  82 80  85 80  /92 89  Radical Structuralist  Radical Humanist  [nterpretivist  Functionalist  35  68  30  30  25%  25%  20  20  15  15 62  10  10  5  2%  Functionalist  Interpretivist  Radical Humanist  Radical Structuralist  Figure 6: World View Profiles for Respondents on Average, Instructors, Students and Two Sample Respondents  115 Analysis of Respondents' World Views The section entitled World Views was based on Burrell and Morgan's (1979) 2x2 matrix involving four distinct world views. A s this chapter has demonstrated, respondents tended to endorse Functionalism and Interpretivism and, to a slightly lesser degree, Radical Humanism. Three major commonalities emerged among respondents. The first was the need to accommodate diversity and be accepting of alternate points-of-view. Such tolerance for difference bears tremendous resonance in a university as globalized and multiculturalized as U B C . The second involved an emphasis on logical, pragmatic and rational thinking, indicating a deep-rooted belief in the objectivity of research and knowledge accumulation. The third common thread was the importance of intellectual curiosity and independence. In summary, respondents revealed themselves to be caring, concerned individuals committed to the expanding and testing of knowledge and to working together. This is not to suggest differences of opinion did not exist among the 665 respondents who participated in this phase of the study. Instead, there were differences between individual respondents. The university is a mixture of convergence and divergence. Individuals within a university belong to an extended community with shared goals and visions. However, they are fundamentally individuals with unique perspectives and understandings of the nature of knowledge and their role within the larger community. The delicate balance between convergence and divergence is part of what makes the academy such a vibrant, compelling site for research and learning. However, it can also lead to deep divisions in assessments of something as intrinsic to the university as writing. I investigate these fractures and the impact world views have upon them a little later in this study. In the next chapter, "Respondents' Backgrounds," I focus on yet another way set of predictive variables: socio-demographic characteristics.  116 CHAPTER EIGHT: RESPONDENTS' SOCIO-DEMOGRAPHIC CHARACTERISTICS "Respondents' World Views" (Chapter Seven) provided an overview of respondents' world views. I discuss predictor/fracture relationships between respondents' world views and assessments of good academic writing later in this study. This chapter, however, focuses on respondents' socio-demographic characteristics, laying the groundwork for later discussions on the influence of personological and situational characteristics on fractures in perspectives on good academic writing. Table 30 launches the investigation into respondents' socio-demographic characteristics by summarizing the four research areas the section of the questionnaire entitled "Background" contained.  Table 30: Socio-Demographic Research Areas and Sub-Areas Research Area Biography Role Language Academics  Sub-Areas Gender, Age, Country of Origin, Year of Arrival in Canada Instructor/Student, Faculty Position, Department, Years at UBC First Language, Academic Language, Years Writing in English, Marking Training Highest Academic Qualification, Year of Graduation Country of Graduation  Biography: Were respondents male or female? When were they born and, if not in Canada, where? If foreign-born, when did respondents arrive in Canada? Role: What percentage of respondents were students as opposed to instructors? What position did instructors occupy? To which faculty or department did they belong and what were their academic specialties? Which department did students belong to and what were their intended majors? How many years had respondents been teaching or studying at U B C ? Language: What were respondents' first languages? In what language did they earn their highest academic qualifications? How many years had they been producing academic papers in English? Had any of the instructors received formal training in the marking of first-year English, and if so, when and were had such training occurred? Academics: What were respondents' highest academic qualifications? In which country had such qualifications been earned, and when had they graduated?  117 A total of 680 respondents participated in this study. I discuss who they were and what they revealed about themselves in the tables and discussions below.  Socio-Demographic Characteristics of Respondents Among 157 instructors, 89 (56.7%) were men and 66 (42.0%) were women. Two faculty respondents (1.3%) did not divulge gender. Of the 523 students, 229 were men (43.8%) and 289 (55.3) women. Five student respondents (.96%) did not divulge gender. Faculty ranged in age from 24 to 78 years, with a mean age of 46. Five faculty did not divulge age. Students ranged in age from 16 to 56. The mean age was 20. One student did not divulge age. The table below categorizes respondents by region of birth.  Table 31: Instructor and Student Tallies by Region of Birth Region Canada United States Western Europe East Asia Eastern Europe Latin America South/Southeast Asia West Asia. Africa Australasia Missing Total Canada East Asia South/Southeast Asia United States Eastern Europe Latin America West Asia Africa Missing Total  Frequency Instructors 87 26 21 5 5 4 4 2 1 1 1 157 Students 276 171 20 13 12 8 7 5 1 157  Percentage 55.4 16.6 13.4 3.2 3.2 2.5 2.5 1.3 .6 .6 .6 100.0 52.8 32.7 3.8 2.5 2.3 1.5 1.3 1.0 .6 100.0  The majority of faculty (55.4%) were born in Canada. Faculty also came from the United States (16.6%), Western Europe (13.4%), East Asia (3.2%), Eastern Europe (3.2%), Latin America (2.5%), South or Southeast Asia (2.5%), West Asia (1.3%), Africa  118 (.6%), and Australasia (.6%). While faculty respondents came from across the globe, most were North American or Western European. Student numbers reveal a slightly different picture. While most were Canadian born (52.8%), there was a sizable Asian contingent (37.8% collectively). Thus, there was a geographical shift from the North American/Western European nexus to Asia.  Table 32: Instructor and Student Tallies by Year of Arrival in Canada Year of Arrival 1940-49 1950-59 1960-69 1970-79 1980-89 1990-1999 2000-2005 Total 1980-89 1990-99 2000-2005 Total  Frequency Instructors 2 1 9 13 6 12 23 66 Students 17 138 80 235  Percentage 3.0 1.5 13.6 19.7 3.8 18.2 14.6 100.0 7.2 26.4 34.0 100.0  Sixty-six instructors (42.0% of total faculty respondents) had immigrated or come to Canada between 1940 to 2005. The largest sub-set within this group had arrived in Canada between 2000 and 2005 (34.8%). Close to 50 percent of student respondents had immigrated or arrived in Canada between 1980 to 2005. The biggest sub-set within this group had arrived between 1990-1999 (58.7%).  Respondents' Roles Six hundred and eighty respondents participated in the study. They comprised 157 instructors (23.1%) and 523 students (76.9%). Of the instructors, 47 (29.9%) were sessional instructors, 46 (29.3%) were tenured professors, 38 (24.2%) were on the tenure track, and 14 (8.9%) were teaching assistants. The rest of the instructor respondent group consisted of five Emeritus Professors (3.2%), three Post-Doctoral Fellows (1.9%), two Assistant Professors (1.3%) and two Senior Instructors (1.3%). The table below breaks down the number of years instructors had been teaching partor full-time at U B C .  119  Table 33: Instructor Tallies by Length of U B C Employment Years <3 4-6 7-9 10-12 13-15 16-18 Total <5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 Total  Frequency Part-Time Teaching 33 11 5 2 3 1 55 Full-Time Teaching 54 18 15 3 2 11 10 4 117  Percentage 60.0 20.0 9.1 3.6 5.5 1.8 100.0 46.2 15.4 12.8 2.6 1.7 9.4 8.5 3.4 100.0  Of the 55 instructors who had been teaching part time, 80% had done so for up to six years. Four instructors (7.3%) had been teaching part-time for more than 13 years. Of the 117 full-time instructors, 46.2% had been teaching for up to five years. This was the largest group among full-time teaching staff. Twenty three percent of the full-time instructor contingent had taught at U B C for more than 20 years. Among students, 504 (97.1%), had been studying at U B C for two years or less. Thirteen students (2.5%) had been at U B C for between three and four years. Three (.4%) had been at U B C for between five and six years. Having established the length of time respondents had been teaching or studying at U B C , Tables 34 and 35 distribute them by department or faculty.  120  Table 34: Instructor Tallies by Department Department Frequency English 43 History 14 Political Science 13 Geography 11 Social Work 10 Economics 9 Classical Studies 8 Philosophy 7 Sociology 6 Writing Centre 6 Psychology 5 Foundations 4 French, Hispanic and Italian Studies 4 Language and Literacy Education 4 Linguistics 4 Arts One 2 Central, Eastern, Northern European Studies 2 Creative Writing 2 Technical Writing 2 Art History 1 Total 157  Percentage 27.4 8.9 8.3 7.0 6.4 5.7 5.1 4.5 3.8 3.8 3.2 2.5 2.5 2.5 2.5 1.3 1.3 1.3 1.3 .6 100.0  Table 35: Student Tallies by Faculty of Enrolment Faculty  Frequency  Science Arts Applied Science Commerce Agricultural Science Human Kinetics Forestry Music Computer Science Dental Hygiene Education Fine Arts Nursing Total  219 188 39 35 14 6 5 4 1 1 1 1 1 515  Percentage 42.5 36.5 7.6 6.8 2.7 1.2 1.0 0.8 0.2 0.2 0.2 0.2 0.2 100.0  The largest instructor respondent group, 27.4%, was from the Department of English, the department responsible for teaching and running first-year English courses.  121 Remaining instructors were spread over the Faculty of Arts (62.5%), the Writing Centre (3.8%), 2 the Department of Language and Literacy at the Faculty of Education (2.5%), the inter-disciplinary Foundations program (2.5%), and 1.3% from the Technical Writing program at the Faculty of Applied Science (1.3%). Of the 515 students who provided information about the faculty in which they were enrolled, the largest student group was from the Faculty of Science (42.5%) followed by the Faculty of Arts (36.5%). The remaining students were enrolled in a variety of faculties including Applied Science (7.6%), Agricultural Science (2.7%), Human Kinetics (1.2%) and Music (.8%). Table 36 breaks down student respondents by intended major.  Table 36: Students' Intended Major by Faculty Frequency 164 97 93 50 40 35 6 3 3 491  Faculty Science Undecided Arts/Law Commerce Health Sciences Applied Science Music Education Forestry Total  Percentage 33.4 19.8 18.9 10.2 8.1 7.1 1.2 .6 .6 100.0  Of the 491 student respondents who listed their intended major, 33.4% hoped to major in the Sciences.  A little under 20% were undecided, while 18.9% intended to  major in the Arts, 8.1% in the Health Sciences, and 7.1% in the Applied Sciences. The remainder expressed a leaning towards majors in Music, Education or Forestry.  Respondents' Languages The next phase of investigation involved respondents' use of language. Seventynine percent of instructors spoke English as their native tongue, in contrast to 52.4% of students. East Asian languages were within striking distance with 37.7% of students citing these as their first languages. Table 37 provides further information on respondents' first languages. Languages are grouped according to geographic region.  122 Table 37: Instructor and Student Tallies by First Language  English Western European East Asian Eastern European South/southeast Asian West Asian African Total English East Asian South/southeast Asian Western European Eastern European Missing West Asian Total  Frequency Instructors 124 22 4 2 2 2 1 157 Students 274 197 13 13 12 6 2 523  Percentage 79.0 14.0 2.5 1.3 1.3 1.3 .6 100.0 52.4 37.7 2.5 2.5 2.3 1.1 .3 100.0  Table 38 summarizes the years respondents had spent writing academic papers in English. Table 38: Instructor and Student Tallies by Years Writing Academic Papers in English Instructors 1-5 6-10 11-15 16-20 21-25 26-30 Missing Total  17 26 50 7 25 31 1 157  1-5 6-10 11-15 Missing Total  342 150 22 9 523  Students  10.8 16.6 31.8 4.5 15.9 19.7 .6 100.0 65.4 28.7 4.2 1.7 100.0  Instructors had spent anywhere from less than five to 30 years writing academic papers in English. The largest subset within this grouping had had between 11 to 15 years writing experience (31.8%). Most students had less than five years writing experience (65.4%) with 4.2% claiming between 11 to 15 years writing experience.  123 Table 39 distributes respondents by language of highest academic qualification. Languages are categorized by geographic region. Table 39: Instructor and Student Tallies by Language of Highest Academic Qualification Language English Western European Missing East Asian Total English East Asian Missing Western European Eastern European South/southeast Asian West Asian African Total  Frequency Instructors 148 6 2 1 157 Students 463 37 8 6 4 2 2 1 523  Percentage 94.3 1.1 3.8 .6 100.0 88.5 7.1 1.5 1.1 .8 .4 .4 .2 100.0  Most instructors and students had earned their highest academic qualifications in English (94.3% and 88.5% respectively). A total of 7.1% of students had earned their highest academic qualification in an East Asian language (Cantonese, Mandarin, Japanese or Korean) compared to only .6% of instructors. Of the 154 instructors who provided information on formal marking training, 127 instructors (82.5%) had not received any such training. Twenty seven instructors (17.5%) had received marking training. Of the 22 instructors who provided additional information, 54.5% had received training between 1995 and 2004 in such varied programs as E S L instructor courses, LPI markers' training, and teaching assistants' orientation seminars. Instructors' distribution by year of marking training is summarized in Table 40. Table 40: Instructor Tallies by Year of Marking Training Year 1995-2004 1985-1994 1975-1984 1955-1974 Total  Frequency 12 4 3 3 22  Percentage 54.5 18.2 13.6 13.6 100  1 2 4 Respondents' Academic Qualifications R e s p o n d e n t s T a b l e 4 1  w e r e  a s k e d  t o p r o v i d e i n f o r m a t i o n a b o u t t h e i r a c a d e m i c q u a l i f i c a t i o n s .  c a t e g o r i z e s r e s p o n d e n t s  T a b l e 4 1 : I n s t r u c t o ra n d  S t u d e n t T a l l i e s b y H i g h e s t A c a d e m i c  Qualification  Frequency Instructors 7 25 4 107 14 157 Students 488 5 12 2 8 1 7 523  Bachelor Master Doctoral Candidate Doctorate Post Doctorate Total High School International Baccalaureate College Diploma Incomplete University Bachelor Master Missing Total  M o s t i n s t r u c t o r s h a d  d o c t o r a l d e g r e e s  f a c u l t yw i t h m a s t e r ' s d e g r e e s ( 4 . 5 % ) a n d ( 9 3 . 3 % ) , a n d  t h e r e w e r e  b a c h e l o r d e g r e e s M o s t  1 9 9 0  a n d  i n s t r u c t o r sh a d 2 0 0 5  1 9 6 0  a n d  1 9 6 9 .  w h i l e T a b l e 4 3 w e r e  a t t a i n e d .  4.5 15.9 2.5 68.2 8.9 100.0 93.3 1.0 2.3 .4 1.5 .2 1.3 100.0  ( 6 8 . 2 % ) , f o l l o w e d , s o m e  d i s t a n c e a w a y ,  ( 8 . 9 % ) , b a c h e l o r h i g h s c h o o l  b y  d e g r e e s  d i p l o m a s  o t h e r q u a l i f i c a t i o n s i n c l u d i n g c o l l e g e d i p l o m a s  ( 2 . 3 % )  ( 1 . 5 % ) . g r a d u a t e d  w i t h t h e i r h i g h e s t a c a d e m i c  ( c o l l e c t i v e l y5 7 . 9 % ) .  q u a l i f i c a t i o n s b e t w e e n  Percentage  ( 2 . 5 % ) . M o s t s t u d e n t s h a d  af e w  q u a l i f i c a t i o n a t t a i n e d .  Q u a l i f i c a t i o n  ( 1 5 . 9 ) , p o s t - d o c t o r a t e d e g r e e s  D o c t o r a l C a n d i d a t e s  t h o u g h  a c c o r d i n g t o h i g h e s t a c a d e m i c  2 0 0 0  T a b l e 4 2  a n d 2 0 0 5  M o s t s t u d e n t s h a d  q u a l i f i c a t i o n b e t w e e n  e a r n e d  t h e i rh i g h e s t  ( 8 4 . 7 % ) w i t h . 6 % o f s t u d e n t s g r a d u a t i n g  c a t e g o r i z e s r e s p o n d e n t s b y y e a r o f a c a d e m i c  d i s t r i b u t e s r e s p o n d e n t s  b y c o u n t r y / r e g i o n i n w h i c h  a c a d e m i c b e t w e e n  g r a d u a t i o n  t h e s e q u a l i f i c a t i o n s  125 Table 42: Instructor and Student Tallies by Year of Highest Academic Graduation Year of Graduation 2000-2005 1990-99 1980-89 1970-79 1960-69 Missing Total 2000-2005 1990-99 1980-89 1970-79 1960-69 Missing Total  Frequency Instructors 47 44 20 22 7 17 157 Students 443 16 2 1 3 58 523  Percentage 29.9 28.0 12.7 14.0 4.5 10.8 100.0 84.7 3.1 .4 .2 .6 11.1 100.0  Table 43: Instructor and Student Tallies by Geographic Region in which Highest Academic Qualification was Earned Region Canada United States Western Europe Missing Australasia East Asia Eastern Europe Total Canada East Asia United States South/southeast Asia Eastern Europe Latin America West Asia Africa Western Europe Missing Total  Frequency Instructors 80 49 19 5 2 1 1 157 Students 421 29 13 5 4 4 2 2 2 41 523  Percentage 51.0 31.2 12.1 3.2 1.3 .6 .6 100.0 80.5 5.5 2.5 1.0 .8 .8 .4 .4 .4 7.8 100.0  Most instructors had earned their highest academic qualification in Canada (51.0%), followed by the United States (31.2%) and Western Europe (12.1%). Students had earned their qualification in more disparate locations, including Africa (.4%), Eastern  126 Europe (.8%) and Western Europe (.4%). The overwhelming majority of students, however, had earned their highest academic qualification in Canada (80.5%) There were 150 instructors who provided information about the university from which they had received their highest academic qualification. Of these instructors, 50% were from Canadian universities, with the largest contingent from U B C (22.7%). Thirty three percent of instructors were from American schools, including such Ivy League establishments as Harvard, Princeton and MIT. Instructors from British schools made up nine percent of the instructor respondent group and included seven participants from Cambridge and Oxford. Other instructors had earned their academic credentials from Western European institutions in Austria, France, Germany and Spain, as well as from Hong Kong, New Zealand and Poland. Thus, while instructors in the participant group had collectively earned their highest academic qualifications from ten different countries, the majority were from North American schools.  Summary of Respondents' Socio-Demographic Characteristics The respondents who participated in this study were an assorted group. They came from six different continents, identified a virtual tower-of-Babel spectrum as their first languages, and were affiliated with a wide range of academic faculties or departments. There were, nevertheless, some key commonalties. Almost 8 0 % of respondents had earned their highest academic qualifications in Canada, and 90.4% had attained these credentials in English. These 680 respondents were also all associated with the University of British Columbia, and must produce or assess writing created within this shared environment. Having identified respondents' world views and socio-demographic characteristics, I focused on relationships between predictor variables and fractures in perspectives on good writing documented in earlier chapters ("Responses to Sample Student Writing" and "Responses to Attributes of Writing"). Specifically, were respondents' assessments of good writing more powerfully shaped by who they were personologically or situationally? "Rival Hypotheses" (Chapter Nine) provides answers.  