UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An analysis of academic achievement among grades 7 to 11 students in a Northwestern Ontario band-controlled… Senior, Sharon 1993

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

Item Metadata


831-ubc_1994-0030.pdf [ 1.24MB ]
JSON: 831-1.0054609.json
JSON-LD: 831-1.0054609-ld.json
RDF/XML (Pretty): 831-1.0054609-rdf.xml
RDF/JSON: 831-1.0054609-rdf.json
Turtle: 831-1.0054609-turtle.txt
N-Triples: 831-1.0054609-rdf-ntriples.txt
Original Record: 831-1.0054609-source.json
Full Text

Full Text

AN ANALYSIS OF ACADEMIC ACHIEVEMENT AMONGGRADES 7 TO 11 STUDENTS IN A NORTHWESTERNONTARIO BAND-CONTROLLED SCHOOLBySHARON SENIORB.Sc., Andrews University, 1981B.Ed., Queen’s University, 1985A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF ARTSinTHE FACULTY OF GRADUATE STUDIES(Department of Educational Psychology and Special Education)We accept this thesis as conformingto j.he req-t1red standardTHE UNIVERSITY OF BRITISH COLUMBIANovember 1993(c) Sharon Senior, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. it is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(SigDepartment of /&N4 YH&LCGY SPEC/AL 1TiA/The University of British ColumbiaVancouver, CanadaDate /“Iv. c,DE-6 (2/88)ABSTR1CTThe main purposes of this study were to determine ifthere were (a) correlational relationships between academicself—concept (ASC) , general self—concept (GSC), attendance,(ATT), socioeconomic status (SES) , and academic achievement(AA) among grades 7 to 11 students in a band—controlledschool in Northwestern Ontario; and, (b) if academic self—concept, global self—concept, attendance, and socioeconomicstatus were predictors of academic achievement.The population sample was 70 Native students; 20 ingrade 7, 27 in grade 8, 13 in grade 9, 4 in grade 10, and 6in grade 11.The results showed: (1) significant correlationalrelationships between (a) ASC and GSC and SC and ATT forthe grades 7—9 students; (b) ?SC and TT for the grades 1011 students; and, (2) academic achievement was not predictedby any of the variables.Future research concerning Native education issuggested.11TABLE OF CONTENTSAbstract. . . . . . . . . . . . . . iiTable of Contents ...... . .. iiiList of Tables. . . . vAcknowledgement. . . . . . viChapter One The Pr oblem . . . . . . .1TheObjectivesoftheStudy 4SignificanceoftheStudy .....8Definitions of Terms 8Chapter Two Review of the Related Literature 10Factors Related to General and AcademicSelf—Concept. . . . . .10Factors Related to Attendance ..31Factors Related to Socioeconomic Status...38Statement of the Hypotheses....Chapter Three Methodology . . . 41Students. . . . . . . . . . .41School. . . . . . 42Instruments. . . . . . . . . . .42Experimental Design .48Procedure . . . . . . . .48Statistical Analyses. . . . . .49Chapter Four Results . . .51Academic Achievement Measure . .. ... .. .52Correlational Relationships among theVariables. . . . 52Predictors of Academic Achievement. 62Chapter Five Discussion and Conclusions...... . 63Discussion. . . . . . . . . . 63Academic Achievement Measure . . . . . . .63Correlational Relationships among theVariables..... ..... .64Predictors of Academic Achievement 68Recommemdations . . . . . . 68LimitationsoftheStudy ...71iiiConclusion. . . . . . . • • .71References...... . . . . . . . . . . . . . . . .74lppendix 1 . 82ivLIST OF TABLES1. Interrcorrelations of Academic Achievement (AA),Academic Self—Concept (ASC), General Self—Concept(GSC), Attendance (ATT), and Socioeconomic StatusSES) For Grades 7 to 9... .. ... 532. Interrcorrelations of Academic Achievement (AA),Academic Self—Concept (ASC), General Self—Concept(GSC), Attendance (ATT), and Socioeconomic Status(SES) For Grades 10 and 11.... 543. Michigan State General Self—Concept of Ability ScaleMean Scores by Grade •. . . . . . 554. Piers—Harris Children’s Self—Concept Scale Mean Scoresby Grade...... . . . . . . . . . . . . 565. Correlation between reading scores and both academicand general self—concept for grades 7—9 and grades10—11. . . . . . . . . . . . . . . . . . . . . . . . . 586. Correlation between reading and attendance for grades7—9 and grades 10—11 . . . . . . 597. Correlation between reading scores and SES for gradesgrades 7—9 and 10—il .. . .... 608. Correlation between SES and both academic and generalself—concept for grades 7—9 and grades 10—11 ...6lvACKNOWLEDGEMENTIt is my sincere pleasure to express appreciation toeveryone who contributed to the realization of this thesis.I am especially grateful to the chairman of mycommittee, Dr. Art More, for his contribution to thisresearch. His advice and support made this thesis possible.I also wish to thank my committee members, Dr. PerryLeslie and Dr. Dave Whittaker, for their roles in correctingthe drafts and for providing me with valuable suggestions.I am indebted to Dr..Walter Boldt for his expertise inthe statistical analyses.I also wish to thank Dr. Jackie Baker—Sennett for hervaluable feedback during the initial stages of the study.Special thanks goes to UBC graduate students, RayLandry and Marie Pighini, for taking time to discuss theresearch results with me.I wish to thank my parents, Robin and Gladys Senior, andmy sister and brother—in—law, Cheryl and Robert Wall, forthe encouragement which they provided throughout this study.Special thanks goes to the administration, staff, andstudents of the Indian reserve who granted me the time tocarry out the research in their school.I am also thankful for the providential guidance of Godthroughout the research and writing.viCHAPTER ONETHE PROBLEM-INTRODUCTIONA serious challenge facing Native peoples today is theeducation of future generations. Governments and religiousgroups, in applying a policy of assimilation, have producedunsuitable results in their attempts to fulfill this self—appointed responsibility (Dawson, 1988; Kirkness, 1981;Luftig, 1983; Pauls, 1984). Inspite of these unsuccessfulefforts, in 1969, the federal government proposed throughits White Paper to grant authority for Indian education tothe provincial governments (Pauls, 1984; Ward, 1986).The National Indian Brotherhood responded in 1972 withits paper entitled, Indian Control of Indian Education(National Indian Brotherhood, 1972). There are two facets tothe concept of Indian Control of Indian Education. Onefactor consists of parental involvement, the other entailslocal control (Richardson & Richardson, 1986). “Parentalresponsibility means Indian influence on the educationalprocess” (Kirkness, 1981, p. 452; see also Pauls, 1984).“Parents who are informed, interested, and willing tocontribute their time, add a broader dimension to education.However, parents must understand their function in order for1efficient educational services to be provided” (Richardson &Richardson, 1986, P. 21). Reyhner (1992) elaborates,“greater Indian parent involvement can reduce the culturaldistance between home and school....Parents need to haveeffective input as to how and what their children aretaught” (pp. 47, 51).The paper published by the National Indian Brotherhood,which was an outline of educational policy, goals, andphilosophy of education from the Native perspective, wasaccepted by the federal government in 1973. Ward (1986)observed:The summary of the Indian position on educationas stated in the (National Indian Brotherhood,1992) policy paper is: Indian parents must havefull responsibility and control of education. TheFederal Government must adjust its policy andpractices to make possible the full participationand partnership of Indian people in all decisionsand activities connected with the education ofIndian children. Pp. 10—11One of the most important results of the NationalIndian Brotherhood’s policy paper, was recognition of theresponsibility that Native parents have in promoting2cultural awareness and cultural development in the educationof their children (Huriburt, Henjurn, & Eide, 1983; Ward,1986). In their Manitoba study, Huriburt et al., (1983),compared 50 Indian students in a local high school that had400 White students and a teaching staff of 18 Whiteteachers, with 60 Indian students in a locally controlledschool where more than 40% of the teachers and 97% of the600 students were Indians. They reported “that students in alocally controlled school run by American Indians achievedhigher academic grades than American Indian studentsattending a White school....[and they concluded] thatAmerican Indian students would be better educated in theirown locally run schools” (p. 20).To counter the argument that perhaps there aredifferences in academic standards between the two groups ofstudents, therefore a difference in academic achievement,the researchers point out that “spelling, which has beenfound by [other] researchers to be positively related toacademic achievement and mental ability was also assessed,and no difference in spelling was noted between the twogroups of students. This lack of difference supportedcomparability of the two groups, and may suggest that thequality of instruction did not vary greatly between the two3schools” (p. 20)The move towards Indian Control of Indian Educationnecessitates an analysis of some of the components that mayaffect Native students present outlook and performance inthe educational system.The Objectives of the StudyNative students’ performances on standardized tests andtheir high drop—out rates suggests that the presenteducational system is not adequately serving their needs(Persaud & Madak, 1992; Riffel, 1991). It behooves educatorsto foresee the factors contributing to this pressing problemand counteract these influences before they becomeestablished.As is documented in chapter 2, there are four prominentvariables known to affect academic achievement. Thus, thepurposes of this study are to investigate(1) intercorrelational relationships between the fivevariables of academic achievement, general self—concept,academic self—concept, attendance, socioeconomic status, andacademic achievement; and, (2) to determine if theindependent variables of general self—concept, academic4self—concept, attendance, and socioeconomic status arepredictors of the dependent variable, academic achievement.The independent variables have been linked to causes ofsuccess and failure in school, primarily in the area of astudent’s perception of locus of control (Barnes & Vulcano,1982; Chapman, Cullen, Boersma, & Maguire, 1981; Madak,1988; Persaud & Madak, 1992; Riffel, 1991) . Thus, thiscombination of variables should provide additionalunderstanding in how a student’s interest and achievement inschool is influenced by the student’s interpretation ofhis/her control within that environment.s a foundation for supporting the choice of variablesinvolved in this study, the attribution paradigm wasselected because it addresses the perception of causality orthe criterion of why a particular event occurred“and...variations in the degree to which people believe theycan determine their own success” (Riffel, 1991, p. 26); inother words, attribution theory looks at the search forcauses of successes and failures (Hunter & Barker, 1987;Kruglanski, 1975). Possible perceived causes are “luck,ability, and personal effort” (Riffel, 1991, p. 26).The assignment of accountability affects futurebehaviour (Weiner, 1972). “Perceptions of causality, rather5than reality, are critical because they influence self—concept, expectations for future situations, feelings ofpotency, and subsequent motivation to put forth effort”(italics authors) (Hunter & Barker, 1987, p. 51).Hunter and Barker (1987) argue that effort is affectedby at least two factors: locus and controllability. Althoughnot the focus of this research, a word of explanation isprovided in order to differentiate between these terms.Locus can be internal or external and a student’s perceptionof the location of the cause of his/her failure of successaffects feelings of self—concept or as Martin and Coley(1984) state, “locus of control is a concept which refers tothe degree to which