127 CHAPTER NINE: RIVAL HYPOTHESES Instructors who participated in this study assessed sample student writing and writing attributes in a multitude of ways, as the findings discussed in "Responses to Sample Student Writing" (Chapter Five) and "Responses to Attributes of Writing" (Chapter Six) revealed. Divergences of opinion among first-year students are expected since they are a polyglot, heterogeneous group of individuals comprising a variety of backgrounds and experiences with presumably little in common beyond the desire to earn a degree. Divergences of opinion among instructors are a different matter. Beginning students could legitimately expect their instructors to subscribe to a common set of standards when assessing writing. However, as this study demonstrated, no such consensus existed, even among faculty from the arts or social sciences. Because instructors are gate-keepers for students' advancement through academia, how they choose to evaluate student writing has tremendous consequences. If students get consistently different messages from instructors, they are less likely to develop into competent and confident writers. Consequently, the university will have failed to produce graduates with a precise and incisive capacity for the written word. Why do instructors within the same well established Western Canadian university subscribe to differing perspectives on good student writing? When sifting through the information gathered in the "World Views" and "Background" sections of the questionnaire, I became aware of two broad predictor variables. The first was personological (Little, 2001), i.e. characteristics inherent among individuals themselves which were fundamental in shaping who instructors were beyond the influence of the university. This corresponds to the first of three levels of personality McAdams (1995) identifies for personality assessment and refers to as broad, decontextualized and generally nonconditional features of individual differences that emphasize certain of the "Big Five " personality factors (McRae & Costa, 1987) such as extraversion, dominance, or neuroticism. The second theme or system of categorization dealt with who respondents were situationally, i.e. the steps they had taken or features they possessed that either led to their membership at U B C or defined the role they played there. This theme corresponds with McAdams' second level of personality assessment which invokes the motivations, skills or other strategic constructs contextualized in time, place  128 or role. (McAdams' third level refers to a psychological integration of identity as an internalized and evolving life story and is beyond the parameters of this study). Within the context of this study, I interpreted personological variables as basic characteristics that defined who respondents were within the world (i.e. qualities they were either born with or born into), and how they perceived that world. Personological characteristics were gender, first language (English or other), country of birth (Canada or other) and world view. I defined situational characteristics as follows: U B C role (if respondents were teaching assistants, sessional instructors, tenured or tenure-track professors, other types of instructors, or students); status of U B C employment (parttime or full-time); U B C employment history; highest academic qualification; departmental grouping; country of highest academic qualification; language of highest academic qualification; year of graduation; and prior marking training. With these categories in place, I sharpened my research question accordingly. Specifically, were instructors' assessments of student writing the result of who they were personologically or situationally?  Personological vs. Situational Rival Hypotheses I built this study around the possibility of fractures (i.e. disjunctures) about what constitutes good student writing. To investigate the origins of writing fractures, I grouped personological and situational characteristics into separate categories and analyzed them using S P S S A N O V A with covariates. I tested situational characteristics collectively as A N O V A ' s fixed and random factors and entered nine personological characteristics as covariates, either as continuous variables or as dichotomies. A series of 19 independent variables were tested for detailed and differential examination of instructors' judgments of rank and grade for each writing sample and each of the seven attribute families. This arrangement enabled nine personological and ten situational characteristics to be tested as rival explanations (both individually and collectively) for variations in writing quality or paragraph preferences. Instructors were the focus of this phase of the study because they have the academic responsibility and power to assign grades and ranks to student writing; consequently, fractures among instructors carry more weight than do similar fractures among students.  129 Table 44 begins the analysis of rival hypotheses by examining the influence personological versus situational factors had upon instructors' ranking and grading of the paragraphs, as well as their scoring of the attributes of writing as a combined whole and as seven individual families. Statistically significant F-test scores are in bold.  Table 44: Variances Accounted for in Instructors' Paragraph Assessments and Importance of Writing Quality Attributes by Collective Personological and Situational Factors Total  Situational Factors (DF=21)  DF  Personological Variables (DF=8) F (.Sig)  Paragraph Assessments Current Traditional Rank Current Traditional Grade Expressivist Rank Expressivist Grade Social Construction Rank Social Construction Grade  133 133 133 133 133 133  1.13 .55 1.19 1.54 .75 1.12  (.35) (.82) (.31) (.15) (.65) (.36)  1.52 .82 1.98 1.12 1.38 .81  (.09) (.69) (.01) (.34) (.15) (.70)  Combined Attributes of Writing  133  2.90  (.01)  1.15  (.31)  Individual Attribute Families Mechanics Author's Voice Social Analysis Paragraph Unity Academic Inquiry Figurative Language Academic Conventions  134 133 133 134 134 133 133  1.18 4.73 1.07 1.66 3.13 2.70 2.06  (.32) (.00) (.39) (.12) (.00) (.01) (.05)  1.08 1.69 .79 .74 .68 2.21 1.30  (.38) (.04) (.73) (.79) (.84) (.01) (.19)  1  l^ial  Of the 14 measures used (six paragraph ranking or grading options; importance of combined attributes of writing; and importance of seven attribute families), there were six fractures between instructor operations (paragraph ranking and grading) and preferences (importance of different writing attributes). In general, situational variables had collectively more influence on "instructor operations" (paragraph rankings and scorings), while personological characteristics were collectively more predictive of "instructor preferences" (importance of different writing attributes). Situational factors (F=1.98; p<01) had statistically significant influence on instructors' ranking of the Expressivist paragraph. In contrast, personological factors had statistically significant influence on combined attributes of writing (F=2.90; p<.01) and  130 such families as Author's Voice (F=4.73; p<.001), Academic Inquiry (F=3.13; p<.001), Figurative Language (F=2.70; p<.01) and Academic Conventions (F=2.06; p<.05). These findings are replicated in Table 45, in which F-test scores for the personological and situational factors were averaged over six paragraph ranking and grading options, as well as seven attribute families Table 45: Average Variances Accounted for in Instructors' Paragraph Assessments and Importance of Writing Quality Attributes by Collective Personological and Situational Factors Six Measures of Paragraph Ranks and Grades Seven Attribute Families Total  Personological 6.28/6 = 1.05  Situational  7.63/6 = 1.23  Total 13.41 /12 = 1.16  16.53/7 = 2.36  8.48/7 = 1.21  25.02/14 = 1.79  22.82/ 13 = 1.75  16.12/13= 1.24  38.92 / 26 = 1.50  In terms of total variance accounted for, personological factors (F-Test Average=1.75) had a .51 edge over situational factors (F-Test Average=1.24). However, situational factors (F-Test Average=1.23) had .18 larger influence on paragraph ranks and grades, while personological factors had 1.15 greater influence on averaged scores among the attribute families (F-Test Average=2.36). While instructors' ranking and grading of the paragraphs was influenced by situational characteristics, their scoring of the attributes was more strongly influenced by personological factors. Previous tables hinted at limited agreement between instructor operations and instructor preferences. Not surprisingly, there were relatively few positive and statistically significant correlations between paragraphs and attribute families. Instructors who favoured a subjective, authorial voice were most likely to be consistent in their assessment of the paragraphs. Those who assigned higher importance to Author's Voice were likely to rank (r=.24) and grade (r=.25) the Expressivist paragraph highly. There was also a statistically significant correlation (r=.22) between Social Analysis and Expressivist grade, demonstrating a self/social overlap between Expressivist and Social Construction, suggesting a link between author-centred writing and social analysis. Table 46 summarizes the correlation between instructors' ranking and grading of the paragraphs and scoring of the attribute families.  131 Table 46: Correlations between Instructors' Paragraph Assessments (Ranks and Grades) and Importance of Attribute Families Expressivist Grade Rank  Attribute Family Mechanics  Current Traditional Grade Rank  -.17*  -.10  .10  .08  Author's Voice  -.18*  -.10  .24**  Social Analysis Paragraph Structure Academic Inquiry Figurative Language Academic Conventions  -.13  -.15*  .09  Social Construction Rank Grade  .14  .10  .25**  -.02  .05  .10  .22**  .06  .10  .02  .05  .11  -.06  .01  -.04  .08  -.03  -.10  .03  .01  -.05  .01  .13  .14  -.05  .00  -.01  .01  -.02  .05  .07  .06  p<.01; ** p<.001  In addition to the few statistically significant and positive correlations discussed earlier, the table reveals several negative and statistically significant correlations. One could hypothesize that instructors who favoured Author's Voice or Social Analysis would be less inclined towards the Current Traditional paragraph with its objective, authorvacated tone and relative lack of social inquiry. Consequently, the negative Author's Voice/Current Traditional ranking (r=-.18) and Social Analysis/Current Traditional grading (r=-.15) correlations are not surprising. However, the negative correlation between Mechanics and Current Traditional ranking (r=-.17) suggests instructors who claimed to endorse grammatically oriented attributes may not always rank or grade such writing accordingly. This negative correlation, combined with the relatively few statistically significant and positive correlations indicates instructors' assessment of the paragraphs did not always echo their attribute family assessment, and vice versa.  Rival Hypotheses: Statistically Significant Fractures in Instructors' Paragraph Rank and Grade A s s e s s m e n t s  My objective within this section of the study was to investigate reasons for disagreements between instructor operations and preferences. To begin, I identified the specific points in instructors' ranking and grading of the paragraphs in which personological versus situational characteristics were statistically significant.  With  these fractures established, I determined the direction in which they occurred. Table 47  132 begins this analysis by presenting the fractures; subsequent tables summarize directionality.  Table 47: Personological and Situational Influence on Instructors' Paragraph Assessments (Ranks and Grades) Category  Current Traditional  Social Construction Paragraph  Expressivist  Grade F(Sig)  Rank F(.Sig)  Rank F(.Sig)  Grade F (Sig)  .55 (.82)  1.19 (.31)  1.50 (.15)  .75 (.65)  1.12 (.36)  .17 (.69)  1.73 (.19)  1.60 (.21)  .26 (.62)  .01 (.92)  (.02)  .59 (.44)  .48 (.49)  .33 (.57)  1.99 (.16)  .98 (.32)  Radical Humanist Radical Structuralist  .54 (.46)  .00 (.99)  1.58 (.21)  1.45 (.23)  .01 (.93)  1.08 (.30)  .60 (.44)  1.93 (.17)  .81 (.37)  5.78  .07 (.79)  .19 (.67)  Gender  .11 (.75)  .36 (.55)  .47 (.23)  .73 (.39)  .63 (.43)  .00 (.98)  Canada Born  .19 (.66)  .17 (.68)  .41 (.52)  .99 (.32)  1.71 (.19)  1.64 (.20)  First Language  .39 (.53)  .04 (.95)  .11 (.74)  .61 (.44)  1.28 (.26)  2.97 (.09)  Aqe  1.28 (.26)  1.19 (.28)  2.95 (.09)  .83 (.37)  .02 (.89)  2.06 (.15)  Situational Variables Collectively  1.52 (.09)  .82 (.69)  (.01)  1.12 (.34)  1.38 (.15)  .81 (.70)  Grade F (Sig)  Rank F(.Sig) Personological Variables Collectively Functionalist Interpretive  1.13 (.35)  .55 (.46) 5.40  (.02)  J  1.98  Role Full/Part Time Employment  .12 (.89)  .72 (.49)  .58 (.56)  1.23 (.30)  1.38 (.15)  .50 (.61)  1.13 (.29)  .02 (.90)  11.55(.00)  2.70 (.10)  4.17  (.05)  .41 (.52)  Qualification  1.39 (.24)  .23 (.92)  2.06 (.09)  1.26 (.29)  1.54 (.20)  .96 (.43)  Departmental Grouping  3.48  (.02)  .68 (.57)  .65 (.59)  .32 (.81)  2.52  (.05)  2.04 (.10)  .27 (.60)  Length of U B C Employment Country of Qualification Academic Language Year of Graduation Marking Training  (.02)  1.44 (.26)  .44 (.78)  1.38 (.25)  3.62  (.01)  1.20 (.31)  .05 (.83)  3.13 (.08)  2.83 (.10)  .40 (.53)  .51 (.48)  1.13 (.29)  .08 (.78)  .01 (.91)  2.85 (.10)  .01 (.94)  .22 (.64)  .54 (.71)  .34 (.85)  .34 (.85)  .18 (.95)  .64 (.63)  .27 (.90)  .90 (.35)  3.17 (.08)  .10 (.75)  1.17 (.28)  .87 (.35)  .05 (.83)  3.52  133 When grouped collectively, situational rather than personological characteristics had more of a statistically significant influence on instructor operations, particularly in their ranking of the Expressivist paragraph (F=1.98; p<.01). When viewed as individual variables, additional variances could be observed. The first involved world view. Current Traditional ranking was most influenced by the Interpretive world view (F=5.40; p<.02) while Expressivist grading was most affected by a Radical Structuralist world view (F=5.78; p<.02). Second, there was a statistically significant relationship between situational factors such as employment status, departmental grouping and length of U B C employment and instructors' paragraph ranking and grading. Instructors' ranking of the Current Traditional paragraph could best be attributed to departmental affiliation (F=3.48;=.02) and length of U B C employment (F=2.52; p<.05). Expressivist ranking most correlated with whether instructors held full-time or part-time positions at U B C (F=11.50/ p<.001; this was the largest F-test score generated in this phase of data analysis), and by the length of U B C employment (F=3.52; p<.01). Social Construction paragraph ranking was best explained by instructors' full-time or part-time employment status (F=4.17; p<05) and their departmental affiliation (F=3.62; p<.02). Having established the points at which fractures occurred, my attention focused on relationships between world views, employment status, departmental grouping and length of U B C employment and paragraph ranks and grades. Table 48 begins the examines the correlation between world views and paragraph ranks and grades.  Table 48: Correlations between Instructors' Four World Views (Functionalist, Interpretive, Radical Humanist and Radical Structuralist) and Paragraph Assessments (Ranks and Grades) World View Functionalist  Current Traditional Rank Grade .20* .06  Interpretive  -.11  .07  .07  .18*  .21**  .30**  .02  .17  .06  22**  .09  .27**  -.03  .16  .07  .26**  -.02  .20*  Radical Humanist Radical Structuralist  * p<.01; ** p<.001  Expressivist Social Construction Rank Grade Rank Grade .21** .08 -.02 .12*  134 Findings revealed a correlation between instructors' world view and paragraph assessment. The Interpretive world view registered positive and statistically significant correlations with the Expressivist grade (r=.18), Social Construction rank (r=.21) and grade (r=.30), but correlated negatively with the Current Traditional rank (r=-.11, p<.02). Thus, an instructor who manifested an Interpretive world view was more likely to assign a higher rank or grade to the Expressivist and Social Construction paragraphs than the Current Traditional paragraph. Instructors holding a Radical Structuralist world view showed higher Expressivist grading (r=.26, p<.02). The orderly, logical Functionalist world view correlated significantly with the objective, structured Current Traditional sample (Functionalism/Current Traditional grade r=.20) while the challenging, socially astute Radical Humanist and Radical Structuralist world views correlated significantly with the socially aware Social Construction paragraph (Radical Humanist/Social Construction grade r=.27); Radical Structuralist/Social Construction grade r=.20). Table 47 showed that U B C employment status, departmental grouping, and length of employment at U B C hinted at fractures in instructors' ranking of the paragraphs. Table 49 breaks down the mean scores generated by these variables.  Table 49: Differences in Instructors' Paragraph Assessments by Situational Characteristics (Employment Status, Departmental Grouping, and Length of U B C Employment) Situational Factors  Full/Part Employment (154) Part-Time (54) Full-Time (100) Department Grouping (155) Foundations of Thinking (26) Languages/Language Arts (32) English (42) Natural/Social Sciences (55) Length of UBC Employment (155) > 5 years (80) 5-10 years (22) 10-20 years (25) 20-30 years (14) 30 years s (14)  Current Traditional  Expressivist  Rank  Grade  Rank  Grade  Social Construction Grade Rank  2.19 2.33  7.80 7.90  1.83 1.52  7.36 6.94  2.04 2.16  7.76 7.75  2.50 2.09 2.05 2.47  7.92 7.73 7.76 8.01  1.54' 1.75 1.48 1.71  7.12 7.22 7.05 7.02  2.08 2.16 2.45 1.85  7.60 7.89 8.06 7.52  2.40 2.27 2.32 2.00 2.28  8.04 7.55 7.86 7.79 7.87  1.46 1.68 1.84 1.86 1.63  6.94 6.91 7.34 7.25 7.08  2.14 2.05 1.96 2.21 2.12  7.82 7.34 7.74 7.89 7.76  135 Mean ranks, unlike grades, provide more restrictive and simultaneously less calibrated view of the variances in instructors' assessments. Nevertheless, they offer a useful starting point for distinguishing writing fractures. Instructors showed differences in their rankings of the Expressivist and Social Construction paragraphs by employment status. Part-timers tended to rank the Expressivist paragraph more strongly than their full-time colleagues did (p<.001). Fulltime faculty generally ranked the Social Construction paragraph higher (p=.05). Differences in instructors' ranking of the Current Traditional and Social Construction paragraphs might be partly explained by departmental grouping. I classified instructors into four roughly equal-sized groups in terms of field of study: Foundations of Thinking (Art History and Visual Studies; Arts One; Foundations; History; Philosophy), Languages and Language Arts (Asian Studies, Central and Eastern European Studies; Classical Studies; Creative Writing; French, Hispanic and Italian Studies; Language and Literacy Education; Linguistics; the Writing Centre), English, and Natural and Social Sciences (Applied Science; Economics; Geography; Political Science; Psychology.; Sociology; Social Work). Faculty associated with Foundations of Thinking and the Natural and Social Sciences tended to assign the Current Traditional paragraph somewhat higher ranks (mean=2.50 and 2.47 respectively) while faculty from Languages and Language Arts as well as English generally ranked this paragraph lower (mean=2.09 and 2.05 respectively; p=.02). Meanwhile, faculty from the English department assigned the Social Construction paragraph a high ranking (mean=2.45), in contrast to instructors from the Natural and Social Science departments (mean=1.85; p=.02). Finally, instructors displayed differences in their ranking of the Current Traditional (p=.05) and Expressivist paragraphs (p=.01) by length of U B C employment. Faculty who had been teaching at U B C for up to five years generally ranked the Current Traditional paragraph high (mean=2.40) while those who had been teaching for between 20 to 30 years assigned it a lower rank (mean=2.00). Faculty who had been teaching at U B C for up to five years ranked the Expressivist paragraph low (mean=1.46) while those who had been teaching on campus for between 20 to 30 years ranked the paragraph higher (mean=1.86).  136 Rival Hypotheses: Statistically Significant Fractures in Instructors' Attribute Family Assessments Having established that instructor operations (i.e. ranking and grading) were generally influenced by situational factors, I investigated instructor preferences (i.e. scoring of the writing attributes). In particular, I ascertained points where personological versus situational factors resulted in statistically significant fractures in instructors' assessments of the importance of the writing attributes, first combined then family-byfamily. Combined attributes refer to how each instructor scored the importance of all 45 writing attributes (or seven attribute families) collectively. Combined attributes thus function as an overall measure of attribute importance, gauging instructors' overall endorsement or lack thereof of the importance of attributes as a whole. Were some instructors more supportive of the importance of the attributes collectively than others, and who might they be? Table 50 begins the analysis by presenting an overview of these fractures. Subsequent tables summarize the direction in which they occurred.  Table 50: Personological and Situational Impact on Instructors' Assessments of Importance of Combined Attributes of Writing Variable Personological Variables Collectively Functionalist Interpretive Radical Humanist Radical Structuralist Gender Country of Birth First Language Situational Variables Collectively Role Full-Time/Part-Time Employment Highest Academic Qualification Departmental Grouping Length of UBC Employment Country of Academic Qualification Academic Language Year of Graduation Marking Training  Combined Attributes of Writing F (.Sig) DF 8 2.90 (.01) 3.27 (.07) 14.26 (.00) .01 (.92) .43 (.51) .22 (.64) .14(71) 4.37 (.04) 1.15 (.31) 1.77 (.18) 2.14 (.15) .15 (.96) 1.23 (.30) 1.32 (.27) 6.88 (.01) .09 (.80) .24 (.92) .98 (.33)  21 2 1 4 3 4 1 1 4 1  137 Personological characteristics collectively (F=2.90;df=8;p<.01) had greater influence on instructors' perspectives on the importance of the combined attributes of writing than did situational characteristics collectively (F=1.15;df=21; p=ns). Within the personological grouping, there were two statistically significant fractures: the Interpretive world view registered the highest F-test score of the characteristics tested (F=14.26;df=1 ;p<.001), followed by whether or not instructors spoke English as their first language (F=4.37;df=1 p<.04). Within the situational grouping, whether or not instructors earned their highest academic qualification in Canada registered the largest statistically significant fracture (F=6.88; df= 1; p<.01). Having identified these fractures, I investigated the direction in which they occurred. The importance of the combined attributes of writing correlated most strongly with the Interpretive world view (r=.28; p<.001), followed by the Functionalist (r=.16; p<.07), Radical Humanist (r=.14; p<.92) and Radical Structuralist (r=.01; p<.51) world views. Thus, instructors who endorsed the Interpretive world view were most likely to assign the highest assessments of importance to the combined attributes of writing, reinforcing the tolerant, inclusive nature of this world view. Respondents identified with Radical Structuralism, however, appeared not to care one way or another about the importance of the combined attributes. Instructors whose first language was not English tended to score the combined importance of writing attributes higher (mean=3.34) than those whose first language was English (3.26), a differential of .08 (p=.04). Faculty whose highest academic qualification was earned in Canada generally scored the combined qualities higher (3.33) than those who had graduated outside of Canada (3.20), a differential of .12 (p=.01). Table 51 narrows the analysis by summarizing the influence of personological vs. situational variables on the seven attribute families which made up the combined measure of importance. The table reveals that, when viewed collectively, personological variables had greater influence on instructors' assessments of the attribute families' importance than situational variables.  138 Table 51: Personological and Situational Influence on Instructors' Assessments of Importance of Seven Attribute Families  Personological Variables Collectively Functionalist Interpretive Radical Humanist Radical Structuralist Gender Canadian Bom First Language Aqe Situational Variables Collectively Role Full/Part Time Employment Qualification Departmental Grouping Length of UBC Employment Country of Academic Qualification Academic Language Year Graduated Marking Training  Paragraph Academic Social Mechanics Author's Analysis Structure Inquiry Voice F(.Sig) F (Sig) F(.Sig) F (.Sig) F (Sig) 3.13(.00) 1.66(.12) 4.73(.00) 1.07(.39) 1.18 (.32)  Figurative Language F(Sig) •  Academic Conventions F(.Sig)  72(.40)  11.23 (.00)  2.70(.01)  2.06 (.05)  .00 (.99) 2.54 (.11) 3.37 (.07)  .41 (.52) 5.75(.02) .63(.43) 15.85(.00) 3.78(.06) .01 (.94) .26(.61) .52(.48)  8.82(.00) 4.64(.03)  10.81 (.00)  .47(.49)  .05 (.83) 2.45 (.12)  .20 (.65)  6.89(.01) 2.65(.11) 3.22(.08)  4.75(.03)  3.99 (.05)  .27 (.60)  .06(.81)  5.11 (.03)  .30 (.59)  .14(71) .18 (.68)  .45(.51)  .16(.69)  1.14 (.29) .51 (.82)  2.56(.11) 3.23(.08)  .32(.57) .08(78)  .25(.62) .90(.35)  4.68(.03)  2.16 (.15)  4.65(.03)  .66(.42)  ,.45(.51)  1.86(.18)  .00(1.00)  1.53 (.22)  .00(1.00) 1.08 (.40)  3.71 (.06)  .63(.43) 79(73)  1.84(.18) 74(79)  .07(.80) .68 (.84)  .23 (.63)  .65 (.42) 1.30 (.19)  2.78 (.07) .37 (.54)  4.79(.01) 2.4.6(.09) 5.96(.02) .28(.60)  .35(71) .14(71)  .52 (.60) .06 (.81)  2.83 (.06)  .48 (.75) 2.12 (.10)  1.75 (.14) 1.26 (.29)  .13(.97) .30(.82)  78(.54) .87(.45)  .09 (.99) 1.38(.25)  1.27 (.29) 1.27 (.29)  5.12 (.00)  1.14 (.34)  .80 (.53)  .42(.80)  .40(.82)  1.07(.38)  4.03 (.00)  .06 (.99)  .79 (.38)  1.49 (.23)  1.96(.16)  .40(.12)  .94(.33)  3.67 (.06)  1.65 (.20)  .10 (.76)  .03 (.86)  .28(.60)  1.70(.20)  .05(.83)  .05 (.82)  .41 (.52)  .69 (.60)  .99 (.42)  .51(73)  .46(77)  .17(.95)  1.49 (.21)  .37 (.83)  .29 (.59)  .55 (.46)  .02 (.88)  2.54 (.11)  1.69(.04)  3.95(.05) 1.51 (.22) 2.84(1.00)  2.21 (.01)  6.04 (.02)  1.99 (.14) .26 (.61) .35 (.85)  Instructors' assessments of importance of Author's Voice was best explained by personological variables collectively (F=4.73; p<001), faculty's Interpretive (F=15.85; p<.001) and Radical Structuralist world views (F=6.89;p=.01), first language (F=4.65; p=.03), situational variables collectively (F=1.69; p<.04), role (F=4.79; p=.01) and employment status (F=5.96; p=.02). Faculty's assessments of Social Analysis could be accounted for by whether they had received marking training (F=3.95; p=.05), and Paragraph Structure by respondents' Functionalist world view (F=5.75; p=.02).  139 Assessments of importance of Academic Inquiry could be best attributed to personological variables collectively (F=3.13; p<.001), faculty's Functionalist (F=8.82; p<.001), Interpretive (F=4.64; p=.03), and Radical Structuralist (F=4.75; p=.03) world views and country of birth (F=4.68; p=.03). Importance of Figurative Language could be best explained by personological variables collectively (F=2.70; p=.01), the Interpretive (F=10.81; p<.001) and Radical Structuralist (F=3.99; p=.05) world views, gender (F=5.11; p=.03), country of academic qualification (F=2.21; p=.01), employment status (F=6.04; p=.02) and length of U B C employment (F=4.03; p<.001). Finally, assessments of importance of Academic Conventions was best explained by personological variables collectively (F=2.06; p=.05), the Functionalist world view (F=11.23; p<.001) and departmental grouping (F=5.12; p<.001). These associations are tested further in the table below.  Table 52: Correlations between Instructors' Scoring of the Seven Attribute Families' Importance and World Views (Functionalist, Interpretive, Radical Humanist and Radical Structuralist) Academic Conventions  World V i e w  Mechanics  Author's Voice  Social Analysis  Paragraph Structure  Academic Inquiry  Figurative Language  Functionalist  .04  .14  .11  .21**  .16*  -.03  .13  Interpretive  .08  .32**  .24**  .13  .06  .16*  .12  Radical Humanist  -.06  .21**  .19*  .13  .02  .00  .12  Radical Structuralist  -.12  .19*  .14  -.03  -.14  .07  -.06  * p<.01; **p<.001 Correlations between world views and importance of attribute families reinforce findings summarized previously while suggesting other links between world views and perspectives on good writing. Author's Voice correlated well with the Interpretive (r=.32), Radical Humanist (r=.21), and Radical Structuralist (r=.19) world views. Social Analysis generated statistically significant correlations with Interpretivism (r=.24) and Radical Humanism (r=.19). Paragraph Structure and Academic Inquiry both correlated well with Functionalism (r=.21 and .16, respectively). Figurative Language correlated significantly with Interpretivism (r=.16), as did Academic Conventions with Functionalism (r=.13).  140 These findings posit a relationship between one's view of the world and preference for writing attributes. Faculty espousing the conservative Functionalist world view preferred writing that demonstrated structure, formal investigation and discourse conventions. Faculty endorsing the Interpretive world view with its interest in individual expression and social harmony were more likely to prefer subjective content, social analysis and metaphorical language. Instructors within the socially challenging, freedom-seeking Radical Humanist world view tended to prefer authorial expression and social analysis. Having established attribute family/world view relationships, I examined the direction in which fractures in instructors' assessments of the importance of the attribute families occurred by gender, country of birth, first language, U B C role, part/full-time employment status, length of U B C employment, departmental grouping and marking training. I examined three personological variables in this phase of investigation: gender, country of birth and first language. Each of these variables had statistical influence on only one attribute family at a time. I also examined five situational variables: U B C role, employment status, departmental grouping, years at U B C and marking training. These associations are presented in Table 53.  141 Table 53: Instructors' Assessments of Seven Attribute Families' Importance by Gender, Country of Birth, First Language, U B C Role, Part/Full-Time Employment Status, Departmental Grouping, Length of U B C Employment, and Marking Training  Male (88) Female (66)  Mechanics  Author's Voice  Social Analysis  3.66 3.78  2.82 2.72  2.55 2.72  Paragraph Structure  Gender (154) 4.08 4.14  Academic Inquiry  Figurative Language  Academic Conventions  4.31 4.28  2.69 2.33  2.78 3.07  Country of Birth (156) Canadian-born (87) Non-Canadian (69)  3.71  2.88  2.66  4.11  4.22  2.63  3.02  3.70  2.62  2.54  4.10  4.38  2.40  2.77  English (124) Other (33)  3.70 3.73  2.75 2.81  4.23 4.53  2.54 2.47  2.88 3.03  Tenured (51) Tenure Track (40) Non Tenure (66)  3.83 3.62  2.73 2.55  First Language(157) 2.59 4.04 2.72 4.15 UBC Role(157) 2.50 4.05 2.61 4.07  4.28 4.37  2.69 2.26  2.72 2.81  3.66  2.92  2.71  4.25  2.56  3.11  Part-Time (54) Full-Time (100)  3.65 3.73  3.05 2.61  Full/Part Employment Status (154) 4.27 2.71 4.17 4.31 2.57 4.06  2.75 2.40  3.09 2.81  Foundations of Thinking (27) Languages/ Language Arts (32) English (43) Natural/Social Sciences (55)  3.86  2.71  Departmental 2.74  (157) 4.40  2.63  2.39  3.83  2.92  2.67  4.16  4.34  2.61  3.19  3.68 3.59  2.82 2.66  2.59 2.54  4.07 4.04  4.16 4.35  2.38 2.56  3.35 2.66  0-5 6-10 10-20 20-30 30-above  3.56 3.87 3.70 3.99 4.00  2.73 2.76 2.83 2.60 3.02  2.35 2.39 2.64 2.71 3.27  2.94 2.98 3.02 2.75 2.60  Trained (27) Non-Trained (130)  3.67 3.72  2.81 2.75  2.43 2.55  3.31 2.82  4.17  Grouping 4.23  Length of UBC Employment 4.35 2.65 4.10 4.25 2.56 4.14 4.16 2.65 4.04 4.23 2.37 4.10 4.33 2.67 4.16 Marking Training (157) 4.26 2.46 4.15 4.30 2.65 4.09  Gender had statistical influence on only Figurative Language (p<.03). This family was assigned more importance by male faculty (2.69) than by female instructors (2.33), a differential of .36 on a five-point scale. Similarly, country of birth showed impact only on Academic Inquiry (p<.03). This family was assigned higher importance by nonCanadian born faculty (4.38) rather than the Canadian born (4.22), a difference of .16.  142 First language only demonstrated impact on Author's Voice (p<.03). This family was assigned more importance by faculty whose first language was not English (2.81) than by native English speakers (2.75), a difference of .06. In terms of situational variables, U B C role had statistical influence on only Author's Voice (p<.01), which was assigned more importance by non-tenured faculty (2.92) than by tenure-track instructors (2.55), a spread of .37. Employment status demonstrated influence on Author's Voice as well (p<.02). Part-time faculty assigned more importance to this family (3.05) than their full-time colleagues did (2.61), a difference of .44. Employment status had an influence on Figurative Language (p<.02), with part-time faculty assigning more importance (2.75) than full-time faculty did (2.40), a differential of .35. Departmental grouping demonstrated influence on Academic Conventions only (p<.001). This family received its highest score from faculty from the English department (3.35) and its lowest from departments categorized as Foundations of Thinking (2.39), a difference of nearly a full point and the largest gap between instructors in this phase of data analysis. Years at U B C had statistical influence on Figurative Language (p<.001), which received its highest assessments from faculty who had been at U B C for more than 30 years (3.2/) and its lowest from those who had been at U B C for less than and up to five years (2.35), a difference of .92. Finally, whether or not instructors had received marking training demonstrated influence on Social Analysis (p=.05), with faculty who had not received training in marking student writing assigning higher importance to this family (2.65) compared to those who had (2.45), a difference of .19.  Summary of Key Findings In this phase of investigation, I tested nine personological variables (personological variables collectively, Functionalism, Interpretivism, Radical Humanism, Radical Structuralism, gender, country of birth, first language and age) and ten situational variables (situational variables collectively, U B C role, employment status, highest academic qualification, department, number of years at U B C , country of academic qualification, year of graduation, academic language and marking training). I tested the relationship between these predictor variables and ranks and grades instructors  143 assigned the three paragraphs, as well as assignments of importance they accorded the 45 writing attributes and seven attribute families. Overall, situational variables had more influence on instructors' ranking and grading operations than did personological variables. Of 60 possible relationships (ten situational variables x 6 paragraph assessments), seven statistically significant relationships were evident - a rate of 11.67% on instructor operations. The variables involved were situational variables collectively, employment status, departmental grouping and length of U B C employment. These variables had an impact on instructors' ranking of the three paragraphs. With regard to personological variables, of 54 possible relationships (nine personological variables x six paragraph assessments), only two statistically significant relationships were evident - a rate of 3.70% on instructor operations. The predictor variables involved were the Interpretive and Radical Structuralist world views. Conversely, personological variables had more influence than situational variables on instructors' assessments of importance of the seven attribute families. Of a possible 63 relationships (nine personological variables x seven attribute families), there were 16 significant relationships between personological variables and instructors' assessments of the attribute families - a rate of 25.39% on instructor preferences. The personological variables involved were personological variables collectively, world views, gender, country of birth and first language. The attribute families that were most frequently influenced by these variables were Author's Voice and Figurative Language. With regard to situational variables, of 70 possible relationships (ten situational variables x seven attribute families), only seven statistically significant relationships were evident - a rate of ten percent on instructor preferences. The variables concerned were situational variables collectively, U B C role, employment status, departmental grouping, length of U B C employment, and marking training. Why did so many fractures involve Authors Voice and Figurative Language and not families like Mechanics or Paragraph Structure? One explanation could be that grammar and structure have exerted such a long, historical hold on instructors' understanding of good student writing that there is generally consensus about their importance. In contrast, the subjectivity of the author and of language still occupies a contentious presence in academic discourse.  144 Finally, when designing the questionnaire for this study, I assumed that major fractures would involve gender and age. Instead, these variables had relatively muted impact (in the case of age, no impact at all). Fractures did, however, involve world views and outsider/insider themes, particularly surrounding place of birth (Canadian versus foreign-born), language (native speaker versus ESL-speaker) and employment status (part-timer versus full-timer). These are the characteristics that appear to divide faculty; their resonance in writing classrooms is an area deserving of much future study.  