individuals perceive themselves ashaving control or influence over their environment(internal) or as lacking such control (external)” (p. 517).Controllability is related to a student’s “feeling ofpotency to affect the outcome by controlling the cause”(Hunter & Barker, 1987, p. 51; see also Kruglanski, 1975).Effort is the only causal attribution completely underthe control of an individual (Frieze, 1976). A student whobelieves his/her effort will influence the outcome will bemore likely to put forth effort (Hunter & Barker, 1987).Research has demonstrated that (1) high achievers exert6effort regardless of the domain; and, (2) that there aredevelopmental trends among students from grade 5 to collegein their concepts of ability versus effort in school work(Bloom, 1985; Gardner, 1983; Raviv, Bar—Tal, Raviv, & Levit,1983; Weiner & Peter, 1973). Younger students appreciateeffort more than ability, whereas the opposite evaluationhas been detected in older students. Weiner and Peter (1973)suggest that this may be the result of a change in values asa student matures. Riffel (1991) elaborates:Virtually all students begin school with highself—expectations for academic success....Overthe long term, self—concepts are strengthenedif students are able to attribute their successto their efforts and not to external forces(for example, luck or excellent teaching) overwhich they have no control....Success is notenough——how the student interprets that successis more important to self—concept. (p.p. 26—27)It is not suggested that the variables in this studyare the only factors affecting Native education.Nevertheless, these five variables have been shown to beessential elements in the educational process (Metcalfe,1981; Riffel, 1991; Wall & Madak, 1991).7The present study does not take published test normsinto consideration when looking at students’ tests scoresfrom any standardized test. n individual student iscompared only within the sample. The use of standardizedtest scores is problematic when these tests are used incultures for which they have not been designed. However,justification for continued use of these tests lie in thefact that a satisfactory alternative has yet to be found.Significance of the studyThe main significance of this exploratory study lies inthe combination of factors to be studied. Previous studieshave looked at each of the aforementioned variables, but nostudy presently exist that draws all five variables togetherand investigates them within the Native (or non—Native)student population. The analysis of these five variablesshould provide a more complete picture of their roles andhow their interactions cortribute to Native education.Definitions of TermsDefinitions of the following terms are presented for8clarification:(1) Native/Indian. Individuals who are Native people bybirth and heritage (i.e. Indian or descendants of Indians,whether or not they are classified as such under the termsof the Indian Act).(2) Standardized test(s). Refers to any published testthat is commonly used in educational settings to assessacademic achievement, mental abilities, etc., and has astandard administration and includes norms.(3) Academic self—concept. “Behaviour in whichindividuals evaluate (publicly or privately) their abilityto achieve in academic tasks as compared with others engagedin the same tasks” (Wall & Madak, 1991, p. 44).(4) Self—concept. “In general terms, SC is ourperception of ourselves; in specific terms, it is ourattitudes, feelings and knowledge about our abilities,skills, appearance, and social acceptability” (Byrne, 1984,p. 429).9CHAPTER TWOREVIEW OF THE RELATED LITERATUREThe literature review focuses on the academicachievement factors of (a) general and academic self—concepts; (b) attendance; and, (c) socioeconomic status. Itconcludes with a statement of the hypotheses.Factors Related to General and Academic Self—ConceptsSelf—concept, self—esteem, self—image, and self—worthhave all been used interchangeably in the literature. Aneffort will be made here to distinguish between these terms.It was the work of Wilbur Brookover that really set thestage for exploring self—concept. Canton (1981) observes:In the application of self—concept theories tolearning....[He] discovered that academic self—concept and grades remained solidly correlatedeven after the effects of intelligence had beenpartialled out....[however, although] as Brookoverdiscovered, virtually all high achievers hadrelatively high self—concepts...not all thosewith high self—concept had high performance levels.10(p.p. 78, 82—83)Some researchers see self—concept as a part of self—perception, with self—esteem and values being the other twocategories (Beane, Lipka, & Ludewig, 1980). Beane et al.(1980) define self—concept as how an individual woulddescribe himself/herself and self—esteem as “the level ofsatisfaction [one would] attach to that description....Self—esteem decisions, in turn, are made on the basis of what isimportant to us...our values” (p. 84). These researcherssuggests that adolescence is a period in which questions ofthe self become important as this age group “confronts theclassic identity crisis” (p. 84). They also point out thatschool achievement is one of several variables that has beenfound to be related to self—concept. Beane and Lipka (1980)further argue that “self—concept is typically defined as theperception one holds of oneself, totally and with regard toseveral dimensions, and which is influenced by environmentalinteraction” (p. 1). They see:Self—concept...[as] the descriptive perceptionof self in various roles and is judgementalonly in that one may assign some qualitativeassessment to the role performance....[and]self—esteern...as the valuative assessment one11makes regarding personal satisfaction withrole(s) and/or the quality of performance....In referring to self—concept and self—esteem,distinction must also be made in evaluativeterminoiogy....An individual’s actual orinferred self—concept is described as clear orconfused, complete or incomplete, general orspecific, or by other descriptive, but non—valuative technology. Self—esteem and inferredself—esteem are described as strong or weak,positive or negative, or by other value—relatedjudgemental terminology which suggests therelative comparative value or ranking theindividual attaches to the total self—conceptor its particular dimensions. (p.p. 2—3)Reid (1982) defines self—image as “how people seethemselves” and self—esteem as “what they put uponthemselves” (p. 179). His study investigated therelationship between persistent school absenteeism and self—concept. Self—concept was measured using “the Brookover(1967) Self—Concept of Academic Ability Scale” (p. 180).Results showed persistent school absentees as having lower12academic self—concepts than the control groups. They alsocame from lower socioeconomic backgrounds than the twocontrol groups.Martin and Coley (1984) report that “Coopersmith (1981)defined self—esteem as a set of attitudes an individualholds with regard to him— or herself” (p. 517). Hornett(1990) defines self—image or self—esteem “as an individual’sperception about self—competence, ability, goodness, anddesirability relative to other persons” (p. 43). Atherley(1990) sees self—esteem as:What a person feels about the discrepancybetween the way they are (the self—image) andthe way they would like to be (the ideal self).Thus a person whose actual and ideal self arevery distanced from each other and who thereforehas a negative perception of self is said to havelow self—esteem. Research supports a positiverelationship between the level of self—esteem andacademic achievement. (p. 225)Riffel (1991) suggests that self—concepts are “theideas that students develop about themselves——how theyunderstand themselves, what skills they think they possess,what interests they have, and how motivated they are to13develop their skills or pursue their interests” (p. 26)Further, this researcher points out that:Some empirical studies show [self—concept]...to be a relatively weak correlate of achievement(.20 or so), accounting for about five per centof the variance in student achievement....[also]academic self—concept is thought to operate asone of many variables which mediate the relationshipbetween teaching and learning. (p. 26)From her brief review, Parry (1982) states, “self—concept theorists all accept a view of the developing self—concept as being dependent on the individual’s perception ofhis total appraisal of how significant others view him”(p. 12) . Luftig (1983) stresses, “the self—concept has beendefined as the sum total of how an individual views himselfor herself” (p.p. 251—252). And according to Martin andColey (1984) “self—concept refers to the self—knowledge onepossesses regarding one’s strength and weaknesses. It isthat part of one’s personality of which one is aware”(p. 517) .Marshall (1989) sees self—concept as:The perceptions, feelings, and attitudes thata person has about himself or herself. The term14self—concept and self—image are often usedinterchangeably to designate a global conceptionof self. This global self—concept is made up ofmany dimensions.One dimension is self—esteem (or self—worth). Self—esteem refers specifically to our self—evaluations———that is, our judgments about our own worth———whereasself—concept refers to other aspects as well———physical characteristics, psychological traits,and gender and ethnic identity.As children develop, self—concept becomesincreasingly differentiated into multiple domains.(italics author’s) (p. 45)Pepper and Henry (1991) argue that:Self—esteem is about feeling good about oneself,feelings of personal worth, and feelings ofpersonal effectiveness in how a youngster valuesand regards his or her performance. Because self—esteem is a feeling, it always expresses itselfin the way a youngster acts....Self—esteem ishard to identify because it is experiencedcontinuously and constantly and changes from dayto day, from situation to situation, even from15minute to minute....self—esteem is different fromself concept. Self concept is a ‘theory’ or beliefthat youngsters have about themselves. Self conceptrefers to the individual’s personal perceptions ofhis or her view of life and of self. A child mayhold a view of himself or herself that does notcorrespond to his or her behavior. ...Self conceptis more stable than self—esteem and is alteredgradually. Self—esteem can and does change formsituationally. It is self—esteem that directlyinfluences the child’s performance or behaviorin a specific situation. (p.p. 146—148)Hoge, Smit, and Hanson (1990) seem to use self—esteemand self—concept interchangeably. They conducted a two yearlongitudinal study with 322 grades 6 and 7 students from twoschools, ranging in age from 11 to 13, looking at severalfactors including self—esteem. Each academic year thestudents completed two questionnaires, one in the fall andone in the spring. The students at school 1 had a highersocioeconomic status than the students at school 2. Globalself—esteem was measured using the Rosenberg Self—Esteem Scale. Academic self—esteem was measured using theSelf—Concept of Schoolwork Ability General Scale. “At the16ort‘—iiDDI—i(ft‘.0.—..U)U)oTHMix_.<C)IIVrJCDrZi-•r1MiP3‘.0Cl)CDP3Di•(1)oU)CDCD—C)101MiC)Cl)o•i--CDCDCDI—I.rC)Diii-b10U)i—’C)i-h(t0C10(ftCD-3e-CDl--‘.00(ft(ftCrH•(ftCDZCCD-ZC)C)Mi—ZC)CDC)ZC)zCD(Uw“010Cl)0•C)oDiDiCCDrcMirtCDCDMiH-Cl.<U)0CDi—’HiI—i0rtC11rt0..CD‘-i(U.-1i--CDCDOCDD0CDZCDH-CDCH-H-(0CD(0i-ii—s0—iI—’H-CDZ3C)uC)ZC)(UrC)HiCrri(I)C)‘.0(1)H-CrMiCiJCri--iCTCD.C))IY0C)C3C)C)—I-..2.Cr•.P3(1)MiCDU)I—’CDCDCDI-ti—UCD00COCCD0CDIC)CI—iU)C)Z‘0(ftZCD<i—.CDci—.t-----tiCDCDC))CMiZCDCrH-ZH-CDCDMi(ftCrZC)CDCD0..CTCD2.0Cti--’CDZC)C)ICrMiCOCD0CDCDCD‘<-MixCrCD(0H-C)Cl)I--Ci--iSI-C-ZICrCD0..CDCrCD0CD0(ft(ft0CDC0CTC(t5(ft(ftC)CtCDZ000‘0CTH-Ci.‘<C)0CDCl)CDCD‘<ciCDCDC)CZMirI-Cl)Cl)CDrrU)(ft5CDC))IDCi)C)CDH-30‘zj0)(I)CDC—•CDC)It0..‘Q•I—iCDCrCtU)CtftMir-rC)ID0..(ftCr(ftJ13<rØyU)C)H-0CD3CrH-H-CD-CrCDtCTCD(I)CTCTCrCl.CTCDCDCDCt0’,C))‘5H-0ii)CDCD“<CDCTCD‘i<itCDz•0--n0or-<CDi-iCD‘.030•CDCDH•0..0..(ftC)3I—CD(0CD(ftIt‘iCS10uiH-CTCDC))r-•CD1CDftCDCT‘—MiCD0CD3()I—i•(I)Cl.CDC)3CTCtH-0i-i-C)0X0..CDI(ftC)CD03’ctIrtClCD5I—’i-tii-5C)00..CT(j)l—5CDI-’Cl)Cl01CCDftc-I-C)000CDzciiCrI—iCDCrriCTH-0CT0H-0CrClCDCTI—’0Ct33’H-00ClCDCTt-103’05H-I-’3H<0CD(ft3Ci.CDCDCDI—CD0MiH-(ft3CD(ftH-3<CDCtCiCD0‘-<(ftCTC)c-fCDriCD03CDX3’CT0C)03’00Cl)‘00‘.00..‘(1CDCDCCuC<DiU)SCDCD(1)CDHiCD(ftCDCtrt(1)I—’(ftCDI-.C)‘t)CrSX0)1rrHi0ClCDCDClCDC))CDCD‘dzC)CTCH-H01(0Q..00..U)0ClCD0..3CDCiCDH-3CD(1)H-CDCD0CDU)c-i-Cl‘0(ftC)CD3U)i—‘.0CrCrMi—iCiH-CD•H-CI-c-I-00..(ft0C)Mi-i103’0(ftc-I-•CDH-0CH-CDH-ICDCDCDDiClU)c-I-Cr•3CD0ClU)H-‘.0U)‘0CD0..3I-1Cr3’01•03CD33•HU)Cl)CDCD•(ftCCDMiH-CDU)H-H-CI-Cl<i-rj2.CTU).—CI--3CD‘0DiCD(I)CDU)CH’3’H-I-hCDCl)0I—’ClI—’C)3CDCt‘.0(1)C)0=5I1CD‘.0H-CrI—’CD‘.0CD‘1310i-hU)Mi—H-CrCDt)HIc-I-‘CI3C)H 31rizona, using the Tennessee Self—Concept Scale. Resultsshow that “the difference between the total positive score,or overall level of self—esteem, for the Navajo and whitestudent samples was not significant” (p. 12).Wall and Madak (1991)Compared the academic self—concept of Indianstudents who live on the same reserve but whoattend either a public school or a band—controlledschool....[They also) compare[d] the two groupsof students with regard to the levels ofeducational aspirations that they perceived theirparents and their favourite teacher held for them.(p. 44)In this study, there were 20 students attending the band—controlled school and 22 students attending the publicsecondary school. These researchers used the Michigan StateGeneral Self—Concept of Ability Scale and cite others whohave used this scale “to study academic self—concepts ofboth Native and non—Native grade 7—12 students” (p. 46).Included in the above scale was the Perceived ParentalEvaluation of Ability and Perceived Teacher Evaluation ofAbility. “These two instruments were designed by Brookoverto elicit students’ perceptions of the academic expectations18held for them by parents and teachers” (p. 46) . Resultsshowed no significant difference between the two groups ofstudents in terms of academic self—concept. However:Students attending the band controlled schoolperceived their parents as holding significantlyhigher educational aspirations for them than didstudents who attended the public school....[alsojstudents attending the band—controlled schoolperceived that their favourite teacher hadsignificantly higher educational aspirationsfor them than did their peers who attended thepublic school. (p. 47)Scheirer and Kraut (1979) report that “severalcorrelational studies [which) have found strongrelationships between children’s self—concepts and theiracademic achievement” (p. 132). Luftig (1983) citesresearchers within whose work:Self—concept has been shown to be positivelyrelated to school achievement in elementaryschool children.. .and in school achievement withintermediate and secondary school pupils....Academicsuccess and positive self—concept have also beenshown to be positively correlated in Black and19Hispanic children. ..in the mentally retarded...inphysically handicapped children...in the sensoryimpaired...and in learning disabled pupils. (p. 252)Beane, Lipka, & Ludewig (1980) discuss research byWilliam McGuire who used interview techniques to determinethe self—concept of children. He worked with 560 children ingrades 1, 3, 7, and 11, and found age trends in the data.bout 15% of a grade 11 student’s self—concept centredaround school as opposed to 5% for a grade 1 student.McGuire concluded “that a student’s sense of self is tied toacademic performance and the quality of the relationshipshe/she has with fellow students and teachers” (p. 87) . Theseresearchers found similar results using McGuire’s methodwith 1,102 grades K—12 students. They concluded “that abouta fifth of a child’s sense of self is derived from theschool experience” (p. 87) and noted that students tended todescribe themselves in increasingly negative terms as theymoved through the school system.Altinann and Dupont (1988) conducted a study in whichthey attempted “to test the hypothesis that academic self—concept... [would] be a better predictor of report cardgrades than general self—concept” (p. 170). With a sample of198 students in grades 3 to 6, they administered the20piC‘Cl-Qit(i-iC)CDCl)CCl3L)3itCDYl—ui—ip3CDt-sCCl‘-1P3CDitClCi)I—iQCDCDitP3H(DCDHCDH-ClU)(I)3H-L—•0it(_)l—•itC)Cu3ZpiP3H-Cs)CDitC)Cl)3CDC)Cs)CflCDC)C)3CD0DiCD30CD3CDH-CDc0-Di3Cl3DiCli-riDi‘DiCDP3iC)ClClCs)H•Cs)flCD—-I-’-‘-Cl•H-HiCT)Ut3ClU)Cs)3nCl)H-U)l—3ClID(I)CT)C)ICD3CDCHiCi—i-zC)ClP3C)<DiCCli—iClçCli-tiHiCl0ClCDCDCTCDH-it0P3Ci)uH-DiCDCDt-..0CDClCDCl<I-riCD—C)CD3CClCt—I—Cs)C)CCl)çtCD3C)CD0itrCDCl)C)H-ClSi-’-Di3CL)zCs)-CD1DP3CDP3CDP3it0ciCl)03ClH-CDCl)tiCDCD‘DitCTi-i-‘d3Cl<33ClC)rtiit“0Cl-3DiCDC)H-TitC)Cl)0P33dH-U)H-itH-3-H-f—sP3CDCTC)ClF—iU)CDCDCl-CD0zH-çtU)<0C)H-CCDCDP3P333Dirtitii)H-3C!)itC)Cl3ClI—h303’H-Cs)<C)3itC)C)3’CDU)HiP3DiCL)I0ClitCD3C)3’3’U)ClH’0itC)I-ti3’3’U)H-CL)CuCDCCuC)0<3C0U)CDCl)CD0ClF—iP3C)ClEC)P3U)0CL)Cl)Cl3otI(1)Cs)0CC)ClCl)DiCl)CD3’P33H-H-t)CDC)CTC)I—’rtClCClC5H-ClCs)C)ctU)Cl3CDH-H-Cl)HiH-•U)CDCl(C)T)’UtC)CDH-Cl)CD1-’-i—I-I—’01)IH-3CDU)3’3’‘030tfl(I)itHF-’Ir-C)P3-ClH--p3H-3it(1)itC-c-I-CDHi0Cl-‘Ti3U)C)ç-H-C)H-Cl)Cl)C)Cl)“<C)3CD•H-3’Cl)U)I-bCiDiClititiDC)oitC)Cl‘O<DiC)UiCl)H-P33(j)H-I3P3Cl)CDCDI—’H-CDC)C)C)DiC)C)ClClI—’Cs)ClI-’C)o‘itC)Clit3’0Di3’CDCDP3it0ICDP3S5c-i-ZwP3C)3H-CClSH-35P3itH-50I—’H-Cs)CD3CDCDCD3CD0CDitH-C)H-C)CDc—ti-CDC)--it‘tI<itSCl<C)P’oP33IC)Cl3I—’c-i-C)CDH-CDCD-<Cl3I-’it‘<U)-(I)•CD<<H-U)SC)SCl)CDCs)CDflU)it•I-h0C)CDCl)‘-0CD3’CDS3’5C)P3U)U)H-0Cl3Cl)30Cl)F-’3H-I.—’H-H-CD0Clit3itCDCi)•CrClCD0it)frC)raiti-IU)CP3CD3’I—’•F-’CT3’ICs)3’r-ClP3-Cs)3ClCDCD0•itU)I—hCi’-CDC)Di0CDCD3ClU)itCl)•3’H-II—’‘TICl0C)0-ClI—’0CDCl’H-CDCl)tQC)33CDCl)P303—CTCs)NClit3’3305A()H-ititClC)Cl‘Ti0CD3H-H-3CDCDCD*H-it0•it-ClitCDCs)‘-0I-hC)P3•‘TI<CDct0Clit3’Di0H-I—’H-CDCs)C)itCDCD33C)Cl3’I-’)3l-1Ci)0C)Cu-i53CDP3CDCDitit(3’P3ClCDs)3ClF-’u-i3’CDi3-CD33QIP3CDClI—’ct•itClitH-•H-‘-<I-’•CDCDClQIClHICs)3Ithis study, 213 11—year—olds from three schools wereclassified socially’ based on their school records and theclassification of the father’s occupation by the Office ofPopulation, Censuses and Surveys. The Piers—Harris Self—Concept Scale was used to measure self—concept. Resultssuggests “that it is the child’s level of academic abilitywhich influences the self—concept, not their socio—economicstatus” (p. 227)Chapman, Lambourne, and Silva (1990) conducted a studyin which they “examined the associations between academicself—concept, reading performance (as a primary indicator ofschool achievement) , and antecedent cognitive and familybackground •iieasures over time....home background factorsincluded socio—economic status (SES)” (p. 143). Their samplewas made up of 435 children who were assessed from birth to11—years—of—age. Parental and family characteristics wereassessed using a number of different measures, as werecognitive and achievement levels (see Chapman et al (1990)for details)The longitudinal data in this study do notsupport the belief that family backgroundvariables have any major, long term effecton academic self—concept. Further, family22interest in cultural, political, andintellectual activities were not associatedwith either academic self—concept or readingachievement. Academic self—concept was affectedprimarily by past achievement and past perceptionsof ability... .home background variables appearedto have no direct or major indirect long lastingeffects on achievement or academic self—concept....Our results strongly support thehypothesis that academic self—concept is primarilythe result of different levels of schoolachievement. This result is consistent with studiesof the relationship between achievement and academicself—concept reported by Byrne (1984) and Hansfordand Hattie (1982) . (p.p. 149—150)Chapman et al. (1990) aptly sums:The experience of schooling, involving comparisonwith others and feedback about academic performance,is likely to be the main factor in the developmentof an academic self—concept. From that time, theinterplay between academic self—concept andachievement is probably reciprocal....Stanovich(1986) refers to this reciprocal effect as the23‘Matthew effect’, in which competent achievers dobetter over time, and poor achievers become moreself—deficient over time....In essence then,academic self—concept would seem to be closelylinked to actual achievement outcomes in school.(p. 150)Some studies have shown the correlations betweengeneral self—concept and academic achievement to “have beenmildly to moderately positive. ...These results have beeninterpreted by some as indicating (that] the importance ofself—concept in education has been overemphasized,especially if viewed as a potential predictor of academicachievement” (Lyon & MacDonald, 1990, pp. 1135—1136; seealso Canton, 1981; Chapman, Lambourne, & Silva, 1990;Madak, 1988; Riffel, 1991). Chapman et al. (1990) state,“academic self—concept has a stronger relationship withschool achievement than does general self—concept withachievement” (p. 142).Lyon & MacDonald (1990) note that the results of recentstudies suggests that “there may be many facets to theconstruct of self—concept” (p. 1136) (see also Carlton,1981). In viewing self—concept as having severalcharacteristics, some researchers have come up with a24hierarchical model (Shavelson, Hubner, & Stanton, 1976).General self—concept is assumed to be at the top of thisstructure. At the next level down, are foundtwo specific components: academic self—concept andnonacademic self—concept. Academic self—concept would thenbe divided into specific subject areas and nonacadernic self—concept would be divided into its emotional, physical, andsocial factors. However, there is no consensus amongresearchers for a model on the structure of self—concept.Shavelson, Hubner, and Stanton (1976) who have proposedseven characteristics of self—concept and believe that thereare both multidimensional and hierarchical facets to self—concept, have received support from recent research(see Byrne & Shavelson, 1987). In their study, Byrne andShavelson (1987) tested “the invariance of a multifaceted,hierarchical SC [i.e. self—concept] structure for adolescentmales and females” (p. 379) . Using a self—report battery ofSC which included three general SC scales, three academic SCscales (Self—Concept of Ability Scale——Form A was found tobe the most reliable measure of academic SC), three EnglishSC scales, and three mathematics SC scales, they found “thatthe assumption of an invariant SC structure for males andfemales cannot be taken for granted; relations among SC25facets do differ across gender. The findings also show thatSC instruments measure particular facets of SC in differentways, and with different reliabilities for males andfemales” (p. 382)Lyon & MacDonald (1990) suggests that previouscontradictory findings “may be partially a function ofinadequate operationalizatiori of the self—concept” (p.1136) . In their study of 122 3rade 6 students in a suburbanmidwestern school district, they assessed 67 girls and 55boys on cognitive abilities, general self—concept, academicself—concept, and locus of ccntrol (academic experiencies inthe academic environment) . They reported significantlyhigher correlations between academic self—concept andacademic achievement than between general self—concept andacademic achievement and between locus of control andacademic achievement (italics mine) . The researchersconcluded that academic self—concept is a variable worthy offurther study in understanding students’ achievementbehaviour. Butterfield (1983), in commenting on the highdropout rates among high school students pointed out that“It is riot surprising, then, that educators of Indianstudents see a lack of positive self—image in thesestudents” (p. 51). (see also Luftig, 1983; Pauls, 1984).26“[There] is also evidence to suggest that self—concept ofNative American children is negatively correlated withchronological age and years of schooling....