Analysis and Implications Why do instructors from within the same university and from academic programs with similar epistemological, historical and cultural traditions respond to writing samples and attributes in such a conflicted variety of ways? Instructors' perspectives on good academic writing are shaped by who they are situationally, i.e. their role and level of participation within the university, as well as who they are personologically, i.e. characteristics they have developed or were born with that transcend or pre-date academia. When instructors assess student writing, they are entitled to assign a grade because they have been hired by the university to do so. However, their response might also be tempered by socio-demographic characteristics that have little to do with the university - their country of origin, for instance, or way they see the world. This internal dichotomy can express itself in different ways. When an instructor is presented with a piece of student writing to assess, he or she is governed by two different and sometimes opposing motivations. On one hand, there are instructor preferences. What an instructor likes or dislikes for reasons that may have very little to do with the academy; on the other, there are instructor operations n the official business of the academy and the assigning of grades to student work. Whether or not these two sides can be reconciled and how this reconciliation is to take place results in the tension which constantly surrounds assessments of writing. As I demonstrated in this chapter, when instructors had to rank and grade what they assumed were samples of student writing, they appeared influenced by situational factors and responded in their capacity as faculty, perhaps mindful of their pedagogical responsibilities and philosophies. When they were asked to score the writing attributes, however, they were informed by personological factors and operated as individuals with idiosyncratic world views and personal histories.  145 On a positive note, this almost equal distribution suggests that no matter how long instructors are part of the academy, they never relinquish the characteristics that make them unique or subscribe entirely to institutional dictates. On a less-than-positive note, this means instructors are likely to be self-contradictory when assessing student writing. They may not always grade what they like highly, or vice versa, to the further confusion of students receiving those grades. As the relatively few statistically significant and positive paragraph/attribute family correlations revealed, instructors were not always consistent in their ranking and grading of the paragraphs on one hand, and their scoring of the attribute families on the other. The one exception was the consistency between Author's Voice and the Expressivist paragraph. Instructors who endorsed the Expressivist sample were most likely to score Author's Voice highly. This consistency could be due to the somewhat marginalized position this pedagogy occupies. Supporting subjective, author-centred writing means incurring ridicule and, in some cases, condemnation from the rest of the academic community; thus, instructors who sanction this writing style are more likely to do so with passionate conviction. In keeping with the debate surrounding Expressivist-derived writing, most of the fractures uncovered in this chapter revolved around the Expressivist paragraph and such attribute families as Author's Voice and Figurative Language. The Expressivist paragraph was ranked highly by part-time faculty, those who had taught at U B C for more than 10 years, and who were affiliated with the Radical Structuralist world view. In contrast, the Current Traditional and Social Construction paragraphs were ranked highly by full-time faculty; the former by faculty from the English department, the latter by instructors from departments categorized as Foundations of Thinking (i.e. Art History and Visual Thinking, Arts One, Foundations, History and Philosophy). Author's Voice was statistically preferred by part-time faculty, instructors without tenure, and those who did not speak English as their first language, reinforcing the theme of outsidership. Finally, Figurative Language was endorsed by male faculty, part-time instructors, and faculty who had been teaching at U B C for more than 30 years. In contrast, this family received poor scores of importance by instructors who had been teaching at U B C for no more than five years. One might surmise that instructors who, for one reason or another do not belong to the status quo, or who are close to retirement are willing to encourage experimentation in student writing.  146 The reasons why faculty endorse one style of writing over another are complex and result from the delicate interplay between institutional role and private self. Consequently, a uniform, dissension-free, university-wide perspective on good academic writing is unlikely. Instead, as this study has repeatedly demonstrated, fractures are a tangible presence in assessments' of student writing, and there are numerous alternative and legitimate perspectives on good academic writing. In "Fractures in Perspectives on Good Academic Writing" (Chapter Ten), I delve into these themes more deeply.  147 CHAPTER TEN: FRACTURES IN PERSPECTIVES ON GOOD STUDENT WRITING This has been a study documenting fractures in perspectives on good academic writing. Previous chapters have revealed fractures among faculty and student respondents. In this chapter, I concretize the discussion of fractures in two ways. First, using statistical analysis, norms sheets and dendograms showing the underlying structure of the attribute families, I provide a pictorial representation of fractures and their impact on assessments of student writing. Second, I connect these fractures and findings to the major themes of this study. I began this study with the observation that there were divisions between instructors and students, and especially among faculty, about what constitutes good academic writing. This observation was reinforced by the scholarship of researchers like Bartholomae, Bizzell, Elbow, Britton and particularly Berlin on distinct composition pedagogies. Study findings demonstrated that such divisions exert a compelling influence on how hundreds of faculty and students at U B C assessed first-year writing.  Fractures of Writing between and among Respondents A major subtext of this study has been the difficulty beginning students face as they grapple with writing conventions and expectations of the academy. A s extensive scholarship has demonstrated, there are definite gaps between what instructors and students know about good academic writing (Bartholomae, 1988; Bizzell, 1986; Clark & Ivanic, 1997; Ivanic & Lea, 2006; Lea, 1994; Lea & Street, 2000; Lillis, 1997; Lilis, 2006; Street, 1999). The findings of this study confirm the scholarship by demonstrating fractures between faculty and students. A s "Responses to Sample Student Writing" (Chapter Five) and "Responses to Attributes of Writing" (Chapter Six) made clear, instructors and students did not consistently agree on the ranking and grading of sample paragraphs, or scoring of the 45 writing attributes. Figure 7 pinpoints these fractures in assessment through a norms sheet detailing how instructors' and students' differed on various measures of the importance of writing attributes. The numbers on the far left and right of the figure report cumulative percentages from the respondents collectively. Gaps between lines are fractures.  148  Statistically significant fractures are bolded and underlined. Scale points were multiplied by 1 0 to avoid decimal points.  Writing Quality Attributes of Student Composition Name:  (N=680)  F=157 Faculty; S=523 Students  95 90 85 80 75% 70 65 60 55 50% 45 40 35 30  27  25  37 34  34  25%  24 36  33  20  32  24  23  15  31  23  22  35  31  10  30  21  20  33  30  32 15  20  10  31  5  28  19  17  30  28  15  2% 1%  26  14  14  28  24  10  24  13  13  24  22  Mechanics  Author's Voice  Social Analysis  Parag. Structure  Acad. In  Figure 7: Fractures in Assessments of Student Writing between Instructors and Students  149 On average, the 680 respondents assigned the following scores of importance to the attribute families: 37 for Mechanics, 30 for Author's Voice, 28.5 for Social Analysis, 40 for Paragraph Structure, 38 for Academic Inquiry, 27 for Figurative Language, 28.5 for Academic Conventions, and 34 for all 45 writing attributes Combined. There were statistically significant differences between instructors' and students assessments of the importance of the following families: Author's Voice (F=31.32; p<001), Social Analysis (F=50.45; p<.001), Figurative Language (F=9.57; p<.001), Academic Conventions (F=9.31; p<.001), Academic Inquiry (F=116.76; p<.001) and Combined Attributes (F=8.34; p<.001). Students on average assigned Author's Voice 30.5 points while instructors gave it only 27.5. Students assigned Social Analysis a mean of 30, in contrast to instructors (mean=26). Students assigned Figurative Language a mean of 31 while faculty allotted it a mean of 25. Academic Conventions received a mean of 32 from students but only 29 from faculty. The most compelling fracture involved differences between students' and instructors' scoring of Academic Inquiry. While instructors assigned this family a mean of importance of 43, students assigned it a mean of 37, suggesting clear divergence between faculty and student about the definers of good academic writing. Academic Inquiry earned the highest mean score of importance among instructors, suggesting it is the bedrock of academic writing. Students, however, did not espouse this view. Academic Inquiry thus represents a central fracture between faculty and students. Finally, students assigned Combined Attributes a mean of 34 while instructors gave this measure a mean of 33. Combined Attributes represents the assessment of importance each respondent assigned all 45 attributes (or seven attribute families). Consequently, this variable provides an overall measure of how important academic writing is considered by individual respondents or groups. The findings depicted in Figure 7 thus demonstrate that students generally found the attributes more important to academic writing than their instructors did. The importance of inquiry has particular connotations given the imperiled position the culture of thought holds in present-day universities. A s explained in "The Myth of Good Academic Writing" (Chapter One), a prevailing tension at institutions of higher learning is the emphasis accorded thinking as opposed to superficial, easily quantifiable excellence. Students, for instance, gave Mechanics the highest mean score of importance among all attribute families. It is relatively simple to decide if a piece of  150 writing is grammatically excellent and to structure writing courses around mechanics. It is far more difficult to assess the quality of thought in a piece of writing and to design a writing course that promotes student thinking. Nevertheless, as these findings indicate, first-year writing must revolve around thought and its expression in written discourse to provide legitimate preparation for students entering academia. Another key study theme has been the globalized university as it promotes, at least on paper, the culture of inclusion rather than exclusion. The global university's allure stems from the dismantling of barriers, allowing students from all over the world, and from diverse ethnic, cultural or socio-economic backgrounds access to higher learning. U B C has more than 40,000 students from close to 140 countries. How such a diverse body of students assesses the same samples and qualities of writing is the focus of the next phase of investigation. "Rival Hypothesis" (Chapter 9) revealed fractures among instructors. Sociodemographic variables with strength of association included gender, country of origin, department, and first language. Using this lens, I investigated whether these variables had statistically significant influence on students' scoring of importance. Department or faculty had no statistically significant influence on students' assessment of any of the attribute families. However, variables such as gender, country of birth and first language resulted in statistically significant fractures: These fractures are illustrated in Figure 8.  151 Attribute Families of Writing (N=523) M=229 Men W=289 Women; C=277 Canadian Born NC=238 Non Canadian Bom; E=274 English First Language NE=243 English Not First Language  Figure 8: Fractures in Students' Assessments of Attribute Family Importance by Gender, Country of Origin and First Language  152 The fractures depicted in Figure 8 run along categories of gender, country of birth and first language. •  G e n d e r had statistically significant influence on students' scoring of the importance of Mechanics (F=23.28; p<.001), Paragraph Structure (F=19.53; p<.001), Academic Conventions (F=16.44; p<.001) and Overall Importance (F=14.41; p<.001). Male students assigned Mechanics a mean of 37 in contrast to women who gave it a mean of 39. Female students assigned Paragraph Structure a score of 42 while men gave it a lower score of 39. Academic Conventions received a mean of 33 from women while male students assigned it a mean of 30. Finally, men assigned Overall Importance a mean of 33 while women assigned it a mean of 34.  •  Country of birth had a statistically significant influence on students' assessments of the importance of Mechanics (F=9.74; p<.001), Author's Voice (F=15.89; p<.001) and Social Analysis (F=18.66; p<.001). Canadian-born students gave Mechanics a mean score of 39, in contrast to 37 from nonCanadian-born students. Author's Voice received a mean of 32 from students born outside of Canada and a mean of 30 from the Canadian born. Social Analysis was preferred by non-Canadian-born students (mean=31) than by students born in Canada (mean=29).  •  S t u d e n t s ' first language resulted in statistically significant fractures in three families: importance of Author's Voice (F=10.79; p<.001), Social Analysis (F=7.70; p<.01) and Academic Inquiry (F=6.90; p<.01). Author's Voice was preferred by students whose first language was not English (mean=32) as opposed to those whose first language was English (mean=30). Students whose first language was English assigned Social Analysis a mean of 29 while students whose first language was not English gave it a mean of 31. Finally, students whose first language was English preferred Academic Inquiry (mean=38), in contrast to students whose first language was not English (mean=37). Native English speakers thus had a slightly better appreciation of the role of inquiry in academic writing than did their E S L counterparts.  U B C , like so many other institutions of higher learning in North America, has a multicultural, diverse student body. However, as these findings reveal, students do not consistently share perspectives on good writing. Thus, while some students might  153 already have insight into what their instructors expect, others may not, particularly if they were born outside of Canada or speak English as a second or third language (see Andrade, 2006 and Hutchings, 2006 for more on difficulties international students face when learning writing). For instance, these students were more likely to score Author's Voice highly than their Canadian-born, native-speaking counterparts did, suggesting personal-themed, subjective writing might be a way to introduce students on the linguistic or cultural margins to academic discourse. The crux of the issue is this: while it helps the reputation of a university to appear inclusive, such openness and tolerance are ultimately pointless if foreign-born or E S L students are not provided necessary support to become viable, accomplished producers of academic discourse. Whether or not they receive such support at U B C , and the efficacy of this assistance, are subjects for further and necessary research. In short, fractures are a viable and tangible presence in assessments of good writing. They exist because respondents have differing perspectives on what good writing must look like. I document these perspectives below.  Perspectives on Good Student Writing I began this study with the hypothesis that there are at least three distinct and viable perspectives on good student writing: Current Traditional Rhetoric, Expressivism and Social Construction. Using cluster analysis, the ranks, grades and scores assigned to the writing variables were tested to determine how paragraphs and attributes were grouped together in the minds of the 157 instructors.  Findings revealed at least seven  perspectives on good student writing. These perspectives are represented in Figure 9 and demonstrate how instructors grouped and clustered ranks and grades assigned the three paragraphs together with the 45 writing attributes based on their perceived importance. I used a dendogram to illustrate instructors' perspectives on good student writing. The figure was generated by S P S S , hence the truncated names assigned to each attribute. On the left of each attribute are the letters C, S, and E, representing the Current Traditional Rhetoric, Social Constructivist, and Expressivist pedagogies from which the phrases originated.  154  Figurative  EIMAGERY  12  .EMETAPHR  1 3  EAUPVIEW  8  EPPRNOUN  11  EAUVOICE  5  EINCLSNA  10  EAUINVLV  7  EHNSTYAV  9  SATTNRDR  24  RANK3-EXP  50  GRADE3  51  EAUTBACK  2  EAUTEXPR  3  EPTRUTH  6  SSOCREF  IS  Language=2  J Author's  Voice—7  Author's  8ackground=7  4  EAUTFEEL Social  25  20  15  C A S E Label Num Expressivism  Constr  SCONTXT  14  SCUREVNT  16  SWRLDEVN  17  SPWRSTR  20  SSOCIMP  21  SSOCINFL  18  RANK2-SCON  48  GRADE2  49  SCONVENT  26  SRHESTRG  27  55 A C AD I HQ  22  SCRITHNK  23  SRELSCHO  19  SSTANDRD Current  Social  1  EBIAS  Academic  A n a j.vs:i.s~  Inquiry=7  J  2b Traditional  CSYNTAX  28  CGRAM  31  CARTICLE  34  CPREPUSE  35  CPUNC  32  CSPELL  33  CVOCAB  41  CVARSNT  45  CTRANSTN  36  CLINKS  37  CWORDCHO  39  RANK1-CTR  46  G R A D E 1.  47  CCLARITr  29  CCOHEREN  30  CPLACEMT  38  CORGANI  40  CTHESIS  42  CTOPIC  43  CUNIPRGH  44  Mechanics=  11  J  J Paragraph  Structure=9  J  Figure 9: Fractures in Instructors' Assessments of the Paragraphs and Attribute Family Importance  155 The figure invites two observations. First, there were distinct groupings in instructors' clustering of attribute importances and paragraph ranking and grading.  