[The]relationship between low self—concept in American Indianchildren as a function of academic achievement has also beenreported by a variety of...experimentors” (Luftig, 1983, p.p. 252—253). In addition, some studies “suggest[s] that theschool and non—school environment can have a profound effecton the child’s self—concept” (Metcalfe, 1981, p. 66).Rampaul, Singh, and Didyk (1984) published a studywhich investigated the relationship between academicachievement and self—concept among Native Indian students.In their study, 22 grade 3 students and 19 grade 4 studentswere tested for academic self—concept using the MichiganState General Self—Concept of Ability Scale. Academicachievement was measured by the canadian Tests_of BasicSkills. They found significant positive correlations [therange was .43 to .61 1 between “self—concept and academicachievement across all grade levels and for both sexes” (p.219)Pepper and Henry (1991) point out that “the self—concept of school—related ability is a better predictor ofsuccess in school than is overall self—concept. It appears27that an important limiting factor of achievement in schoolis the student’s concept of his or her ability” (p. 149).Teachers’ perceptions of Indian students in theirunderstanding of English in the classroom may also presentproblems for the students’ self—concept. “Perhaps localcontrol of their children’s education would produce betterresults” (Huriburt, Henjum, & Eide, 1983, p. 17).Although there is still not a clear, acceptedoperational definition of self—concept, it “has beenidentified by many educators as a critical variable in theeducation of children” (Madak, 1988, p. 4).Madak (1988) suggests the following guidelines whenmeasuring self—concept:(1) select measuring instruments which have highvalidity and reliability ratings;(2) select measuring instruments that are appropriatefor the population being tested (eg., know thereading level of your population, ethnicbackground, etc.)(3) make sure that the instruments are administeredand scored appropriately; and,(4) realize that no self—concept instrument willproduce an exact score. Therefore, all scores28must be viewed as an imperfect estimate. (p. 6)He also cites research which indicates that Nativechildren have lower self—concept ratings than their non—Native counterparts. Madak (1988) concludes from his reviewthat for Native students, the concept of self includes thestudent’s relationship to his/her culture., However, he alsopoints out that some researchers believe “it is poverty, notcultural differences which account for poor self—concept andschool performance” (p. 19).Chapman, Cullen, Boersma, & Maguire (1981) investigatedthe “interrelationships between....general and academicself—concept, academic locus of control, and self—expectations of future academic performance, along with theschool achievement one year later...in elementary schoolchildren” (p.182). There were 376 students in grades 3 to 6from predominantly middle class families in their sample.Variables were assessed using available scales such as thePiers—Harris Children’s Self—Concept Scale. Results showedacademic self—concept and expectations individuallycorrelating with report card grades. They also found that:Attributions of responsibility over successfulschool outcomes are moderately related to schoolachievement, whereas perceived control over29failure is not....The implication of this findingis that success in school is partly related tobeliefs in individuals that they have somecontrol over the cause for successful outcomes.Failure outcomes, on the other hand, are notseen as related to achievement, at least at theelementary level in this sample. (p. 188).In addition, their results showed “low and non—significantcorrelations between general self—concept, as measured bythe Piers—Harris, and all the other variables investigatedin the study. Such a finding is consistent with the notionthat global measures of self—concept are less fruitful inpredicting school achievement than more specificinstruments” (p. 188). Chapman et al. (1981) also point outthat “considered together, the results of the regressionanalyses indicate that, as would be expected, 1977 gradesare the strongest predictor of 1978 grades” (p. 189).Zarb (1981) explored the relationship between a numberof variables, including academic self—concept and academicachievement. 128 grade 10 students, 15—16 years—of—age,attending an inner—city secondary school were selected forthis study. Most of the students were from “the lowestsocio—economic category” (p. 894). Some of the students had30repeated a grade in the past, some had failed more than twoof eight courses at the end of grade 10, and some had failedeither one or two courses. Results showed that “only poorAcademic Self—Concept was significantly related to pooracademic performance... .Academic Self—Concept was the bestsingle predictor of Grade Point Average for both the maleand female samples....when the effects of the other fivevariables were partialled out” (p.p. 897—898)In sum, this is still obviously an area in which muchresearch needs to be done in order to more fully understandthe relationships between general self—concept, academicself—concept, and academic achievement.Factors Related to Attendance“The relationship between acceptable levels ofattendance and the high school diploma is crucial. Theassumption is that for learning to take place, the teachermust have learners in attendance, and a consistently highlevel of attendance is necessary for learning” (Brodbelt,1985, p. 64).High dropout rates among Indian students is a wellestablished fact. “The reasons for dropping out of school31can be broken down into three main categories: (a) socialfactors; (b) student factors; and, (C) school factors”(Persaud & Madak, 1992, P. 236). Among severalcharacteristics, “belonging to a language minority group”has been identified as a social factor associated withdropout behaviour” (p. 237). Among student factors, someresearch suggests that high school dropouts did not feelthat school had something “important to offer to them”(p. 237). Among school factors, “one of the best predictorsof school dropout behavior has been reported to be academicperformance....furthermore, students who dropped out ofschool were reported to have...lower locus of control thangraduates....[in addition] school attendance has been shownto be another strong predictor of school dropouts behavior”(p.p. 237—238)Persaud and Madak (1992) cite researchers who “havehypothesized that, when taken together, the evidenceconcerning why students drop out of school strongly suggeststhat dropout behavior is a long—term process in which thestudent slowly becomes disenfranchised from school and theschooling process” (p. 248).Among the Native students who remain in schoolabsenteeism is a prevalent problem (Berger, 1973;32Butterfield, 1983; Chavers, 1991; Dunn & Griggs, 1990; Gade,Huriburt, & Fuqua, 1986; Huriburt, Krocker, & Gade, 1991;Kirkness, 1981; Pauls, 1984; Riffel, 1991).Brodbelt (1985) proposes that “absenteeism must beexamined from two bases: (a) the nature of out—of—schoolfactors that give rise to poor attendance, and (b) the in—school factors that are influential in causingabsenteeism....[because in the long run] it is theindividual pupil who is the real loser when he or she isconsistently absent” (p. 64). Brodbelt (1985) further citesWilbur Brookover who found that academic achievementincreases as attendance increases. In addition:Attendance is influenced by a combination ofmany overlapping factors. Wilbur Brookover(1982) found that effective attendance relatedprimarily to the school ideology. The ideology‘refers to the general beliefs, norms,expectations, and feelings which characterizethe school social system. The belief thatstudents can learn arid that teachers can teachis an important characteristic of an effectivelearning environment’. (p. 65)Further, Brodbelt (1985) cites a 1982 study by John Easton33I_t,rcift—.ri-CD‘-C)t--Cii‘c.I0C:CDUH(I)Ui—i•‘-CTInDDi0Dil.J00ft—.i01Di01Oici)H•I‘Di-”CDc-rCTI—IUi()Cl)-0i-h‘-CDCDZF—iC)i-hi—i-Cn01MiF-•Cl)0I—tiC)•0i-IciiDiC)rIC)cliDiftCC)UCDC:ciizi-ioZ(I)Di0ZCDft:iaft0-’0‘-001CtDiC:CDi—’Di0C)CDC)iCC)01(1)H-ftZI—’-‘-CiU<Ui‘-<H•ZUMiCDi—’-Cl)C)CDciiDii-tiC)CD1‘-CltZ(C)I—<Di1.1)C:CD0C)01rMi(I)C)CTC:Q0‘-ci-•3CDCD1H-r-C:113C)CD0HU.C)‘‘-ClZC)Z01(iiCDi(I),—-Q1)(1)çrci)C)UC)I—-’C—’hC—aC:—i-a-.Ct)ftI—’DiCL)CY’i—CIClCl)‘-<Ct--iiU—U.ii)‘-<Cl)-‘(C)i-DiciCDCDCD,—-ciia(t•C)rClDiSZIICDt2,ft(C)ft(DiClCDCl-.C)ftL)—‘Di0ClUici)—“CDi---’•Di-L)ClCDDi2“DiCDCDHH-iD•(C),—‘01—aWZftC)DiDil—Dic,i-.(iQHCDZCi)‘-<Z-1I—I—C)(SICliC:‘-CiCDCliMi11)ft0C)“i-Z(SICii0C:I—’H-0‘-<ClC:Difti-iftCl0Di<c-I-0‘-<NC—’C-’-I—’-ti‘-ClCDCliCDCDDiH-UUDiftCDDlc-c015CDk<0Cl)‘-iiClCl)DiDift01Ui—’-•I—’Z‘-(3CDCri1.ZI—’C)0ci)<C)U-‘iaDici)CDDiCDci)UCDci)ZI—’ZHC)-1o01DiClClft‘-<c-c0MiCDC)U)CliCliC--a0cft<ftC:‘-Cii—.DiC:(C)CDCliH-CDSUCDH-Cl(0CDI—’IIC:ClH<<<I—’<ftC)DlDiCC)Cl)DiI-tiCl)rC)ftftftftC-”H-CDH-CDN0Uc-lCDCl)001Cl)ZftDiDiC)CYUCDF--(C)IDZ C)CDCD CD(C) ZDi(1•r- Cl)H ZU) Hc-fc-U (C)C-’-CDh-cCDH-C)‘-0Dic—czDiftClCD‘-ciCDc-1-(D•ClH-C)‘-cift•0 CDUi—0 I-b c-c CD cii 1. 1.0ci)Dia’c-i(0C)rtCDHCUftDl‘-0Dic-”CD01ZftCDiH-HH-<ZI-hoCDC)cC)C-.0()U)CDZ‘-0Di•ZftI—cZ-Ct‘-1L)ift0ClDiC—’CD‘-<oc-CDi-ciiCDDiDiCDoiic-izI-’(C)C—’ClCl)CL)Dic-Ii0CTCDCD-H’ft0CDCDDi-iiCDZftCDCl0CDUHDi<‘-<ZCTCDC)Cl0i-I-I—’c—’.(lftUCNUUCD01CD iici-Dir-ZCDZCl)‘-0ftDiCl‘-0CDC)i-c‘-CICDUIDCD-1ftCDlI.(0IIc-0Cl)<U)I-.H-C)Difl000(1)UZCc-cCl)(flCDHLQ ci)CD0101CDDi z Ui0 CD 0 1 La (C) Di‘-a H’CD U Dl c—i ClU 0 I—’0 0 CD 01 Di ft c-I-U CD Di ft ft CD z 01 Dl C) CD Di 01CD-1C)CDUci)o01oHi-.i‘-aCDUftDi(C)C:()01UCDI-’.hiZCDoi-<CUiCDz•ClCD z—ft U0DiItic-iCli-iC)Cl)I-ic-I-0CC:a CD0ftMiDiCDUI—’CoCDCD zCD(_)z(1)ft P)c--cc-iDl ft CDDlCl)Z 01Cl) i-iU.CDC: H-0 I-i Uto identify variables which may effect the academicattendance of some grades 3, 6, 8, and 12 Navajo students inArizona. Previous studies within this particular schooldistrict had indicated socio—economic status (SES),stability, and average yearly attendance as possiblyeffecting student achievement as measured by standardizedtesting. In these tests, most of the Navajo studentsconsistently scored below their public school counterpartsacross the state.Stability was defined in the study as the “student’stotal length of uninterrupted enrollment at the publicschool” (P. 24). The median for grades 7 and 8 was 5 yearsand “this was selected as the lower limit at which studentswould be labelled stable” (p. 24). The average dailyattendance was determined by totalling all studentsattendance over the last seven quarters in a grade. Then anaverage was set for each grade which eliminated students whohad less than the seven quarters attendance. “Students wereassigned a status of above—average attendance or below—average attendance for that grade level” (p. 25). Socioeconomic status was determined by a student’s mealclassification. Parents could apply for free meal benefitsfor their children based on government standards for income35level and family size. In this study, a student wasclassified as high SES if he/she did not qualify for free orreduced meal benefits, and as low SES if he/she did.There are three immediate problems in looking at theresults. First, all data were compiled from cumulativerecords kept by others for different reasons. Secondly,sample sizes were not given and cost determined samplingprocedures. The third problem is the lack of reportedstatistical evidence to support the conclusions. However, ofconcern here, are the conclusions drawn regarding the effectof attendance on academic achievement.The effect of school attendance on achievement appearedto be linked to SES. In other words, absences appeared tohave little effect on achievement of high SES students. Theauthors concluded that their results were consistent withthose of other researchers, whom they cited, who found alink between school attendance and a student’s academicachievement.Monk and Ibrahim (1984) point out that “if some periodsof instruction a.re more important for learning than areothers, then the timing of absences can have substantialeffects on how much a given student learns....absence notonly reduces the amount of schooling time but causes a36disruption in the sequence of learning” (p.p. 296—297). Theyfurther observe that “the impact of a day of instructionlost can differ in important ways for an individual student,depending on the number of classmates who are absentsimultaneously” (p. 299). In their study, a sample of 227grade 9 students from a college preparatory algebra programwere selected for investigation because “the effects ofabsentee patterns are likely to be most evident in highlysequenced types of instruction where emphasis is placed onwhole class instruction” (p. 300) . For their variables, theychose to use the results from a state—wide achievement teston the algebra curriculum as the dependent variable, andattendance data from school records. (see Monk & Ibrahim fordetails on determining the quality of a student’s absence).The results demonstrate that apparently “the timing ofabsence makes a difference in terms of students’achievement. However, the data do not indicate that absencesduring one period...of instruction are consistently moreimportant than absences occurring during other periods” (p.308). They sum, “our results show that being absent whenlarge numbers of classmates are absent has an insignificanteffect on learning. It is quite another matter to be absenton a high attendance day” (p. 308).37The present body of research in this area appears tosupport the need to further explore the relationship betweenschool attendance and academic achievement for the Nativestudent. As Brodbelt (1985) observes, “learning is dependentupon the availability of the learner” (p. 66).Factors Related To Socioeconomic StatusChapman, Lambourne, and Silva (1990) cite researcherswho have found an association between self-concept, (inparticular academic self—concept), and familycharacteristics such as its structure, social status, andpsychological characteristics. For example, “Song and Hattie(1984) found that social status was a significant influenceon academic achievement through its effect on familypsychological characteristics and self—concept” (p. 143).Barnes and Vulcano (1982) state that, “Research...hasshown that lower social class students have lower academicself—concept scores than higher social class students evenwhen ability levels are controlled” (p. 61).These results are based on studies conducted among non—Native populations. Cultural differences may be a factor indifferentiating between students’ socioeconomic status.38Statement of the HypothesesBased on the literature reviewed, the followinghypotheses are proposed.1. Academic self—concept will decrease as grade levelincreases.2. There will be no significant difference between generalself—concept and grade level.3. Academic self—concept will positively correlate higherwith academic achievement than general self—concept withacademic achievement. The correlation between generalself—concept and academic achievement will be close tozero. These relationships will be stronger for grades7 and 8 students.4. Attendance will positively correlate with academicachievement.5. The socioeconomic status (SES) of the students willpositively correlate with academic achievement, i.e. thehigher the SES of the student, the higher his/heracademic achievement.6. SES will also positively correlate with both academicself—concept and general self—concept.397. Academic self—concept, general self—concept, attendance,and socioeconomic status, with varying degrees ofstrength, will all be seen as predictors of thedependent variable, academic achievement.40CHAPTER THREEMETHODOLOGYStudentsThe population for the study was grades 7 to 11 Nativestudents from an isolated community in Northwestern Ontario.Ages varied from conventional placements as schoolattendance is volatile. Some students are mobile and havenot have completed all of their education at the school onsite. In addition, dropping out of school starts to occur ingrade 8.The students are instructed through subject basedcurriculum guides supplied by the Ontario Ministry ofEducation. Modifications are made by classroom teachers tosuit the northern environment and Native culture.The total population for grades 7 to 11 was 117students. Of this, a total of 70 students participated inthe study. These students were all volunteers. There were 20students from grade 7, 27 students from grade 8, 13 studentsfrom grade 9, 4 students from grade 10, and 6 students fromgrade 11. It should be noted that a number of students leave41the community each year to attend high schools in largercentres. This affected the number of students available forthis study and may also have had an effect on the obtainedresults.SchoolThe school is located in an isolated NorthwesternOntario community of approximately 1600 Cree Indians,approximately an hour and a half by air northeast ofWinnipeg, Manitoba.The school originated under the direction of Indian andNorthern Affairs Canada. However, plans were eventuallyundertaken to place control in the hands of the community.This was achieved in 1989 and the education of the studentsin now under the control of the local Education Authoritywhich is staffed by community members. This change isconsistent with the position paper of the National IndianBrotherhood, whose aims were (1) parental responsibility and(2) local control of education (National Indian Brotherhood,1972)Education is provided for students from juniorkindergarten to grade 11. The total enrollment is42approximately 500. There are 27 teachers. 17 are Native withBand membership. The remaining 10 teachers are non—Native.InstrumentsThe purpose of this study was to investigate severalvariables that have been shown to affect academicachievement. Of primary interest was the correlationalrelationships between the independent variables of academicself—concept, general self—concept, socioeconomic status andattendance, with academic achievement as the dependentvariable. Of secondary interest was determining which of theindependent variables were predictors of the dependentvariable.To investigate the proposed hypotheses of this study,two instruments were used: The Michigan State Self—Conceptof Academic Ability Scale (Brookover, Erickson, & Joiner,1967); and the Piers—Harris Children’s Self—Concept Scale(Piers & Harris, 1969). These instruments were selected forthis study because of their widespread use andrecommendations by other researchers.The Michigan State Self—Concept of Academic AbilityScale (SCA) was used to measure the students’ academic self—43concept. The SCA is an 8—item Likert—type self—report scale.Each question has 5 responses which are labelled ‘a’ through‘e’, with ‘a’ being the most favourable response. Astudent’s score is determined by assigning values of 5 to 1,with the most favourable answer, ‘a’, receiving a value of5. A summation of values yields a potential score of 40points.A normative sample of 1,050 grade 7 students wasoriginally used to determine the reliability and validity ofthe SCA (Brookover, Paterson, & Thomas, 1962). Internalconsistency reliability measures produced coefficients of.84 and .82 for grade 7 females and males respectively(Paterson, 1966) . Later studies involved students fromgrades 7 to 12. For example, test—retest coefficients ofstability over a one year period with a sample of 5,976grade 8 to 12 students, produced results from .724 to .688for females and males combined (Brookover, Erickson, &Joiner, 1967). In a six year study using a sample of 7,126grade 7 to 12 students, correlations ranged from .48 to .63(Brookover et al., 1967).Validation studies which specifically used the SCA toinvestigate the relationship between academic self—conceptand academic achievement yielded the following results: (a)44Calsyn and Kenny (cited in Byrne, 1984) conducted a fiveyear longitudinal study which involved 556 grade 8 to 12students. They reported a correlation of .56 for the totalsample at the grade 8 level; (b) Shavelson and Bolus (citedin Byrne, 1984) published a correlation of .37 for 99 grade7 and 8 students; (c) Morse, in a 1963 study which involved114 Black students and 1482 White students, foundcorrelations of .43 and .65 respectively; and, (d) in a 1964study of 100 delinquent boys, Haarer reported a correlationof .41 between academic self—concept and academicachievement without IQ partialled out, and a correlation of.39 with the IQ effect controlled.The Piers—Harris Children’s Self—Concept Scale was usedto measure the students’ general self—concept. This scaleconsists of 80 first—person declarative sentences. Thesentences are set up to be answered either ‘yes’ or ‘no’.‘Yes’ is an indication that a student feels the sentencedescribes the way he/she feels about himself/herself.student’s raw score (total number of responses in thepositive direction) can be converted to T—scores and areavailable as an overall self—concept score or as a profileof six cluster scores, (1) Behaviour; (2) Intellectual andSchool Status; (3) Physical Appearance and Attributes; (4)45Anxiety; (5) Popularity; and, (6) Happiness and Satisfaction(Piers, 1984). “The Piers—Harris Children’s Self—ConceptScale...is one of the most recommended global self—conceptinstruments available for use with childreri...and isintended for use with children in grades 3 through12....Reliability in a test—retest situation over a four—month period ranged from .71 to .77” (Altmann & Dupont,1988, p. 171).The Piers—Harris was standardized on 1,183 studentsfrom grades 4 to 12 in a small Pennsylvania town. It has notbeen renormed by its authors since its original developmentin the 1960s. However, as Epstein (1985) notes:Recent reliability studies generally confirmand expand on the results of the originalstudies. Test—retest reliabilities ranged from.42 to .96, with a mean of .73. Studiesinvestigating internal consistency yieldedcoefficients ranging from .88 to .93 on thetotal scale. In another study using the scoresfrom the original norm group, the internalconsistency coefficient for the total scale was.90, with the cluster scales ranging from .73to .81. Thus the instrument appears to be highly46reliable in terms of temporal stability andinternal consistency....In the realm of inventoriesof this nature, the Piers—Harris is apsychometrically adequate instrument whoseusefulness in research has been documented. (p. 1168—1169)Jeske (1985) aptly sums, “the Piers—Harris appears to be thebest children’s self—concept measure currently available. Itis highly recommended for use as a classroom screeningdevice, as an aid to clinical assessment, and as researchtool” (p. 1170)The academic achievement of the grade 7 to 9 studentswas determined from raw reading scores obtained from theCanadian Tests of Basic Skills (CTBS) and the grade 10 to 11students from teacher rankings. These rankings were assessedby the language arts teacher as grade equivalent standings.The CTBS is a teacher—administered and teacher—scoredtest. Hoge, Smit, and Hanson (1990) note that “academicachievement can be measured by grades or by standardizedtest scores” (p.. 117). Since reading forms the basis for allthe other subject areas, data on this subject provided aneffective assessment of academic achievement.lifelong member of the community who is also employed47by the Education Authority, rated the students as beinghigh, middle, or low according to their parents/guardianssocioeconomic status (SES). This rating was done on thebasis of source of income or source of employment. Thus,this rating was not the students’ actual SES.Students of parents who worked at the school were ratedas “high”. Students of parents who were employed by the bandwere rated as “middle”, and students of parents who dependedon Social Assistance were rated as “low”. Ratings were codedaccording to 1 = high, 2 = middle, 3 = low. Thus,significant correlations between