For  clarity, each cluster is presented in a different colour. Second, these clusters indicate seven natural groupings or major clusters, mirroring the seven attribute families. I assigned each of the three pedagogies at this study's foundation a different colour: red-orange for Expressivism, green for Social Construction and blue for Current Traditional. The first cluster (bright red) cluster to form was Figurative Language and consists of Use of Imagery and Use of Metaphor. This cluster is Expressivist in nature. The second (dark blue) cluster to form was Mechanics and began with the aggregation of Correct Use of Articles and Correct Use of Preposition. This family mirrors the Mechanics attributes family almost completely and consists of these eleven attributes: Correct Use of Prepositions, Command of Punctuation Rules, Correct Use of Articles, Command of Spelling Rules, Adherence to Syntax Rules, Command of Grammatical Rules, Effective Use of Transition Words, Precision of Word Choice, Sophistication of Vocabulary, Variety of Sentence Structures and Links between Paragraphs. The third (dark green) cluster to form was Academic Inquiry and began with the clustering of Use of Academic Rhetorical Strategies and Use of Academic Discourse Conventions. This cluster consists of Social Construction rank and grade, as well as the aforementioned Use of Academic Rhetorical Strategies and Use of Academic Discourse Conventions - two attributes within the Academic Conventions family. The cluster also comprises four of the five attributes within the Academic Inquiry family: Evidence of Academic Inquiry, Familiarity with Related Scholarship, Evidence of Critical Thinking and Faithfulness to Academic Standards. Thus, in the minds of instructors the Academic Conventions and Academic Inquiry attribute families, together with ranking and grading of the Social Construction paragraph, occupied approximate levels of importance. The fourth (lime green) cluster to form was Social Analysis and began with the aggregation of Critical Analysis of Current Events and Discussion of World Events. This family also includes Acknowledgement of Author's Biases, Acknowledgement of Social Context, Investigation of Power Structures, Investigation of Society's Impact on  156 Knowledge, and Examination of Society's Influence on Language - a total of seven attributes. This cluster mirrors the Social Analysis attribute family almost exactly. The fifth (dark orange) cluster to form was Author's Background and began with the aggregation of Inclusion of Author's Background and Inclusion of Author's Experiences. This cluster is also made up of the Expressivist paragraph rank and grade as well as Inclusion of Author's Personal Truths, Call for Social Reform, and Inclusion of Author's and Feelings. These attributes all belong to the Author's Voice attribute family, with the exception of Call for Social Reform. The clustering of these items suggest a conceptual overlay in the minds of instructors between author-centred Expressivism and need for social change. The sixth (plum-red) cluster to form was Author's Voice and began with the aggregation of Inclusion of Author's Voice (Style) and Inclusion of Author's Voice (Content). In addition, the cluster consists of Author's Point of View, Incorporation of Personal Pronouns, Author's Involvement with the Subject, Honesty of Author's Voice and Attention to Reader. All seven items correspond with the Author's Voice attribute family. Finally, the seventh (bright blue) cluster to form was Paragraph Structure and began with the aggregation of Structure of Thesis Statement and Structure of Topic Sentence. In addition, the cluster consists of the Current Traditional paragraph rank and grade, as well as Clarity of Sentences and Paragraphs, Coherence of Essay, Placement of Main Ideas, Proof of Organization, and Unity of Paragraphs. The attributes in this cluster echo that of the Paragraph Structure attribute family. The inclusion of the Current Traditional Rhetoric rank and grade not only hint at the conceptual overlay between the pedagogy and paragraph structure, but also suggest these items shared nearly equal importance in respondents' eyes. There are at least seven alternative and defensible ways to understand good writing as well as a conceptual link between the sample paragraphs preferences and scores and writing attributes. Each cluster has a distinct central thrust, representing either figurative language, authorial expression, author's background, grammar, structure, social inquiry or academic conventions. Chapter Five, "Responses to Attributes of Writing," demonstrated hierarchies of importance among the 45 writing attributes and, subsequently, the seven attribute families, suggesting not all clusters were equal in the minds of instructors. Nevertheless, clustering the families begs a vital question: can  157 different families co-exist or are they mutually exclusive? C a n an instructor encourage students to develop an authentic voice in their writing, for instance, with the caveat that voice be grammatically correct? Or, is such compatibility epistemologically impossible? These implications will be discussed in the following chapter. Figure 9 reveals other, broader variable groupings. More precisely, the seven clusters easily lend themselves to three broad groupings when the eye moves rightward. The first grouping comprises the first three major clusters, beginning with Use of Imagery and ending with Inclusion of Author's Feelings. This grouping is largely Expressivist, except for Attention to Reader. The second grouping is made up of the next two major clusters, starting with Acknowledgement of Author's Bias, and stopping at Faithfulness to Academic Standards. These items draw almost entirely from Social Construction, with the exception of Author's Bias. Finally, the third grouping consists of the last two major clusters, beginning with Adherence to Syntax, and ending with Unity of Paragraph. This grouping mirrors the Current Traditional pedagogy completely. The three broad groupings mirror the three pedagogies (Current Traditional, Expressivist, Social Construction). A s such, they provide pictorial and statistical verification for diverging opinions among composition researchers. There is a reason why the writing of an Expressivist scholar like Peter Elbow is so different in terms of tone, language and subject from that of a Social Construction researcher like Patricia Bizzell or a writing handbook inspired by Current Traditional Rhetoric. Simply put, each pedagogy advocates particular attributes in order to accomplish a list of specific goals. These distinctions lend themselves to detailed and explicit ways of thinking about and assessing writing.  Conclusions Individuals in a university as multiculturally, ethnically and socio-demographically diverse as U B C are bound to have differing opinions about what constitutes good academic writing. Participation in the university does not mean individuals cast aside socio-demographic characteristics like gender, first language, or country of birth. Instead, these characteristics influence how faculty and students assess student writing. I discuss the implications of fractures, the seven perspectives on good writing, as well as recommendations for writing instructors, program administrators and composition scholars in the final chapter of this study.  158 CHAPTER ELEVEN: THE REVOLUTION BEGINS WITH WRITING CONCLUSION, DISCUSSION AND RECOMMENDATIONS  I began this study with two broad goals: to identify attributes which instructors and students associated with good writing; to ascertain if there was consensus between students and instructors, and among instructors, about what constituted good writing. An additional aim was to determine the socio-demographic characteristics associated with endorsements of good writing. I have discussed these investigations at length in previous chapters. Along the way, this also became a study of how writing is taught, revealing major weaknesses in composition instruction in North American universities in general and at U B C in particular. For too long, composition research and writing instruction have occupied the margins of academic policy and practice. Writing instruction has been viewed as sufferance, a service the university unwillingly provides students with poor linguistics skills who should have been excluded from academia in the first place. Students have been told that writing is a basic skill, that writing proficiency is a simple matter of understanding syntax and grammar and that their failure to produce good academic writing is due to poor learning skills or apathy. For too long, therefore, composition instructors have been content to participate in the marginalization of their field and, in so doing, have betrayed their students, colleagues and intellectual mandate. What is needed is revolution, one that begins in writing classrooms and affects academic programs, departmental heads and university-wide policy. Specifically, I call for two fundamental changes: first, relocating composition studies from the obscure margins of academic to the intersection between pedagogical policy and practice; second, transforming the aims, direction and content of writing instruction for beginning students. I provide justification for this revolution by reviewing the major findings of this study, considering their implications for writing instruction and composition research, and providing detailed recommendations for instructors, administrators and policy makers to have a truly dramatic and necessary impact.  The Globalized University and Fractures of Good Writing At present, universities in North America are under unprecedented pressure to become globalized. The research they produce must have international reach and  159 impact; they must implement cross-national academic, student-exchange and distance learning programs; and they must attract a broad spectrum of culturally, geographically and ethnically diverse international students. Over the past three decades, the number of students leaving home each year to study abroad has grown at an annual rate of 3.9 percent, from 800,000 in 1975 to 2.5 million in 2004 (Newsweek, 2006). While most of this academic migration is from one developed country to another, the number of students from developing countries is also increasing. Canada alone receives more than 130,000 international students every year and foreign-student enrolment increased by more than 15 percent across the country in 2003, with many provinces showing jumps of 20 percent or greater (Association of Universities and Colleges of Canada, 2006). U B C has more than 5,420 international students and exchange students from over 138 countries (University of British Columbia, International Students, 2006) and international students make up almost 12% of the undergraduate student body (University of British Columbia, Welcome, 2006). Added to the mix are students who are recent immigrants. Of the 515 students in this study who disclosed their country of birth, 238 or 4 6 % had been born outside Canada. First language adds another layer of complexity: of the 517 students who provided information about their first language, 243 or 4 7 % listed a language other than English as their mother tongue. These students are expected to demonstrate university-level proficiency in the English Language. However, budget cutbacks, funding freezes and general apathy for the plight of international or E S L students mean little linguistic or writing supported is provided these students at major North American universities. Students at U B C with a less-than-stellar grasp of the English language, for instance, have limited options if they want to get through first-year English. They can enroll in (and pay for) non-credit courses at the Writing Centre, or hire private tutors. The high student/instructor ratio results in faculty being unable to provide the hands-on assistance students desperately need. The first-year students I spoke to alternated between frustration and despair: they were painfully aware of their shortcomings as writers but had no inkling of how to overcome these deficits. Many saw the linguistic and cultural differences that surrounded them as impassable ravines: they knew instructors found their writing poor but were given only vague direction on how to improve. Worse, they knew getting affirmation from one instructor was no guarantee they would please any other instructor.  160 The hard work of figuring out and delivering what an instructor wanted had to be done anew with each class. What these students had stumbled upon was the "dirty little secret" of the globalized university. For all the rhetoric of inclusion and tolerance, the university is also a site of fractures, especially when it comes to assessments of writing. Fractures have real and tangible impact, but are rarely acknowledged by faculty or administrators. Instead, the onus is unfairly placed on students. They must make their way past craters of dissent with little support, direction or expectation of success. The situation is especially dire, however, because this study revealed at least three major tectonic plates upon which fractures occured: among faculty, between faculty and students, and between native and non-native speakers of English. As "Responses to Sample Student Writing," (Chapter Five) revealed, there were numerous seismic breaks among instructors themselves. Each of the three paragraphs was ranked top, middle or bottom by a sizable number of instructors (see "Results of Sample Student Writing," Chapter Five). There was no majority of opinion favouring any one paragraph. In keeping with the lack of consensus, grade ranges overlapped for all three rankings, signifying lack of agreement about what distinguishes good or weak writing. There was also inconsistency between the way instructors ranked/graded the paragraphs and scored the attribute families, resulting in few positive and statistically significant paragraph/attribute families correlations. One explanation could be the division between who faculty were personologically (qualities they were either born with or born into as well as their world views) as opposed to who they were situationally (the traits that defined or had led to their role at UBC). Instructors were generally informed by situational variables when they ranked and graded the paragraphs, and by personological variables when they scored the attributes. Part-time faculty tended to endorse the Expressivist paragraph more strongly than full-time colleagues who generally endorsed the Social Construction paragraph. Faculty were also divided by departmental affiliation and number of years they had taught at U B C . Personological variables had greater influence on instructors' scoring of the attribute families than did situational variables. There were statistically significant and positive correlations between the Interpretive world view and Author's Voice and Figurative Language, indicating that respondents who espoused this paradigm were more likely to  161 score these families highly. There were also statistically significant correlations between Functionalism and Paragraph Structure and Academic Inquiry. Thus, faculty who had a conciliatory, tolerant and creative world view were more likely to endorse writing that was subjective, author-centred and inventive. In contrast, instructors with a structured, logical and proof-seeking world view tended to prefer structural and inquirydriven writing. These findings provide credence to Berlin's contention of a link between rhetoric and reality. Perspectives on good academic writing are coloured by personal perceptions of the world. Writing assessments are idiosyncratic, a finding that punctures any lofty claim of standardized, objective marking practices, but makes students' progression through academia even more challenging. The fractures of writing uncovered also suggest convergent boundaries between instructors and students, suggesting plates moving towards and colliding against each other. Instructors generally did not share their students' endorsement of a challenging, socially critical, open-ended writing style.  "Responses to the Attributes of Writing"  (Chapter Six) revealed further divisions: faculty endorsed inquiry-driven writing that focused on the nature of scholarship and exhibited critical thinking and academic awareness.  Students, in contrast, were more likely to associate good writing with  mechanics and social analysis.  This finding serves as a useful reminder of the  crevasses students must negotiate before they can become fully fledged members of the academic discourse community.  If the purpose of first-year writing is, as Social  Constructionists argue, to prepare students for such membership, do existing writing programs and courses successfully deliver on these aims, or add to the problem by providing students with a cursory and ultimately bewildering view of academic writing? If students learn to write by participating in the negotiation of meaning as members of the discourse community (Bruffee, 1983), are university administrators,  program  planners and instructors ensuring students are indeed welcome negotiators? Are they provided with guidance and support to ask questions and collaborate with instructors on the politics of language? Or are they expected to fill knowledge gaps by themselves? A third set of tectonic plates involved native speakers of English from non-native speakers. The ensuing fractures of writing are worthy of note in light of the large number of students who identified themselves as non-native speakers of the language. Importance of Author's Voice received a higher mean score from students whose first language was not English than it did from native English speakers. It also received a  162 higher score from students born outside of Canada than from the Canadian born. Author's Voice was also scored more highly by faculty whose first language was not English than by those who spoke English as their first language. Thus faculty and students who identified themselves as non-native English speakers shared at least one commonality. Author's Voice offers E S L writers an entry into academic writing that they might not otherwise have. The freedom to discuss personal themes, issues and context that might not be apparent or of concern to the mainstream, Canadian-born-and-raised, English-speaking community provides these writers an opportunity for self-expression and might help facilitate their development into more competent and confident writers. The use of Expressivist-derived techniques, writing assignments and teaching styles in a writing course for E S L students at U B C is a useful topic for future research . Seismic breaks between instructors and students are to be expected and can be used to justify the existence of the university. Students, university administrators could explain, enter institutions of higher learning so they can be sufficiently educated. A s novice members of the academy, they bring erroneous or naive perspectives on discourse that must be challenged and amended. The onus is therefore on students to submit to their instructors' superior knowledge. What is less convenient to explain are the colliding plates among instructors themselves. If, as this study has demonstrated, instructors cannot agree on what is good academic writing, what are students to do? Must students take it upon themselves to quiz their instructors on their employment status, country of origin, world view, or whatever other collisions future studies reveal? Since most first-year students rarely get to meet their instructors, let alone inquire into their personal lives, such investigations are doomed. It would be unrealistic to expect fractures in perspectives on good writing to disappear from academia. Universities are not monolithic, homogeneous institutions; they are comprised of many activity systems (Cole and Engestrom, 1993) or communities of practice (Lave and Wenger, 1991), each with its own unique way of seeing or using the tools or writing or language. Furthermore, as universities become increasingly larger and more internationalized, fractures in perspectives on good student writing will only multiply until the foundation of the university is irrevocably weakened. Writing instructors and composition researchers are uniquely positioned to stand in the gap between beginning students and university-wide expectations. They can act as  163 interpreters between other faculty and students: first, determining specific departmental or program perspectives on good writing; second, identifying and making sense of faculty contradictions; and third, translating these expectations in language that students can understand and make practical use of. Composition researchers and writing instructors are among the very few who have a macro-level view of fractures in perspectives on good writing. Thus, rather than ignore or downplay fractures, they have an obligation to determine points where fractures occur, make them known to other faculty and students alike, and provide students with strategies for dealing with them. If student writing is ever to improve, composition researchers have to look closely at the obstacles that prevent students from becoming confident writers. In particular, they must march purposefully from the margins to the nexus of the academy to expose the divisions within. Anything less would be tantamount to forfeiting students' development as academic writers. I discuss the steps that composition instructors and researchers must take a little later in this chapter.  