SES and any of the othervariables have a negative sign in the Tables found inchapter 4.Attendance records for each student were accessed todetermine the number of days absent from school during thepresent school year. This number was used in the analyses.Hence, significant correlations here also appear with anegative sign in the summary Tables found in chapter 4.The attendance data for the present study was notanalysed to account for the importance of when absenteeismoccurred as noted in the chapter two literature review.48Experimental DesignA correlational design incorporating multipleregression and analysis of variance was used in this study.The choice was based on the number of variables involved andthe number of grades in the sample. Also, from theliterature review it was determined that this was the choiceof design primarily used for studying these variables.The design was selected to investigate therelationships between academic achievement and the followingvariables: academic self—concept, general self—concept,attendance, and socioeconomic status. Of secondary interestwas determining which of the independent variables werepredictors of the dependent variable.The study involved grades 7 to 11 Native students froman isolated community in Northwestern Ontario.ProcedurePermission to conduct the study was given by the localEducation Authority.A total of seventy students participated. They wereassessed over a period of four days in their classrooms.49Assessments lasted from 15 to 20 minutes. Prior to andduring each testing session, students were given theopportunity to ask questions.Statistical Analyses1 Kruskal—Wallis test was performed among grades 7 to 9students (group 1) and grades 10 and 11 students (group 2),to ascertain the use of a standardized reading score as anindicator of academic achievement.The data for academic self—concept and general self—concept were descriptively analyzed by computing the meansand standard deviations. An analysis of variance was carriedout between the different grades for the means associatedwith (a) academic self—concept; and, (b) general self—concept. Internal consistency measures of the Michigan StateSelf—Concept of Academic Ability Scale and the Piers—HarrisChildren’s Self—Concept Scale were calculated for the sampleusing the Pearson r. The Spearman—Brown prophecy formula wasapplied to the resulting coefficient in order to estimatethe reliability of each test.Correlations were calculated to determine the strengthsof any relationships among the variables.50A multiple regression analysis was carried out withregard to the predictive relationship between theindependent variables and academic achievement.51CHAPTER FOURRESULTSThe results pertaining to the research hypothesestested in this study are presented in this chapter in threesections.The grade 10 and 11 students were analyzed separatelyfrom the grade 7 to 9 students, since academic achievementwas measured differently for both groups (see chapterthree).First, the effectiveness of the measure of academicachievement was investigated. This analysis (Kruskal—WallisOne—Way Anova on the reading means, i.e. academicachievement for each grade group, [group 1——grades 7—9;group 2——grades 10—111) is presented under the heading‘Academic Achievement Measure’.Second, correlational relationships were determinedfor the five variables. Analyses concerning thecorrelational relationships between academic self—concept,general self—concept, socioeconomic status, attendance, andacademic achievement are presented under the heading‘Correlational Relationships among the Variables’.Third, predictors of academic achievement wereexplored. The analysis concerning the stepwise multiple52regression analysis among the five grade levels is presentedunder the heading ‘Predictors of Academic Achievement’.Academic Achievement MeasureThe Kruskal—Wallis (K—W) procedure is a nonparainetricalternative to analysis of variance. In this analysis it wasused to evaluate the measure of academic achievement.The K—W procedure was used for both groups. The result(b = 3.5014; p > .05) show that differences between grades7—9 students in relation to their reading achievement scoreswere not statistically significant (i.e. there is nodifference in the measure of academic achievement based onstandardized reading scores between the 7, 8 and 9 gradelevels). A correction made for ties in the data was notsufficient to alter the above conclusion. The result (b=1.1364; p > .05) obtained for the grade 10 and 11 studentswas also not statistically significant, even with acorrection made for ties.Correlational Relationships among the VariablesThis section outlines the findings for the53correlational analyses among the five variables.As an introduction, summary matrices are presented for thetwo grade groups. Table 1 is a presentation of theintercorrelations for grades 7 to 9 among the fourindependent variables and the dependent variable. Table 2facilitates similar information for grades 10 and 11.Table 1Intercorrelations of Academic Achievement (AA) , AcademicSelf—Concept (ASC), General Self—Concept (GSC), Attendance(ATT), and Socioeconomic Status (SES) For Grades 7 to 9AA ASC GSC APTASC 0.115GSC —0.132 0.642**ATT —0.109 _0.344** —0.078SES 0.056 —0.192 —0.244 0.162n = 60; *p < .05; **p < .0154Table 2Intercorrelations of Academic Achievement (AA), AcademicSelf—Concept (ASC), General Self—Concept (GSC) , Attendance(ATT), and Socioeconomic Status (SES) For Grades 10 and 11AA ASC GSC ATTASC 0.087GSC —0.320 0.003ATT 0.275 _0.660* 0.045SES 0.345 —0.517 —0.039 0.228n = 10; *p < .05; **p < .01As would be expected, there is a significantcorrelation between attendance and academic self—concept forboth groups. The minus sign in both tables is indicative ofthe fact that attendance was measured as the number of daysabsent from school.The grades 7—9 students also showed a significantcorrelation between general and academic self—concepts. Thiswas also expected since academic self—concept is seen as afacet of general self—concept. A zero correlation was found55between general and academic self—concept for the grade 10to 11 group although this was not predicted by theliterature.The first hypothesis tested in this section was thatacademic self—concept would decrease as grade levelincreased. Results from the Michigan State General Self—Concept of Ability Scale (SCA) are presented in Table 3.(This scale despite its name is a measure of academic self—concept).Table 3Michigan State General Self—Concept of Ability ScaleMean Scores by GradeMean Standard Deviation Range SampleGrade 7 25.5 4.35 17—34 n=20Grade 8 26.1 4.12 18—34 n=27Grade 9 23.9 3.48 17—29 n=l3Grade 10 28.8 5.32 21—33 n=6Grade 11 25.8 4.45 20—31 n=4*p < .05; **p < .0156A one—way analysis of variance of academic self—conceptacross grade levels (F = 1.1398, p > .05) revealed nostatistical significance. Thus, hypothesis one was rejected.The Spearman—Brown prophecy formula was used toestimate the internal consistency reliability of SCA for thesample. The internal consistency reliability measure for theSCA yielded a coefficient of .78 for the five grades.The second hypothesis tested was that there would be nostatistically significant difference between general self—concept and grade level. Results from the Piers—HarrisChildren’s Self—Concept Scale are presented in Table 4.Table 4Piers—Harris Children’s Self—Concept ScaleMean Scores by GradeMean Standard Deviation Range SampleGrade 7 44.7 13.53 20—65 n=20Grade 8 49.7 14.83 19—74 n=27Grade 9 47.5 15.20 21—70 nl3Grade 10 47.3 16.03 32—66 n=6Grade 11 51.0 12.73 31—69 n=4*p < .05; **p < .0157A one—way analysis of variance of general self—conceptacross grade levels (F = 0.4239, p > .05) revealed nostatistical significance. Thus, hypothesis two was acceptedas tenable.The internal consistency reliability measure (Spearman—Brown) for this scale yielded a coefficient of .94 for thefive grades.The third hypothesis tested was that (a) academic self—concept would correlate higher with academic achievementthan general self—concept with academic achievement and dueto the huge dropout rate prevalent among this population (b)these correlations would be lower for the high schoolstudents.Academic achievement scores (i.e. raw reading scores)for grades 7 to 9 were obtained from the Canadian Tests ofBasic Skills (CTBS). Since CTBS score weren’t available,academic achievement for grades 10 and 1]. was determinedfrom rankings of the students provided by the language artsteacher.As shown in Table 5, the results indicate a highercorrelation between academic self—concept and academicachievement for grades 7—9 compared to grades 10—11. Thereverse can be seen for the correlations between general58self—concept and academic achievement.At the p < .05 level of significance for criticalvalues of the correlation coefficient, neither group wasstatistically significant. Thus, hypothesis three (a) wasaccepted, although the results are not significant.Hypothesis three (b) was rejected.Table 5Correlation between academic achievement (AA) and bothacademic and general self—concept for grades 7—9 andgrades 10—11Grades 7—9 Grades 10—11(AA) (AA)Academic Self—Concept 0.115 0.087General Self—Concept —0.132 —0.320n=60 n=l0*p < .05; **p < .01The fourth hypothesis tested was that school attendancewould positively correlate with academic achievement.59In order to test hypothesis 4, attendance records foreach student were obtained from their teachers. The totalnumber of days absent was correlated with their academicachievement.The results of the correlation between attendance andacademic achievement are shown in Table 6.Table 6Correlation between academic achievement (AA) andattendance for grades 7—9 and grades 10—ilGrades 7—9 Grades 10—il(AA) (AA)Attendance —0.109 0.275n=60 n=l0*p < .05; **p < .01The correlations are low but nevertheless, in the caseof the grade 7 to 9 group, positive. This contrasts withgrades 10 to ii where the correlation is negative. (Recallthat number of days absent was correlated with academic60achievement). At the p < .05 level for critical values ofthe correlation coefficient, neither correlation wasstatistically significant. Thus, hypothesis four is acceptedfor grades 7 to 9 and rejected for grades 10 and 11.The fifth hypothesis tested was that the socioeconomicstatus (SES) of the students would positively correlate withacademic achievement, i.e. the higher the SES of thestudent, the higher his/her academic achievement.Results of the correlation of both groups are shown inTable 7.At the p < .05 level of significance, the correlationsare not significant for either group. Recalling that SES wasTable 7Correlation between academic achievement (AA) and SESfor grades 7—9 and grades 10—11Grades 7—9 Grades 10—11(AA) (AA)Socioeconomic Status 0.056 0.345n=60 n=l0*p < .05; **p < .0161coded as High = 1, Middle = 2, and Low = 3, negates thepositive correlations reported in this table. Thus,hypothesis five is rejected.The sixth hypothesis tested was that SES wouldpositively correlate with (a) academic self—concept and with(b) general self—concept.The results of hypothesis 6 are shown in Table 8.Again, recalling the method of coding SES negates thenegative signs in Table 8. At the p < .05 level ofsignificance for critical values of correlationalcoefficients these correlations are not statisticallysignificant. However, hypothesis six is still accepted.Table 8Correlation between SES and both academic and generalself—concept for grades 7—9 and grades 10—11Grades 7—9 Grades 10—11SES SESAcademic Self—Concept —0.192 —0.517General Self—Concept —0.244 —0.039n=60 n=l0*p < .05; **p < .0162Predictors of Academic AchievementIt was noted in the literature review presented inchapter 2, that academic self—concept, attendance, andsocioeconomic status have all been known to be predictors ofacademic achievement. However, general self—concept has notbeen demonstrated to be a useful predictor.A stepwise multiple regression analysis was used toexplore the relationships