Alternate Perspectives on Good Academic Writing Fractures exist because there are differing perspectives on good academic writing. "Responses to Attributes of Student Writing" (Chapter Six) indicated seven attribute families: Mechanics, Author's Voice, Social Analysis, Paragraph Structure, Academic Inquiry, Figurative Language and Academic Conventions. Furthermore, "Perspectives on Good Student Writing" (Chapter Ten) presented a pictorial representation of seven alternate perspectives on good writing. These perspectives corresponded almost identically with the seven attribute families. They also function as a wake-up call to composition researchers and writing instructors: teaching students to write involves more than correcting syntax or spelling errors with a red pen. While mechanics and structure are important components of good writing, the defining feature of academic writing is inquiry, as attested by the instructors who scored this attribute family highest out of the seven in terms of importance. What sets academic writing from other genres is the writer's ability to weigh, test and dissect conjectures. Academic discourse exists to stipulate what is not known and fill those gaps in knowledge. Unfortunately, writing courses for beginning students frequently overlooks academic inquiry to focus on mechanics, structure or style. For instance, Writing 098, a popular course at U B C ' s Writing Centre that prepares students for first-  164 year English, revolves almost exclusively around parts of speech, grammar and structure (University of British Columbia, Writing Centre, 2006). "Strategies for University Writing," a mandatory first-year English course offered by U B C ' s Department of English shifts the focus from mechanics to rhetorical principles and strategies. While mechanics, structure and style are important, the removal of inquiry is problematic because it lies at the very heart of academia. The attribute family Academic Inquiry consisted of the following: Evidence of Academic Inquiry, Familiarity with Related Scholarship, Evidence of Critical Thinking, Coherence of Essay and Faithfulness to Academic Standards. Evidence of Critical Thinking was deemed critically important by 79 percent of faculty surveyed, the highest scoring of the 45 attributes and one of the few points of consensus in this study. Three of the remaining four items placed in the top ten of highest scoring attributes, emphasizing the importance of this family among instructors. The statistical grouping of these attributes signify that instructors expect good student writing to exhibit three subthemes: critical thinking, coherence, knowledge of their discipline. Critical thinking is therefore central to good writing, making its omission from writing handbooks and course curricula all the more troubling. Furthermore, if clear thinking is integral to clear writing (Zinsser, 1998), ignoring the impact of thought dooms students to producing incoherent, rambling discourse that renders whatever knowledge they might have of their field moot. Writing courses must be radically redesigned to focus on critical thinking. Students must learn, for instance, how to frame an argument, test hypotheses, synthesize and integrate information, assess the authenticity and accuracy of knowledge claims, differentiate between evidence and opinion, identify logical relationships between ideas, generate creative and effective solutions to problems, and, most importantly, evaluate the quality of their own thinking. In-depth training on how to think is the bedrock for good writing. Mechanics, structure and rhetorical strategies, in comparison, are simply window dressing. Nevertheless, thought must operate within a context, raising the question: what are beginning students to think and write about? One option would be to draw upon two other attribute families generated in this study: Author's Voice and Social Analysis. Author's Voice comprises such author-centred attributes as Inclusion of Author's Voice, Inclusion of Author's Experiences, Inclusion of Author's Thoughts and Feelings,  165 Inclusion of Author's Background, Incorporation of Personal Pronouns, and Development of Author's Point-of-View. This attribute family offers students the chance to incorporate and evaluate their opinions and histories as part of their writing. This is especially useful since even though beginning students may not have a broad understanding of an academic discipline, they have expertise that nobody else does about who they are and what they have experienced. The old adage of write what you know has particular resonance here, particularly in light of this study's earlier findings of the strong impact that personological factors have upon perspectives on good writing. Students can develop into confident writers if they begin by writing about themselves. My own experience as a writing instructor is that students most enjoy assignments when they are invited to incorporate who they are into their writing. There is less anxiety about finding material to write about, and the writing that emerges is usually moving, thought provoking and conspicuously devoid of gnarly syntax and muddled structure. When students have thought deeply about the subject and are confident about what they are writing about, the writing usually takes care of itself. Author's Voice thus offers students (particularly students who are foreign-born or from non-English-speaking cultures, as this study has previously referenced) a launching pad from which to begin writing. However, as this study also demonstrated, because there is considerable overlap between the personal and the social context, themes that emerge from personalized writing can be expanded to include the broader milieu. Personal experience can be the lens through which cultural rites, mores and beliefs are first identified. By determining the place they occupy within the larger environment, students can begin to ascertain and critically evaluate the historical, epistemological, cultural, economic and religious forces that have shaped who they are as well as the framework within which they exist. The seven attribute families reveal that the focus of academic writing instruction must be broadened considerably. Writing instruction must provide students the tools to think deeply, and become more critically aware of the environment within which they operate as well as the very nature of knowledge. The integration of writing, critical thinking and learning is the focus of Writing Across the Curriculum (WAC), a movement that seeks to connect writing to learning in all subject areas, not just composition or writing programs housed in English departments (see Bazerman, Little, Bethel, Chavkin, Fouguette, Garufis, 2005; Fulwiler & Young, 1986). There are numerous W A C  166 departments and programs in universities across North America, and Simon Fraser University in Burnaby, B.C. - U B C ' s closest university neighbour - hosts a WritingIntensive Learning Office that collaborates with faculty and departments across campus to assess the implementation of new or modified approaches to the uses and teaching of writing. Despite their good intentions and commitment to facilitating student learning and writing proficiency, writing programs are generally underfunded, understaffed and overlooked by the rest of the academy (see Bridwell-Bowles, 2003; White, 1991 for more extensive discussion on challenges W A C programs face). This persistent marginalization can only be overcome if writing directors coordinate the research, administrative and pedagogical aspects of their programs (Jones & Comprone, 1993). In order for the teaching and researching of academic writing to have lasting and widespread impact, writing program directors will have to coordinate a lengthy list of responsibilities, including but not limited to the following: training writing instructors; developing course curricula; collaborating with departments across the university; securing funding for daily operations; assessing student learning and writing proficiency; maintaining or increasing student enrollment; facilitating research that pertains to writing; and disseminating research findings across the academy and composition research communities. Considering their limited financial and staffing resources, it would be impossible for writing program directors to successfully implement even a fraction of these ambitious goals. University presidents and senior administrators in general, and at U B C in particular, should be called upon to demonstrate their commitment to improving student writing in practical, sweeping and relevant ways. Instead of allowing the crucial work of writing instruction and research to languish on the margins of the academy, university administrators must facilitate the centralization of a Writing Department, ideally situated within the university's Academic Offices, to fulfill a three-pronged administrative, pedagogical and research approach. Specifically, this Writing Department should have the following mandate: to establish and supervise writing standards for students across the university; develop and implement writing programs across the university's academic disciplines; and facilitate the generation and dissemination of research in academic writing.  167 Recommendations for Administrators, Instructors and Researchers within the Writing Department Fractures in perspectives on good academic writing occur because, to paraphrase W.B. Yeats, there is no centre capable of holding the whole. The recommended Writing Department will function as that whole. While its presence will never completely eradicate gaps and crevasses, its aims will be to investigate the points at which fractures occur (research), navigate students through these fissures (pedagogy), and provide scaffolding within which writing instructors, researchers and program planners can operate most effectively. Specifically, the Writing Department will assume the following roles: •  Collaborate with heads of academic departments throughout the university to determine general as well as discipline-specific writing proficiency standards;  •  Administer entrance- and senior-level examinations that assess students' general and discipline-specific writing abilities at the start and end of their undergraduate tenures;  •  Work with academic departments through the university to determine the points at which fractures are most likely to occur between instructors and students, and among instructors;  •  Work in tandem with instructors and departmental heads across the disciplines to develop course curricula and standards of assessments that provide students with the skills to participate confidently in their discipline-specific discourse communities, as well as awareness of the fractures in perspectives on good writing that can occur within their disciplines;  •  Recruit tenure-track instructors and program planners trained in composition studies and rhetoric to work within the Department;  •  Facilitate university-wide symposiums, seminars and conferences at which findings generated by the Department can be disseminated;  •  Provide faculty within the Department resources and opportunities to attend and present their findings at international conferences;  •  Develop and implement graduate-level programs in composition studies and rhetoric at which graduate students will be able to get hands-on experience researching composition theory, practice and policy throughout the university, as well as at universities throughout the province or Canada.  168  Final Thoughts and Future Research Projects I began this study some years ago following the realization that there were deep and occasionally acrimonious chasms among instructors about what constituted good writing, and that these divisions often had painful and negative impact upon students' development as writers. Study findings have provided statistical evidence that such collisions exist across U B C ' s Faculty of Arts. Instructors and students disagree because there are definite fractures in perspectives on good student writing which are informed by who instructors and students are. The university is the site of infinite fractures of opinion and the challenge for any administrator is to coalesce them into a workable whole. Yet, for all the heartache, confusion and frustration they cause, fractures and the tectonic plates upon which they occur also make for rich and fascinating research subjects. Findings from this study invite at least three or four research projects. First, in order to make the attribute families a more finely-honed instrument of calibration, the number of definers in each family will have to be more or less consistent. At present, families generated in this study like Academic Context and Figurative Language comprise only two attributes each, in contrast to Mechanics and Author's Voice which consist often definers each. Future researchers will need to draw upon more definers for Academic Context and Figurative Language from the literature to impose numerical symmetry upon all attribute families. Second, the study can and should be expanded beyond U B C to determine just how widespread fractures in assessments of good student writing are. One potential and useful study would be to compare high-school teachers of English with first-year university instructors of English to contrast what university-bound high school students are taught about good writing, and what their future instructors expect. The study should also be expanded to universities and community colleges across B.C. to investigate potential fractures in English departments and writing centres. A third possible expansion would be to take the study of fractures in academic writing to the realm of scientific and technical writing. This study was restricted to instructors in the faculty of Arts. However, with some tweaking of the attribute families and sample writing, this study could be used to investigate the presence of the personal in scientific or technical writing. Both genres have traditionally relied almost exclusively  169 on objectivity, with little tolerance for subjectivity. However, subjectivity has made more of an inroad in recent years. Genetics, for instance, is a hard, biological science that requires detached reporting of data. However, in recent years genetic counselors have become a crucial part of the field and are increasingly conducting qualitative, interviewdriven research that requires respondents to comment on the effect certain genes have had upon their emotions, daily lives and future expectations - none of which can be measured with traditional scientific precision. Thus, the sciences contain a minefield of data about how an academic discipline juggles with two very different perspectives. Fourth, the relationship between world views and perspectives on good student writing should be tested further. A smaller scaled, ethnographic study could be conducted in a few writing classrooms over several weeks to determine how instructors differentiated by world views established objectives for their writing classrooms, as well as taught and assessed student writing. Specifically, this study should investigate the impact of world view upon how instructors teach writing. Finally, careful attention should be paid to the theoretical implications of the practical findings uncovered in this study. How do fractures in perspectives on good student writing enrich or complicate accepted or prevailing theoretical assumptions in composition theory? Having already identified the points at which fractures occur, future research should establish a conception basis from which to measure fractures, as well as analytical processes for explaining how and why fractures are shaped by situational and personological variables. Fractures can be used to focus many of the themes that comprise the modern university: multiculturalism, globalization, the nature of research, and the university's pedagogical objectives - to name a few. Additionally, fractures enrich the teaching and learning of writing. They remind instructors and students that academic discourse is complex and multifaceted and crucial to the very existence of the university. Without writing and the subsequent disseminating of knowledge, the academy is simply the site of pointless, obscure research. The objective, therefore, is not to ignore the existence of fractures but to embrace their possibilities for the teaching and researching of academic writing. I began this study by talking about a series of marking meetings that went awry. In the course of data gathering, I came across an instructor who had been at the same department at that time. He had been a teaching assistant then and had been part of  170 the organizing committee. When reminded of the marking meetings, he laughed and said, "The results were so embarrassing to the department that they locked away their findings in a drawer and never mentioned them again." Not mentioning conflicting perspectives on good writing ever again does not make these dissenting opinions disappear. Eventually they poison faculty and student relationships. Instead of avoiding fractures, instructors can and should use them as a basis for guiding students to become better writers. Without seismic breaks and their ensuing fractures, the academic landscape would be dull and flat, devoid of peaks and valleys that provide contour and character. Such uneven, unpredictable terrain can offer writers rich and inspiring source material. The objective, therefore, is to arm students with the tools and strategies they will need to traverse such terrain successfully.  171 REFERENCES Acker, S. & Armenti, C. (2004). Sleepless in academia. Gender and Education, 16, 324. Aguirre, A., Jr. (2000). Women and minority faculty in the academic workplace: Recruitment, retention, and academic culture. ( A S H E - E R I C Higher Education Rep. No. 27-6). San Francisco: Jossey-Bass Aisenberg, N., & Harrington, M. (1988). Women of academe: Outsiders in the sacred grove. Amherst, MA: University of Massachusetts Press. Altbach, P . G . (1999). Patterns in higher education development. In P.G. Altbach, R.O. Berdahl, & P.J. Gumport (Eds.), American higher education in the 21st century (pp.15-37). Baltimore, MD: John Hopkins University Press. Andrade, M.S. (2006). International students in English-speaking universities. Journal of Research in International Education, 5, 131-154. Angelova, M. & Riazantseva, A . (1999). If you don't tell me, how can I know? Written Communication, 16, 491 -525. Aronowitz, S. (2000). The knowledge factory: Dismantling the corporate university and creating true higher learning. Boston: Beacon Press. Association of Universities and Colleges and Canada. (2006). Information for International Students. Retrieved November 13, 2006, http://www.aucc.ca/can_uni/student_info/intl_student/intlstudents_e.html Baardman, S . P . (1994). Shifting perspectives: Developing social conceptualizations of writing. In S . P . Baardman, S . B . Straw, & L.E.Atkinson (Eds.), Social reflections on writing: To reach and realize (pp.1-11). Winnipeg, M B : Literacy Publications. Babin, E. & Harrison, K. (1999). Contemporary composition studies: A guide to theorists and terms. Westport, C T : Greenwood Press. Bartholomae, D. (1988). Inventing the university. In E.R.Kintgen, B.M. Kroll & M. Rose (Eds.), Perspectives on literacy (pp.273-285). Carbondale: Southern Illinois University Press. Bartholomae, D. (1995) Writing with teachers: A conversation with Peter Elbow. College Composition and Communication, 46, 62-71. Bawarshi, A . S . (1997). Beyond dichotomy: Toward a theory of divergence in composition studies. Journal of Advanced Composition, 17, 69-82. Bazerman, C , Little, J . , Bethel, L., Chavkin, T., Fouquette, D., & Garufis, J . (Eds.). (2005). Reference guide to writing across the curriculum. West Lafayette, Indiana: Parlor Press. Bercuson, D., Bothwell, R. & Granatstein, J.L. (1997). Petrified campus: The crisis in Canada's universities. Toronto: Random House. Berlin, J.A. (1984). Writing instruction in nineteenth-century American colleges. Carbondale & Edwardsville, IL: Southern Illinois University Press. Berlin, J.A. (1987). Rhetoric and reality: Writing instruction in American colleges. Carbondale and Edwardsville, IL: Southern Illinois University Press. Berlin, J . (1988). Rhetoric and ideology in the writing class. College English, 50, 477-494. Berlin, J . (1991). Composition and cultural studies. In C M . Hulbert & M. Blitz (Eds.), Composition and resistance (pp.47-57). Portsmouth, N H : Boynton/Cook Publishers.  172 Berlin, J . A. (1993). Composition studies and cultural studies: Collapsing boundaries. In A.R. Gere (Ed.), Into the Field: Sites of Composition Studies (pp.99-116). New York: MLA. Berthoff, A. (1972). From problem-solving to a theory of imagination. College English, 33,636-649. Berthoff, A. (1981). The making of meaning. Montclair, N J : Boynton/Cook. Bizzell, P. (1982) Cognition, convention, certainty: Knowledge about writing, Pre/Text, 3, 213-243. Bizzell, P. (1986). Foundationalism and anti-foundationalism in composition studies. Pre/Text, 7, 37-56. Bizzell, P. (1992). Academic discourse and critical consciousness. Pittsburgh: University of Pittsburgh Press. Bizzell, P. (1994). Are shared discourses desirable: A response to Nancy McKoski. Journal of Advanced Composition, 14, 271-277 Boshier, R.W. (1989). Epistemological and theoretical foundation of education about HIV/AIDS. Paper presented to the Annual Conference of the Comparative and International Education Society, Anaheim, California. Boshier, R.W. (1996). An expanded theoretical perspective on fishing vessel accidents. Paper presented at a National Defence/SAR Secretariat Conference on Search and Rescue, Dartmouth, Nova Scotia. Boyd, R. (1991). Imitate me; don't imitate me: Mimeticism in David Bartholomae's inventing the university. Journal of Advanced Composition^ 11, 335-346. Birdwell-Bowles, L. (2003). Literacy and Minnesota's academic culture: A time for institutional change. (Tech. Rep. No. 3). Minneapolis, M N : University of Minnesota, The Centre for Interdisciplinary Studies of Writing. Britton, James N. (1970). Language and learning. Coral Gables, FL: University of Miami Press. Britton, J . N., Burgess, T., Martin, N., McLeod, A., & Rosen, H. (1975). The development of writing abilities (11-18). London: MacMillan Educational for the Schools Council. Browne-Miller, A. (1996). Shameful admissions, the losing battle to serve everyone in our universities. San Francisco, C A : Jossey Bass, Inc. Bruffee, K.A. (1983). Writing and reading as collaborative or social acts. In J . N . Hays, R.D. Foulke, J.A. Ramsey, & P.A. Roth (Eds.), The writer's mind: Writing as a mode of thinking (pp. 159-169). Urbana: N C T E Bruffee, K. A. (1986). Social construction, language, and the authority of knowledge: A bibliographic essay. College English, 48. 773-790. Burnham, C. (1992). Expressive rhetoric: A source study. In T. Enos & S. Brown (Eds.), Perspectives on twentieth century rhetoric: Essays toward defining the new rhetorics (pp. 154-70). Los Angeles: Sage. Burnham, C. (2001). Expressive pedagogy: Practice/theory, theory/practice. In G . Tate, A. Rupiper & K. Schick (Eds.), A guide to composition pedagogies (pp. 1935). New York: Oxford University Press. Burrell, G . & Morgan, G . (1979). Sociological paradigms and organisational analysis. Brookfield, VT: Ashgate Publishing. Campbell, D.T. & Fiske, D.W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105. Campbell, Donald T. & J . C . Stanley (1963). Experimental and quasi- experimental designs for research. Chicago: Rand-McNally.  173 Canadian Association of University Teachers (2006). Almanac of post-secondary education 2006. Retrieved April 1, 2006, from http://www.caut.ca/en/ publications/almanac/default.asp Canarajah, S . (2001). Addressing issues of power and difference in E S L academic writing. In J . Flowerdew & M. Peacock (Eds.), Research perspectives on English for academic purposes (pp. 117-131). Cambridge: Cambridge University Press. Clark, R. & Ivanic, R. (1997). The politics of writing. London: Routledge. Coles, W.E., Jr. (1978). The plural I: The teaching of writing. New York: Holt, Rinehart & Winston. Cole, M., & Engestrom, Y . (1993). A cultural-historical approach to distributed cognition. In G . Salomon (Ed.), Distributed cognitions: Psychological and educational considerations (pp. 1-46). Cambridge, UK: Cambridge University Press. Coles, W.E., Jr. & Vopat, J . (1985). What makes writing good: A multiperspective. D.C. Heath: Lexington. Connors, R.J. (1997). Composition rhetoric: Backgrounds, theory and pedagogy. Pittsburgh: University of Pittsburgh Press. Connor, U. (1996). Contrastive rhetoric: Cross-cultural aspects of second-language writing. Cambridge: Cambridge University Press. Crowley, S. (1986). The current-traditional theory of style: an informal history. RSQ 16, 233-50. Crowley, S. (1990). Composition in the university: Historical and polemical essays. Pittsburgh: University of Pittsburgh Press. Crowley, S. (1998). The methodical memory: Invention in current-traditional rhetoric. Carbondale: Southern Illinois Press. Curtis, J.W. (2005). Trends in faculty status: 1975-2003. American Association of University Professors. Retrieved June 12, 2006, from http://www.aaup.org/ research/FacStatTrends.html Deetz, S. (1996). Describing differences in approaches to organization science: Rethinking Burrell and Morgan and their legacy. Organization Science, 7(2), 191-207. Diederich, P.B. (1974). Measuring growth in English. Urbana, IL: National Council of Teachers. Emig, J . (1971). The composing processes of twelfth graders. Urbana, IL: National Council of Teachers of English. Elbow, P. (1973). Writing without teachers. New York: Oxford University Press. Elbow, P. (1987). Closing my eyes as I speak: An argument for ignoring audience. College English, 49 (50-69). Elbow, P. Introduction. In P. Elbow (Ed.), Landmark essays on voice and writing (pp. xixvii). Davis, C A : Hermagoras. Elbow, P. (1995). Being a writer vs. being an academic: A conflict in goals. College Composition and Communication, 46, 72-83. Faigley, L. (1992). Fragments of rationality: Post modernity and the subject of composition. Pittsburgh: University of Pittsburgh Press. Fish, S. (1989). Doing what comes naturally: Change, rhetoric, and the practice of theory in literary and legal studies. Durham, N C : Duke University Press. Fishman, S . M . & McCarthy, L. (2002). Whose goals and whose aspirations: Learning to teach underprepared writers across the curriculum. Logan, UT: Utah State University Press.  174 Fitts, K. & France, A.W. (1995). Post/face. In K. Fitts & A . W . France (Eds.), Left margins: Cultural studies and composition pedagogy (pp.319-324). Albany, NY: State University of New York Press. Flower, L. (1994). The construction of negotiated meaning: A social cognitive theory of writing. Carbondale, IL: Southern Illinois Press. Fogarty, D. (1959). Roots for a new rhetoric. New York: Teachers College Press. Fulkerson, R. (1979). Four philosophies of composition. College Composition and Communication, 30, 343-348. Fulwiler, T. & Young, A . (1986). Writing across the disciplines: Research and practice. Upper Montclair, N J : Boynton/Cook. Gale, X . L. (1996). Teachers, discourses, and authority in the postmodern composition classroom. Albany: State University of New York Press. Geertz, C. (1983). Local knowledge: Further essays in interpretive anthropology. New York: Basic Books. Gillespie, P., Gillam, A., Brown, L.F., & Stay, B. (Eds.), (2002). Writing centre research: Extending the conversation. Mahwah, N J : Lawrence Erlbaum Associates. Giroux, H.A. (1994). Beyond the ivory tower: Public intellectuals and the crisis of higher education. In M. Berube, & C . Nelson (Eds.), Higher education under fire: Politics, economics and the crisis of the humanities (pp.238-258). Routledge: New York & London. Giroux, H.A. (2001). Critical education or training: Beyond the commodification of higher education. In H. A . Giroux & K. Myrsiades (Eds.), Beyond the corporate university: Culture and pedagogy in the new millennium (pp. 1-12). Lanham, M D : Rowman & Littlefield. Giroux, H.A. & Myrsiades, K. (2001). (Eds.). Beyond the corporate university: culture and pedagogy in the new millennium: Culture and pedagogy in the new millennium. Lanham, MD: Rowman & Littlefield. Goles, T. & Hirschheim, R. (2000). The paradigm is dead, the paradigm is dead...long live the paradigm: The legacy of Burrell and Morgan. Omega, 28, 249-268. Goodley, D. & Lawthom, R. (2005). Epistemological journeys in participatory action research: alliances between community psychology and disability studies. Disability and Society, 20, 135 - 1 5 1 . Gough, H. G . , & Heilbrun, A .B. (1983). The adjective check list manual. Palo Alto, C A : Consulting Psychologists Press Grabe, W., & Kaplan, R. B. (1989). Writing in a second language: Contrastive rhetoric. In D. M. Johnson & D. H. Roen (Eds.), Richness in writing: Empowering ESL students (pp. 263-283). New York, London: Longman Gradin, S . (1995). Romancing rhetorics: Social Expressivist perspectives on the teaching of writing. Portland, O R : Portland State University. Graff, G . (1992). Beyond the culture wars: How teaching the conflicts can revitalize American education. New York: W.W.Norton. Graves, D. (1982). Writing: Teachers and children at work. Exeter, N H : Heinemann Books. Graves, R. (1994). Writing instruction in Canadian universities. Winnipeg, M B : Inkshed Publications. Hairston, M. (1992). Diversity, ideology, and teaching writing. College Composition and Communication, 43, 179-95 Hamp-Lyons, L. & Zhang, B.W. (2001). World Englishes: Issues in and from academic  175 writing. In J . Flowerdew & M. Peacock (Eds.), Research perspectives on English for academic purposes (pp. 101-116). Cambridge: Cambridge University Press. Harvey, W . (2002). Minorities in higher education: Nineteenth annual status report. Washington, D C : American Council on Education. Herndl, C . (1993). Teaching discourse and reproducing culture: A critique of research and pedagogy in professional and non-academic writing. College Composition and Communication, 44, 349-363. Hollander, P. (2000). Long term cultural trends and the problems of higher education in the United States. In R. Weissberg (Ed.),.Democracy and the academy (pp.21 40). Huntington, NY: Nova Science Publishers, Inc. Hounsell, D. (1997). Contrasting perspectives on essay writing. In F.Marton, D. Hounsell, & N. Entwistle (Eds.). The experience of learning: Implications of teaching and learning in higher education (2nd ed., pp.106-126J. Edinburgh: Scottish Academic Press. Hubert, H.A. (1994). Harmonious perfection: The development of English studies in nineteenth-century Anglo-Canadian colleges. East Lansing, M l : Michigan States University Press. Hutchings, C . (2006). Reaching students: lessons from a writing centre. Higher Education Research and Development, 25, 247 - 261. Ivanic, R. (1998). The discoursal construction of identity in academic writing. Amsterdam: John Benjamins Publications. Ivanic, R, & Lea, M.R. (2006). New contexts, new challenges: The teaching of writing in UK higher education. In L. Ganobcsik-Williams (Ed), Teaching academic writing in UK higher education (pp.6-15). Houndmills, Basingstoke: Palgrove Macmillan. Johns, A . M . (1997). Text, role, and context: Developing academic literacies. Cambridge: Cambridge University Press. Johnson, N. (1988). Rhetoric and belles letters in the Canadian academy: A n historical analysis. College English, 50, 861-873. Jones R. & Comprone J . J . (1993). Where do we go next in writing across the curriculum ? College Composition and Communication, 44, 59-68. Kimball, B. (1986). Orators and philosophers: A history of the idea of a liberal education. New York: Teachers' College Press. Kinneavy, J . (1980). A theory of discourse: The aims of discourse ( 2 ed.). New York: nd  W.W. Norton. Kirsch, G . E. (1993). Women writing the academy: Audience, authority, and transformation. Carbon dale: Southern Illinois University Press. Knoblauch, C , & Brannon, L. (1984). Rhetorical traditions and the teaching of writing. Upper Montclair, N J : Boynton-Cook. Kubota, R. & Lehner, A . (2004). Toward critical contrastive rhetoric. Journal of Second Language Writing, 13, 7-27. L a u e r J . M . , & Asher, J.W. (1988). Composition research: Empirical designs. New York: Oxford University Press. Lave, J . & Wenger, E. (1991) Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Lea, M. (1994). I thought I could write until I came here. In G . Gibbs (Ed.), Improving student learning (pp.216-226). Great Britain: Oxonian Rewley Press Ltd.  176 Lea, M. (1999). Academic literacies and learning in higher education: Constructing knowledge through texts and experience. In C . Jones, J.Turner, &. B. Street. Student writing in the university: Cultural and epistemological issues (pp. 103-124). Amsterdam: John Benjamins Publications. Lea, M. R., & Street, B. V. (1998). Student writing in higher education: A n academic literacies approach. Studies in Higher Education, 23, 157 - 72. Lea. M.R. & Street, B.V. (2000). Student writing and staff feedback in higher education: A n academic literacies approach. In M. Lea & B Stierer (Eds.), Student writing in higher education (pp. 32-46). Buckingham: S R H E and Open University Press. LeFevre, K.B. (1987). Invention as a social act. Carbondale: Southern Illinois University Press. Leki, I. (1995). Good writing: I know it when I see it. In D. Belcher & G . Braine (Eds.), Academic writing in a second language: Essays on research and pedagogy (pp.23-46). Norwood, N J : Albex Publishing Corporation. Leki, I. (2003). Research insights on second language writing instruction. Center for Applied Linguistics. Retrieved September 12, 2005, from http://www.cal.org/resources/digest/0306leki.html Li, X - M . (1996). Good writing in cross-cultural context. Albany: State University of New York Press. Lillis, T. (1997). New voices in academia? The regulative nature of academic writing conventions. Language and Education, 7"/, 182 - 199. Lillis, T . M . (2006). Moving toward an academic literacies pedagogy: Dialogues of participation. In L. Ganobcsik-Williams (Ed), Teaching academic writing in UK higher education (pp.30-48). Houndmills, Basingstoke: Palgrove Macmillan. Little, B. (2001). Personality psychology: Havings, doings and beings in context. In J . S . Halonen & S.F. Davis (Eds.), The many faces of psychological research in the 21 century (chap.2). Retrieved August 23, 2006 from http://teachpsych.lemoyne.edu/teachpsych/faces/script/ch05.html Macrorie, K. (1970). Telling writing. Upper Montclair, N J : Boynton/Cook. Maimon, E., Nodine, B.F. & O'Conner, F.W. (1989). Thinking, reasoning and writing. London: Longmann Group United Kingdom. McAdams, D. P. (1995). What do we know when we know a person? Journal of Personality, 63, 365-396. McRae, R.R., & Costa, P.T. (1987). Validation of the five factor model of personality st  across instruments and observers. Journal of Personality and Social  Psychology,  52,81-90 Milam Jr., J.H. (1991). The presence of paradigms in the core higher education journal literature. Research in Higher Education, 32, 651 - 668. Miller, T . P . (1997). The formation of college English. Pittsburgh: University of Pittsburgh Press. Murray, D. (1968). A writer teaches writing. Boston: Houghton. Murray, D. (1984). Write to learn. New York: Holt, Rinehart and Winston. Mysyk, A. (2001). The sessional lecturer as migrant worker. The Canadian Journal of Higher Education, 31, 73-91. National Center for Educational Statistics. (2005). Digest for Education Statistics Tables and Figures. Retrieved October 12, 2006 from http://nces.ed.gov/programs/digest/d05/tables/dt05_227.asp. Newman, J.H. (1976). The idea of a university defined and illustrated. Oxford:  177 Clarendon. (Original work published 1852) Newsweek. (2006). The complete list: The top 100 global universities. Retrieved October 12, 2006, from http://www.msnbc.msn.com/id/14321230/ Nunnally, J . C , & Bernstein, I. H. (1994). Psychometric theory ( 3 ed.). New York: rd  McGraw-Hill. Paley, K.S. (2001). I-Writing: The politics and practice of teaching first-person writing. Carbondale: Southern Illinois University Press. Pelikan, J . (1992). The idea of the university: A reexamination. New Haven & London: Yale University Press. Perry, W. G . Jr. (1970). Intellectual and ethical development in the college years. New York: Holt, Rinehart & Winston. Petraglia, J . (1991). Interrupting the conversation: The constructionist dialogue in composition. Journal of Advanced Composition, 11, 37-55. Phelps, L. (1986). The domain of composition. Rhetoric Review, 4, 182-195. Puplampu, K. (2004). The restructuring of higher education and part-time instructors: a theoretical and political analysis of undergraduate teaching in Canada. Teaching in Higher Education, 9, 171-182. Rajagopal, I. (2002). Hidden academics: Contract faculty in Canadian universities. Toronto: University of Toronto Press. Readings, B. (1996). The university in ruins. Boston: Harvard University Press. Rudy, S.W. (1951). The 'revolution' in American higher education: 1865-1900. Harvard Educational Review, 21, 155-74. Ruszkiewicz, J . J . (1981). Well-bound words: A rhetoric. Glenview, IL: Scott, Foresman Schneider, M.L. & Fujishima, N.K. (1995). When practice doesn't make perfect: The case of a graduate E S L student. In D. Belcher & G . Braine (Eds.), Academic writing in a second language: Essays on research and pedagogy (pp. 3-22). Norword, N J : Ablex Publishing. Schultze, U. & Stabell, C . (2004). Knowing What You Don't Know? Discourses and Contradictions in Knowledge Management Research. Journal of Management Studies, 41(4), 549-573. Scott, M. (1996). Context as text: A course for student writers in higher education. In G.RijIaarSDam, H. van den Bergh, M. Couzijn (Eds.) Effective teaching and learning of writing. Current trends in research (pp.301-330). Amsterdam: Amsterdam University Press. Shaughnessy, M. (1977). Errors and expectations: A guide to the teacher of basic writing. New York: Oxford University Press. Shor, I. (1980). Critical teaching and everyday life. Cambridge, M A : South End Press. Shor, I. (1992). Empowering education: Critical teaching for social change. Chicago: University of Chicago Press. Suzuki, B. H. (1994). Higher education Issues in the Asian American community. In M. J . Justiz, R. Wilson, & L. G . Bjork (Eds.), Minorities in higher education (pp. 258285). Phoenix, A Z : Oryx Press. Snow, C P . (1959). The two cultures and the scientific revolution. Cambridge University Press. Spellmeyer, K. (1996). After theory: From textuality to attunement with the world. College English, 58, 893-913. Statistics Canada. (2004, July 30). The Daily. Retrieved February 17, 2007, from http://www.statcan.ca/Daily/English/040730/d040730b.htm Stewart, D. C . (1988). Collaborative Learning and Composition: Boon or  178 Bane? Rhetoric Review, 7, 58-83. Street B (1999). Academic literacies. In C. Jones, J.Turner, &. B. btreet ' (Eds ) Students writing in the university: Cultural and epistemological issues (oo 193-227) Amsterdam: John Benjamin Publications. Strun k W . & W h ite, E.B. The elements of style (4 ed.). Needham Heights, MA: Allyn & th  bacon. Taylor, G . (1988). The literacy of knowing: Content and form in students' English. In G . Taylor, B. Ballard, V. Beasley, H. Bock, J . Clanchy, & P. Nightingale (Eds.), Literacy by degrees (pp. 53-64). Milton Keynes: S H R E and Open University Press. Tickle, A. (1996). The writing process: A guidebook for ESL students.  