between the independent variables(academic self—concept, general self—concept, attendance,and socioeconomic status) as predictors of academicachievement (hypothesis # 7)The regression analysis for the grade 7 to 9 groupproduced R = 8.8% (F = 1.322, p > .05), which does notachieve statistical significance.The regression analysis for the grade 9 and 10 groupproduced = 60.8% (F = 1.935, p > .05), which is also notstatistically significant.Hence, these independent variables are not significantpredictors of academic achievement for this sample andhypothesis seven was rejected.63CHAPTER FIVEDISCUSSION AND CONCLUSIONSDiscussionThe purpose of this study was to investigate thecorrelational relationships between attendance, generalself—concept, academic self—concept, socioeconomic status,and academic achievement. A secondary interest was thepredictability of academic achievement by the othervariables.This chapter focuses on a discussion of the results ofthis investigation and concludes with suggestions for futureresearch in Native education.Academic Achievement MeasureThe results from the Kruskal—Wallis Analyses revealedno difference between the academic achievement readingscores of the grades within the two groups.This lack of statistical significance may be anindication that the measure of academic achievement does not64discriminate between these grades or is a reflection ofteacher instruction prior to standardized testing.In addition, unique characteristics of the populationmay be contributing to these results. For example, it wasfound that reading levels among the grades 7 and 8 studentsranged from 3 to 8 for both grades. As a result, theICruskal—Wallis procedure may not be able to distinguishbetween the grades.Correlational Relationships among the VariablesThe Michigan State General Self—Concept of AbilityScale evaluated the students’ academic self—concept (ASC).The Piers—Harris Children’s Self—Concept Scaleinvestigated the students’ general self—concept (GSC).Hypothesis 1. The results showed that there was nostatistically significant difference between ASC and gradelevel.Hypothesis 2. The results also showed that there was nostatistically significant difference between GSC and gradelevel.The conclusion here is that due to the dropout rateamong these students and due to the fact that a number of65students leave the community each year to attend high schoolin larger centres, the scales were unable to differentiatebetween the self—concepts of the students who remained inthe school, although research has shown that self—concept isaltered gradually as a student moves through adolescence(Beane, Lipka, & Ludewig, 1980; Hoge, Smit, & Hanson, 1990;Marshall, 1989; Pepper & Henry, 1991).Hypothesis 3. The results of this investigation alsorevealed that there was no statistically significantrelationship between academic self—concept and academicachievement, nor between general self—concept and academicachievement.It appears that dropping out begins earlier than wasanticipated at the start of this study. Perhaps the academicachievement abilities of the students who remained in schoolin conjunction with their cultural differences from non—Natives, negates the appearance of differences between thesevariables. In addition, the students who have not droppedout have similar self—concepts in relation to their academicabilities.Hypothesis 4. This study showed a lack of statisticalsignificance between attendance and academic achievement.This lack of significance may be due to the small sample66size obtained for the grade 10 and 11 students, and may beindicative of students’ efforts to complete their academicwork even when absent from school for the grade 7 to 9students.Hypotheses 5 & 6. The finding of no statisticallysignificant correlations between socioeconomic status (SES)and academic achievement, nor between SES and both academicand general self—concept, is probably related to the methodof determining the SES of the students.The SES was actually accessed as the form of employmentor the source of income held by the parent/guardian of thestudent. Perhaps this view of SES (the community member) , isnot held by the students themselves, and therefore, does notaffect their academic achievement or their self—concepts(Chapman, Lambourne, & Silva, 1990).Hypothesis 7. The lack of finding any of the variablesto be predictors of academic achievement is not surprisingsince the correlational results revealed almost nostatistically significance.A research result not included within the hypotheses.need to be highlighted.The lack of a statistically significant correlation wasfound between academic self—concept and general self—concept67for the grade 10 to 11 group. Actually, a zero correlationwas found. This is contrary to the expectations of theliterature. Perhaps this result more than any otheremphasizes the need for a Native model of self—concept.There are several possible explanations for theresults found in this study.One reason may be the exposure of the students toNative role models. Most of the staff at the school areNative arid original residents of the community. Theirpresence may be providing the role models with which some ofthe students are identifying. As Parry (1982) points out, “Acohesive ethnic group identity, though it may be devalued bythe dominant culture, may be a major stabilizing factor inthe formation of a positive self—concept” (p. 21).A second reason may be that the students who haveremained in school have more stable self—concepts than thosewho have already dropped out.A third reason concerns the sample sizes for eachgrade. Sample sizes decreased as grade level increased. Thismay have limited or skewed the results of the self—conceptscales for this sample because of the probability of both ahigher ASC and GSC in the students who have remained inschool.68A fourth reason pertains to the difficulties inmeasuring psychological constructs. Brookover et al., (1967)defines academic self—concept as a behavior that can bemeasured through a self—report instrument. However, Combs(1984), cited in Wall (1991) , argues that self—concept isnot a behaviour. “He defined self—concept as a perceptualorganization which generated behaviour only as a symptom ofitself. His contention was that while self—report is abehaviour affected by self—concept, it could not be acceptedas being identical with it” (Wall, 1991, p. 46).A fifth reason concerns the nature of the self—reportinstrument. Results are affected by such things as (1) therelationship of the student and researcher; (2) thewillingness of the student to cooperate; (3) the student’scomprehension of the questions; and (4) the student’semotional state (Wall, 1991). Parry (1982) aptly sums:“An individual’s reporting of his assessment of his ownself—awareness may easily be faked, or may fluctuatedepending on conditions at the time of testing....In thefinal analysis one can only have what an individual iswilling to reveal about himself, which may or may not be howhe sees himself” (p. 13).69Predictors of Academic AchievementThe sample size in this study proved to be a factor inthe running of a stepwise multiple regression analysis. Theoriginal sample had to be divided into two groups due to theassessment of academic achievement.The choice of variables and the influence of unknownvariables contributed to the lack of statisticallysignificant results of the multiple regression.It may be that the variables chosen for this study,although supported by the literature, do not have the sameinfluence within a Native context. Perhaps there are othervariables influencing these relationships. For example,parental and/or teacher expectations as perceived by thestudents.Limitations of the StudyThe limitations of the study are as follows:The study used standardized tests with a population forwhich they have not been normed. The effect of thislimitation was controlled by comparing students in the studyonly with each other. Students were not compared to the norm70group, or with any other students outside of the population.It was not possible to randomly select students becauseof the small population and logistic considerations.Furthermore, the use of self—report data was limited to thewillingness of the subjects to participate.Generalization is limited since this study focused on aspecific school and a specific community.ConclusionThe results of this study have demonstrated that thereare few significant correlational relationships among thesevariables for this sample. It may be that there are othervariables operating in this forum that were not included inthis research. In addition, the accuracy or sensitivity ofthe instruments must be taken into consideration, since apart of the data was collected from existing records (i.e.attendance and reading scores).The suggestion to continue Native educational researchcannot be overstated.71RecommendationsThe following suggestions for future research withNative students are presented as a result of this study:1. A study is needed to determine if there is astatistically significant relationship between academicself—concept and academic achievement, using the results ofan entire standardized achievement test, standardizedreading scores, and teacher assigned grades, of Nativestudents. Consideration must be given to the sample sizes,i.e. n > 30, for each grade in such a study. Since there isno research providing support for a strong positiverelationship between these two variables within a Nativecontext, any meaning as to the significance of thesevariables in Native education is presently inapplicable.Madak (1988) aptly sums, “...without knowing the importanceof the relationship, one cannot make decisions as to howmuch time, energy, or money to invest in trying to improveself—concept with the hope that academic achievement willfollow” (p. 8)The importance of this relationship can be seen inconnection to the dropout rate among Native students. Asindicated in the literature review, a student’s self—concept72is not only related to his/her attendance, but becomesincreasing a factor in whether or not that student viewsschool as a place that is of value to him/her and worthinvesting time in.2. More evidence is needed to support the notion of theeffect of parental socioeconomic status on academicachievement and on self—concept among Native students.Previous research indicates that a relationship does existbetween these variables within non—Native cultures (Barnes &Vulcano, 1982; Chapman, Lambourne, & Silva 1990). Perhapsthe communal nature of Native culture negates theeffect of this variable within the academic achievement ofthe students. Further investigation into this communalnature would perhaps shed light on other facets of educationwithin the Native culture, now perceived as important withina non—Native context. An investigation of this type may alsoreveal aspects of the educational process that are unique toNative culture.4. Some of the variables involved in this research areaccumulative. Academic self—concept, general self—concept,and academic achievement fit into this category.Socioeconomic status is relatively constant and attendancefluctuates. Hence, it is suggested that teaching73effectiveness can be added to the list of factors which needto be investigated in future research.5. Another recommendation involves looking atattendance patterns. In this study, no consideration wasgiven to which days of the week students were absent. Asindicated in the literature review (see chapter two) , thispattern may have an effect on academic achievement aridshould be part of any future research in this area.6. A final recommendation concerns use of theattribution paradigm in further research. The suggestionhere would be to focus on the ‘locus of control’ forstudents in future studies.74REFERENCESAltmann, H. A., & Dupont, S. F. (1988). The relationshipsbetween academic self—concept, global self—concept, andacademic achievement. Canadian Journal of Counselling,22(3) , 170—174.Atherley, C. A. (1990). The