New York:  Longman. Troyka, L. Q. (2002). Simon & Schuster handbook for writers (6 ed.). Englewood Cliffs, th  N J : Prentice Hall. University of British Columbia, Department of English. (2005). Strategies for university writing. Retrieved May 18, 2005, from http://www.english.ubc.ca/ courses/ summer2005/112.html University of British Columbia, International Students. (2006). International students: Welcome. Retrieved February 18, 2007 from http://www.international.ubc.ca/ University of British Columbia, Welcome. (2006). Welcome to UBC. Retrieved April 3, 2006, from http://www.welcome.ubc.ca/ University of British Columbia, Writing Centre (2006). Writing 098: Preparation for university writing and the LPI. Sample course outline. Retrieved November 16, 2006, from http://www.writingcentre.ubc.ca/pdf/w098.pdf Vardi, I. (2000) What lecturers want: A n investigation of lecturers' expectations in first year essay writing tasks. Proceedings of first year in higher education conference, Brisbane. Wardle, E.A. (2004). C a n cross-disciplinary links help us teach academic discourse in F Y C ? Across the disciplines: Interdisciplinary perspectives on language, learning and academic writing (1). Retrieved August 23, 2006, from: http://wac.colostate.edu/atd/articles/wardle2004/ White, W . (1991). Shallow roots or taproots for writing across the curriculum? ADE Bulletin (Spring), 32-34. Willmott, H. (1993). Breaking the paradigm mentality. Organization Studies 14, 681-730. Yancey, K. B. (Ed.). (1994). Voices on voice: Perspectives, definitions, inquiry. Urbana, IL: National Council of Teachers of English. Young, R. (1978). Paradigms and problems: Needed research in rhetorical invention. In C R . Cooper & L. Odell.(Eds.), Research on composing: Points of departure (pp.29-47). Urbana, IL: N C T E . Zinsser, W . (1998). On writing well. New York: Harpers.  APPENDICES Appendix A.  Email to Respondents  Appendix B.  Questionnaire (Instructor Version)  Appendix C.  Tables 54-56  Appendix D.  U B C Certificate of Ethics  180 APPENDIX A  Dear Professor : My name is Sunita Wiebe and I am currently researching first-year writing in English as part of my doctoral degree at UBC's Department of Educational Studies. Specifically, I am examining which attributes U B C instructors associate with "good" firstyear student academic writing. To that end, I have devised a four-section questionnaire that will take approximately 15 minutes to complete. The questionnaire has been approved by U B C ' s Ethics Committee. No names will be asked and any information respondents provide will be kept strictly confidential. I would be happy to leave a hard copy of the questionnaire in your mailbox, or send it to you electronically. I believe that the participation of instructors like yourself will provide invaluable insight into some of the difficulties and misconceptions students face when producing acceptable academic writing. Thus, I would be extremely grateful if you would consider participating. Yours sincerely, Sunita Wiebe Doctoral Candidate Department of Educational Studies University of British Columbia  181  APPENDIX B  Questionnaire (Instructor Version)  r  What is "Good Student Academic Writing"? :  Instructors and students sometimes disagree about what constitutes "good"first-yearwriting.  :.HT:  This four-section questionnaire examines reasons for these differences and should require about fifteen minutes of your time. T h a n k you!  Good Student Academic Writing  Fall: 2004  182  Section A - Sample Student Writing: Here are three opening paragraphs from a take-home essay for first-year English. Please read all three paragraphs first, then do two things (1) Rank the three paragraphs according to the writing style you prefer most (bottom, middle, or top rank). Circle your response. (2) Assign each paragraph a numerical grade for writing quality out of 10 points (half-points are allowed). Your writing style rankings and assessments of writing quality may differ, but please assess each paragraph twice and make certain that no two paragraphs receive the same ranking or same numerical grade.  The Importance of a University Education #1:  Each year, thousands of students enroll in institutions of higher learning across Canada. Their aim is to earn a university degree and they are prepared to spend a lot of money, time and effort to achieve this goal. Why are so many intelligent people prepared to sacrifice so much for a flimsy piece of paper? There are three reasons. First, universities prepare students for satisfying and lucrative careers. Second, they enable students to pursue a wide spectrum of academic disciplines. Finally, because they attract people from diverse backgrounds, universities offer students the chance to meet a unique range of intelligent and interesting individuals. Style Preference: Bottom /Middle / Top Rank  Grade:  out of 10 points  The Importance of a University Education #2:  The statistics are grim: rising tuition costs, a volatile job market and underemployed PhDs. In short, a university education no longer promises comfortable entry into professional, white-collar life. While this is not to suggest that the value of an education be measured by potential earnings, it does signal a conundrum. In an age of budget cutbacks and crushing student debt, university administrators and students must consistently consider their practices in terms of financial returns . How then to define such concepts as "importance" and "university education" subjectively at a time when figures, profit margins and bottom lines are more vital than ever before? Style Preference: Bottom /Middle / Top Rank  Grade:  out of 10 points  The Importance of a University Education #3:  When I was growing up in rural British Columbia, a university education seemed as remote as the moon, an extravagance my cash-poor family could ill afford. Yet, now that I am a first-year student, I have learned that a university education is not a luxury but, rather, a necessity. Being at UBC will not only help me escape the cycle of poverty that has plagued my family but will also broaden my mind and spirit immeasurably. My university education will teach me about the person I am, and make me aware of the creative and academic potential that lies within. Style Preference: Bottom /Middle / Top Rank Good Student Academic Writing  Grade:  out of 10 points Fall: 2004  183 Section B - Attributes of Writing: How important is each of the following attributes in a take-home student essay for first-year English? Please do not leave blanks. Simply circle your response.  How Important?  Content Acknowledgement of author's biases  Not at all  Slightly  Moderately  Highly  Critically  Acknowledgement of social context  Not at all  Slightly  Moderately  Highly  Critically  Call for social reform  Not at all  Slightly  Moderately  Highly  Critically  Critical analysis of current events  Not at all  Slightly  Moderately  Highly  Critically  Discussion of world events  Not at all  Slightly  Moderately  Highly  Critically  Examination of society's influence on language  Not at all  Slightly  Moderately  Highly  Critically  Familiarity with related scholarship  Not at all  Slightly  Moderately  Highly  Critically  Inclusion of author's background  Not at all  Slightly  Moderately  Highly  Critically  Inclusion of author's experiences  Not at all  Slightly  Moderately  Highly  Critically  Inclusion of author's thoughts and feelings  Not at all  Slightly  Moderately  Highly  Critically  Inclusion of author's voice  Not at all  Slightly  Moderately  Highly  Critically  Inclusion of personal truths  Not at all  Slightly  Moderately  Highly  Critically  Investigation of power structures  Not at all  Slightly  Moderately  Highly  Critically  Investigation of society's impact on knowledge  Not at all  Slightly  Moderately  Highly  Critically  Evidence of academic inquiry  Not at all  Slightly  Moderately  Highly  Critically  Evidence of critical thinking  Not at all  Slightly  Moderately  Highly  Critically  Adherence to syntax rules  Not at all  Slightly  Moderately  Highly  Critically  Clarity of sentences and paragraphs  Not at all  Slightly  Moderately  Highly  Critically  Coherence of essay  Not at all  Slightly  Moderately  Highly  Critically  Command of grammatical rules  Not at all  Slightly  Moderately  Highly  Critically  Mechanics  Good Student Academic Writing  Fall: 2004  184  How Important? Command of punctuation rules  Not at all  Slightly  Moderately  Highly  Critically  Command of spelling rules  Not at all  Slightly  Moderately  Highly  Critically  Correct use of articles  Not at all  Slightly  Moderately  Highly  Critically  Correct use of prepositions  Not at all  Slightly  Moderately  Highly  Critically  Effective use of transition words  Not at all  Slightly  Moderately  Highly  Critically  Links between paragraphs  Not at all  Slightly  Moderately  Highly  Critically  Placement of main ideas  Not at all  Slightly  Moderately  Highly  Critically  Precision of word choice  Not at all  Slightly  Moderately  Highly  Critically  Proof of organization  Not at all  Slightly  Moderately  Highly  Critically  Sophistication of vocabulary  Not at all  Slightly  Moderately  Highly  Critically  Structure of thesis statement  Not at all  Slightly  Moderately  Highly  Critically  Structure of topic sentences  Not at all  Slightly  Moderately  Highly  Critically  Unity of paragraphs  Not at all  Slightly  Moderately  Highly  Critically  Variety of sentence structures  Not at all  Slightly  Moderately  Highly  Critically  Attention to reader  Not at all  Slightly  Moderately  Highly  Critically  Author's involvement with the subject  Not at all  Slightly  Moderately  Highly  Critically  Development of author's point-of-view  Not at all  Slightly  Moderately  Highly  Critically  Faithfulness to academic standards  Not at all  Slightly  Moderately  Highly  Critically  Honesty of author's voice  Not at all  Slightly  Moderately  Highly  Critically  Inclusion of author's voice  Not at all  Slightly  Moderately  Highly  Critically  Incorporation of personal pronouns  Not at all  Slightly  Moderately  Highly  Critically  Use of academic discourse conventions  Not at all  Slightly  Moderately  Highly  Critically  Use of academic rhetorical strategies  Not at all  Slightly  Moderately  Highly  Critically  Use of imagery  Not at all  Slightly  Moderately  Highly  Critically  Use of metaphors  Not at all  Slightly  Moderately  Highly  Critically  Style  Good Student Academic Writing  Fall: 2004  185  Section C - Differing World Views: People think and act in legitimately different ways. Using the following alphabetical list of words, how would you describe the ways you generally think and act? Simply circle the word that best describes you. Your first impression is usually the most accurate. Work steadily. Do not leave any blanks. How Often? Activist  Never  Rarely  Sometimes  Often  Always  Advanced  Never  Rarely  Sometimes  Often  Always  Adversarial  Never  Rarely  Sometimes  Often  Always  Agreeable  Never  Rarely  Sometimes  Often  Always  Aggressive  Never  Rarely  Sometimes  Often  Always  Anarchist  Never  Rarely  Sometimes  Often  Always  Antagonistic  Never  Rarely  Sometimes  Often  Always  Appeasing  Never  Rarely  Sometimes  Often  Always  Artistic  Never  Rarely  Sometimes  Often  Always  Autonomous  Never  Rarely  Sometimes  Often  Always  Avant-garde  Never  Rarely  Sometimes  Often  Always  Aware  Never  Rarely  Sometimes  Often  Always  Balanced  Never  Rarely  Sometimes  Often  Always  Caring  Never  Rarely  Sometimes  Often  Always  Class conscious  Never  Rarely  Sometimes  Often  Always  Compassionate  Never  Rarely  Sometimes  Often  Always  Concerned  Never  Rarely  Sometimes  Often  Always  Conflict driven  Never  Rarely  Sometimes  Often  Always  Conformist  Never  Rarely  Sometimes  Often  Always  Confrontational  Never  Rarely  Sometimes  Often  Always  Considerate  Never  Rarely  Sometimes  Often  Always  Contentious  Never  Rarely  Sometimes  Often  Always  Controversial  Never  Rarely  Sometimes  Often  Always  Good Student Academic Writing  Fall: 2004  186  How Often? Creative  Never  Rarely  Sometimes  Often  Always  Critical thinking  Never  Rarely  Sometimes  Often  Always  Deferential  Never  Rarely  Sometimes  Often  Always  Disruptive  Never  Rarely  Sometimes  Often  Always  Efficient  Never  Rarely  Sometimes  Often  Always  Emancipated  Never  Rarely  Sometimes  Often  Always  Em pathetic  Never  Rarely  Sometimes  Often  Always  Empirical  Never  Rarely  Sometimes  Often  Always  Extreme  Never  Rarely  Sometimes  Often  Always  Factual  Never  Rarely  Sometimes  Often  Always  Forceful  Never  Rarely  Sometimes  Often  Always  Freedom seeking  Never  Rarely  Sometimes  Often  Always  Humanistic  Never  Rarely  Sometimes  Often  Always  Independent  Never  Rarely  Sometimes  Often  Always  Instinctive  Never  Rarely  Sometimes  Insurgent  Never  Rarely  Sometimes  Often  Always  Intellectual  Never  Rarely  Sometimes  Often  Always  Intense  Never  Rarely  Sometimes  Often  Always  Intuitive  Never  Rarely  Sometimes  Often  Always  Investigative  Never  Rarely  Sometimes  Often  Always  Law abiding  Never  Rarely  Sometimes  Often  Always  Liberated  Never  Rarely  Sometimes  Often  Always  Libertarian  Never  Rarely  Sometimes  Often  Always  Logical  Never  Rarely  Sometimes  Often  Always  Militant  Never  Rarely  Sometimes  Often  Always  Nurturing  Never  Rarely  Sometimes  Often  Always  Objective  Never  Rarely  Sometimes  Often  Always  Orderly  Never  Rarely  Sometimes  Often  Always  Organized  Never.  Rarely  Sometimes  Often  Always  Good Student Academic Writing  Often  Alway  Fall: 2004  187 How Often?  Perceptive  Never  Rarely  Sometimes  Often  Always  Political  Never  Rarely  Sometimes  Often  Always  Potential seeking  Never  Rarely  Sometimes  Often  Always  Powerful  Never  Rarely  Sometimes  Often  Always  Practical  Never  Rarely  Sometimes  Often  Always  Pragmatic  Never  Rarely  Sometimes  Often  Always  Problem solving  Never  Rarely  Sometimes  Often  Always  Progressive  Never  Rarely  Sometimes  Often  Always  Proof seeking  Never  Rarely  Sometimes  Often  Always  Questioning  Never  Rarely  Sometimes  Often  Always  Radical  Never  Rarely  Sometimes  Often  Always  Rational  Never  Rarely  Sometimes  Often  Always  Realistic  Never  Rarely  Sometimes  Often  Always  Reasonable  Never  Rarely  Sometimes  Often  Always  Rebellious  Never  Rarely  Sometimes  Often  Always  Reflective  Never  Rarely  Sometimes  Often  Always  Reformist  Never  Rarely  Sometimes  Often  Always  Self governing  Never  Rarely  Sometimes  Often  Always  Sensitive  Never  Rarely  Sometimes  Often  Always  Solicitous  Never  Rarely  Sometimes  Often  Always  Stable  Never  Rarely  Sometimes  Often  Always  Strong  Never  Rarely  Sometimes  Often  Always  Structured  Never  Rarely  Sometimes  Often  Always  Sympathetic  Never  Rarely  Sometimes  Often  Always  Thought provoking  Never  Rarely  Sometimes  Often  Always  Tolerant  Never  Rarely  Sometimes  Often  Always  Understanding  Never  Rarely  Sometimes  Often  Always  Violent  Never  Rarely  Sometimes  Often  Always  Please continue on to the last page... ^ Good Student Academic Writing  Fall: 2004  Section D- Background: Here are a few final questions about you, your academic training and your professional background.  What is your current position at UBC? •  Teaching assistant  • Sessional instructor  •  Tenure-track professor  • Tenured professor  • Other: How many years have you been teaching at U B C ? Part-time: Full-time: What is your academic specialty?  _years _years  Were you born in Canada? • Yes • N o If no, what is your country of origin? In what year did you arrive in Canada?  What is your first language? How long have you been writing academic papers in English? In what language did you earn your highest academic qualification?  years  What is your highest academic qualification? • Bachelor's degree  • Master's degree  • Doctoral degree  •  Post-doctorate  • Other: Where did you earn your highest academic qualification (please list both institution and country)? Institution Country In what year did you graduate with your highest academic qualification?  Have you ever received any formal training as a marker of first-year English writing? •Yes • No If yes, when and in which program did you receive such training?  Are you a man or a woman?  • Man  In what year were y o u born?  19  / (Year) (Program name)  • Woman  Thank you for your time, effort, and thought in completing this questionnaire! Sunita Wiebe, Doctoral Candidate, Faculty o f Education, U B C .  Good Student Academic Writing  Fall: 2004  189  APPENDIX C Tables 154-156  Table 154: Grades Instructors Assigned the Current Traditional Rhetoric Paragraph when the Current Traditional, Expressivist and Social Construction Paragraphs were Ranked Top, Middle or Bottom. (Number of Respondents in Parentheses.) Paragraph Current Traditional Expressivist Social Construction  Top Rank 8.53 (73) 7.39 (23) 7.34 (61)  Middle Rank 7.42 (53) 7.91 (51) 8.32 (51)  Bottom Rank 7.05 (29) 7.98 (81) 8.08 (43)  Table 155: Grades Instructors Assigned the Expressivist Paragraph when the Current Traditional, Expressivist and Social Construction Paragraphs were Ranked Top, Middle or Bottom. (Number of Respondents in Parentheses) Paragraph Current Traditional Expressivist Social Construction  Top Rank 7.02 (73) 8.07 (23) 6.89 (61)  Middle Rank 6.80 (53) 7.54 (51) 6.94 (51)  Bottom Rank 7.76 (29) 6.52 (81) 7.53 (43)  Table 156: Grades Instructors Assigned to the Social Construction Paragraph when the Current Traditional, Expressivist and Social Construction Paragraphs were Ranked Top, Middle or Bottom.(Number of Respondents in Parentheses) Paragraph Current Traditional Expressivist Social Construction  Top Rank 7.32 (73) 7.13 (23) 8.59 (61)  Middle Rank 8.01 (53) 7.46 (51) 7.80 (51)  Bottom Rank 8.38 (29) 8.12 (81) 6.51 (43)  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0055903/manifest

Comment

Related Items