effects of academic achievementand socioeconomic status upon the self—concept in themiddle years of school: A case study. EducationalResearch, 32, 224—229.Barnes, G. E., & Vulcano, B. A. (1982). School self—acceptance among Canadian Indian, White and Metischildren. Canadian Journal of Behavioural Science,14(1) , 60—70.Beane, J. A., Lipka, R. P., & Ludewig, J. W. (1980).Synthesis of research on self—concept. EducationalLeadership, 38, 84—89.Beane, J. A., & Lipka, R. P. (1980). Self—concept and self—esteem: A construct differentiation. Child StudyJournal, 10(1) , 1—6.Berger, A. (1973) . The education of Canadian Indians: An in—depth study of nine families. The Alberta Journal ofEducational Research, 19, 334—342.Bloom, B. S. (Ed.). (1985). Developing talent in youngpeople. New York: Ballantine Books.Boloz, S. A., & Varrati, R. (1983). Apologize or analyze:Measuring academic achievement in the reservationschool. Journal of American Indian Education, 23(1),23—28.75Brodbelt, S. (1985). Pbsenteeism: Critical problem inlearning. The Clearing House, 59, 64—68.Brookover, W. B., Paterson, A., & Thomas, S. (1962). Self—concept of ability and school achievement: Therelationship of self—images to achievement in juniorhigh school subjects (Cooperative Research Project No.845) . East Lansing: Michigan State University, Officeof Research and Publications.Brookover, W. B., Erickson, E. L., & Joiner, L. M. (1967).Self—concept of ability and school achievement 111:Relationship of self—concept to achievement in highschool (Educational Research Series No. 36). EastLansing, MI: Educational Publications Services.Butterfield, R. A. (1983). The development and use ofculturally appropriate curriculum for American Indianstudents. Peabody Journal of Education, 61, 49—66.Byrne, B. M. (1984). The general/academic self—conceptnomological network: A review of construct validationresearch. Review of Educational Research, 54, 427—456.Byrne, B. M., & Shavelson, R. J. (1987). Adolescent self—concept: Testing the assumption of equivalent structureacross gender. American Educational Research Journal,24(3) , 365—385.Canton, R. A. (1981). Self—concept research: The gapbetween social theory and educational practice.Canadian Journal of Education, 6, 77—91.Chapman, J. W., Cullen, J. L., Boersma, F. 3., & Maguire, T.0. (1981). Affective variables and school achievement:A study of possible causal influences. Canadian Journalof Behavioural Science, 13, 181—192.76Chapman, J. W., Lambourne, R., & Silva, P. A. (1990). Someantecedents of academic self—concept: A longitudinalstudy. British Journal of Educational Psychology, 60,142—152.Chavers, D. (1991). Indian education: Dealing with adisaster. Principal, 70, 28—29.Clifton, R. A., Perry, R. P., Parsonson, K., & Hryniuk, S.(1986). Effects of ethnicity and sex on teachers’expectations of junior high school students. Sociologyof Education, 59, 58—67.Dawson, J. (1988). “If my children are proud”: Nativeeducation and the problem of self—esteem. CanadianJournal of Native Education, 15, 43—50.Dunn, R. & Griggs, S. A. (1990). Research on the learningstyle characteristics of selected racial and ethnicgroups. Reading, Writing, and LearningDisabilities, 6, 223—247.Epstein, J. H. (1985). Review of Piers—Harris children’sself—concept scale (the way I feel about myself). InJ. V. Mitchell, Jr. (Ed.), The ninth mentalmeasurements yearbook (pp. 1167—1169). Highland Park,New Jersey: Gryphon Press.Frieze, I. H. (1976). Causal attributions and informationseeking to explain success and failure. Journal ofResearch in Personality, 10, 293—305.Gade, E., Huriburt, G., & Fuqua, D. (1986). Study habits andattitudes of American Indian students: Implications forcounselors. The School Counselor, 34, 135—139.Gardner, H. (1983). Frames of mind: The theory of multipleintelligences. New York: Basic Books.77Haarer, D. L. (1964). A comparative study of self—concept ofability: Delinquent and non—delinquent boys. In W. B.Brookover, E. L. Erickson, & L. M. Joiner (Eds.),Self—concept of ability and school achievement 111 (pp.231—246). East Lansing, MI: Educational PublicationsServices.Heaps, R. A., & Morrill, S. G. (1979). Comparing the self—concepts of Navajo and White high school students.Journal of American Indian Education, 18(3), 12—14.Hoge, D. R., Smit, E. K., & Hanson, S. L. (1990). Schoolexperiences predicting changes in self—esteem ofsixth— and seventh—grade students. Journal ofEducational Psychology, 82, 117—127.Hornett, D. M. (1990). Elementary—age tasks, culturalidentity, and the academic performance of youngAmerican Indian children. Action in Teacher Education,12, 43—49.House, 3. D. (1992). The relationship between academic self—concept, achievement—related expectancies, and collegeattrition. Journal of College Student Development, 33,5—10.Hunter, M., & Barker, G. (1987). “If at first...”:Attribution theory in the classroom. EducationalLeadership, 45(2). 50—53.Hurlburt, G., Henjum, R., & Eide, L. (1983). A comparison ofacademic, career, and social patterns of AmericanIndian students. Journal of American Indian Education,22, 17—22.Hurlburt, G., Krocker, R., & Gade, E. (1991). Studyorientation, persistence and retention of Nativestudents: Implications for confluent education. Journalof American Indian Education, 30, 16—23.78Jeske, P. J. (1985). Review of Piers—Harris children’sself—concept scale (the way I feel about myself). InJ. V. Mitchell, Jr. (Ed.), The ninth mentalmeasurements yearbook (pp. 1169—1170). Highland Park,New Jersey: Gryphon Press.Kirkness, V. J. (1981) . The education of Canadian Indianchildren. Child Welfare, 60, 447—455.Kruglanski, A. W. (1975). The endogenous—exogenous partitionin attribution theory. Psychological Review, 82, 387—406.Luftig, R. L. (1983). Effects of schooling on the self—concept of Native American students. The SchoolCounselor, 30, 251—260.Lyon, M. A., & MacDonald, N. T. (1990). Academic self—concept as a predictor of achievement for a sample ofelementary school students. Psychological Reports,66(3, Pt. 2) , 1135—1142.Madak, P. R. (1988). Native self—concept: A review of theliterature. Unpublished manuscript, University ofManitoba.Marshall, H. H. (1989). The development of self—concept.Young Children, 44, 44—51.Martin, J. D., & Coley, L. . (1984). Intercorrelations ofsome measures of self—concept. Educational andPsychological Measurement, 44(2), 517—521.Metcalfe, B. M. A. (1981). Self—concept and attitude toschool. British Journal of Educational Psychology, 51,66—76.79Monk, D. H., & Ibrahim, P4. A. (1984). Patterns of absenceand pupil achievement. American Educational ResearchJournal, 21(2), 295—310.Morse, R. (1963). Self—concept of ability and schoolachievement: A comparative study of negro and caucasianstudents. In W. B. Brookover, E. L. Erickson, & L. M.Joiner (Eds.), Self—concept of ability and schoolachievement 111 (pp. 205—210). East Lansing, MI:Educational Publications Services.National Indian Brotherhood. (1972). Indian Control ofIndian Education. Ottawa: National Indian Brotherhood.Parry, R. (1982). Poor self—concept and differentialacademic achievement: An inadequate explanation ofschool performance of Black and Native Americanchildren. Canadian Journal of Native Education, 10,11—24.Paterson, A. (1966). Reliability and validity of self—concept of ability scale. In W. B. Brookover, E. L.Erickson, & L. M. Joiner (Eds.), Self—concept ofability and school achievement 111 (pp. 155—172). EastLansing, MI: Educational Publications Services.Pauls, S. (1984). The case for band controlled schools.Canadian Journal of Native Education, 12, 31—37.Pepper, F. C., & Henry, S. L. (1991). An Indian perspectiveof self—esteem. Canadian Journal of Native Education,18(2) , 145—160.Persaud, D., & Madak, P. R. (1992). Graduates and dropouts:Comparing perceptions of self, family, and schoolsupports. The Alberta Journal of Educational Research,38(3) , 235—250.80Piers, E. V., & Harris, D. B. (1969). Piers—HarrisChildren’s Self—Concept Scale. Los Angeles: WesternPsychological Services.Piers, E. V. (1984). Piers—Harris Children’s Self—ConceptScale, Revised Manual. Los Angeles: WesternPsychological Services.Rampaul, W. E., Singh, M., & Didyk, J. (1984). Therelationship between academic achievement, self—concept, creativity, and teacher expectationsamong Native children in a northern Manitoba school.The Alberta Journal of Educational Research, 30,213—225.Raviv, A., Bar—Ta1, D., Raviv, A., & Levit, R. (1983).Students’ reactions to attributions of ability andeffort. British Journal of Educational Psychology, 53,1—13.Reid, K. (1982). The self—concept and persistent schoolabsenteeism. British Journal of EducationalPsychology, 52, 179—187.Reyhner, J. (1992). American Indians out of school: A reviewof school—based causes and solutions. Journal ofAmerican Indian Education, 31, 37—56.Richardson D. T., & Richardson, Z. A. C. (1986). Changesand parental involvement in Indian education. CanadianJournal of Native Education, 13(3), 21—25.Riffel, J. A. (1991) . Academic self—concept and studentachievement: Some implications for Native education.The Canadian School Executive, 10, 26—27.81Scheirer, M. A., & Kraut, R. E. (1979). Increasingeducational achievement via self concept change. Reviewof Educational Research, 49(1), 131—150.Shavelson, R. J., Hubner, J. J., & Stanton, G. C. (1976).Self—concept: Validation of construct interpretation.Review of Educational Research, 46, 407—441.Wall, C. S. (1991). Indian students’ academic self—conceptin a selected band—controlled school and a provincialschool. Published master’s thesis, University ofManitoba, Winnipeg, Manitoba.Wall, C. S., & Madak, P. R. (1991). Indian students’academic self—concept and their perceptions of teacherand parent aspirations for them in a band—controlledschool and a provincial school. Canadian Journal ofNative Education, 18(1), 43—51.Ward, M. S. (1986). Indian education: Policy and politics,1972—1982. Canadian Journal of Native Education, 13,10—21.Weiner, B. (1972). Attribution theory, achievementmotivation, and the educational process. Review ofEducational Research, 42, 203—215.Weiner, B., & Peter, N. A. (1973). A cognitive—developmentalanalysis of achievement and moral judgments.Developmental Psychology, 9(3), 290—309.Zarb, J. M. (1981). Non—academic predictors of successfulacademic achievement in a normal adolescent sample.Adolescence, 16, 891—900.82Appendix 1SELF-CONCEPT OF ABILITY--GENERAL(FORM A)Michigan State UniversityBureau of Educational ResearchCircle the letter in front of the statement which bestanswers each question.1. How do you rate yourself in school ability compared withyour close friends?a. I am the bestb. I am above averagec. I am averaged. I am below averagee. I am the poorest2. How do you rate yourself in school ability compared withthose in your class at school?a. I am among the bestb. I am above averageC. I am averaged. I am below averagee. I am among the poorest3. Where do you think you would rank in your class in highschool?a. among the bestb. above averagec. averaged. below averagee. among the poorest834. Do you think you have the ability to completeuniver sity?a. yes, definitelyb yes, probablyc. not sure either wayd. probably note. no5. Where do you think you would rank in your class inuniversity?a. among the bestb. above averagec. averaged. below averagee. among the poorest6. In order to become a doctor, lawyer, or universityprofessor, work beyond four years of university isnecessary. How likely do you thik it is that you couldcomplete such advanced work?a. very likelyb. somewhat likelyc. not sure either wayd. unlikelye. most unlikely847. Forget for a moment how others grade your work. In yourown opinion how good do you think your work is?a. my work is excellentb. my work is goodc. my work is averaged. my work is below averagee. my work is much below average8 What kind of grades do you think you are capable ofgetting.a. mostly A’sb. mostly B’sc. mostly C’sd. mostly D’se. mostly F’s85


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items