Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Harvey’s belief systems in adolescents : problems of response categorization Litzcke, Hans Georg 1978

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1978_A47 L58.pdf [ 5.27MB ]
Metadata
JSON: 831-1.0094471.json
JSON-LD: 831-1.0094471-ld.json
RDF/XML (Pretty): 831-1.0094471-rdf.xml
RDF/JSON: 831-1.0094471-rdf.json
Turtle: 831-1.0094471-turtle.txt
N-Triples: 831-1.0094471-rdf-ntriples.txt
Original Record: 831-1.0094471-source.json
Full Text
831-1.0094471-fulltext.txt
Citation
831-1.0094471.ris

Full Text

HARVEY'S BELIEF SYSTEMS IN ADOLESCENTS: Problems of Response Categorization by HANS-GEORG LITZCKE B.Ed., University of British Columbia, 1969 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF EDUCATION in ; THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF EDUCATIONAL PSYCHOLOGY UNIVERSITY OF BRITISH COLUMBIA We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1978 Hans-Georg Litzcke, 1978 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r ag ree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d tha t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department o f Educational Psychology The U n i v e r s i t y o f B r i t i s h Co lumbia 2075 Wesbrook P l a c e Vancouver, Canada V6T 1W5 Date September 12th, 1978 ABSTRACT In 1961 O. J. Harvey, David E. Hunt, and Harold M. Schroder pub-lished a theory of personality structure and development. Major research based on that theory has continued under 0. J. Harvey at the University of Colorado until now with the result that the theory underwent some reforma-tions. Several aspects, especially those dealing with descriptive indexes of personality and the assumed implications of functioning seem largely confirmed through research. However, most reported measurements were done with a "free response" measure on samples from University students. Furthermore, as far as this writer was able to ascertain, no independent evaluation of the effectiveness of the instrument has been reported. For the purpose of attempting such an evaluation, Harvey's "free response measure" (with a slight modification) was coupled with a "forced choice" measure that was theoretically assumed to measure the same variables. Both parts of the instrument, presented in a booklet form, were administered to a non-systematic sample of grade 9 and 10 high school students. Subsequently, results obtained through judgemental categorizations were carefully recorded. Scoring procedures for the "forced choice" measure were based on probability considerations in such a way that the meaning of and confidence in scores would correspond as closely as seemed possible to the meaning of and confidence in judged categorizations. Surprisingly different pictures emerged from comparisons of the results. The "free response" measure, while found very demanding on the level of judgemental expertise and, hence, of questionable r e l i a b i l i t y unless in the hands of rather "finely" tuned judges, was found, nonetheless, to be an instrument of remarkable sensitivity in i t s a b i l i t y to point up inferential positions from the theoretical structures of personality. For this reason, the results from the free response measure were used as a provisionally valid baseline from which the results of the forced choice measure were subsequently interpreted. The results from the "forced choice" measure seemed at f i r s t sight completely unrelated to the "free response" measure, but closer inspection showed a pattern which made the failings of the instrument understandable and lead to directions that would have to be taken to bring the functioning of the "forced choice" measure into closer correspondence with the function-ing of the "free response" measure. With the obtained frequencies of categorizations of the "free response" measure, one "external" and several "internal" comparisons of frequencies of belief system functionings were made. In the "external" comparison of this sample to reported frequencies by Harvey, differences were found - but not a l l of them as expected. It was suggested that these differences might eventually disappear when the chronological age of the respondents would equal that of the respondents in Harvey's samples. Research into distribution pattern of system functioning across different cultures with different patterns of childrearing would shed light on problems of generalizability and developmental aspects of the theory. Internal comparisons showed that frequencies of personality structures (as defined by belief system functioning) were essentially independent of sex and grade level. Contrary to Harvey's indication, levels of "articulateness" were found to be much higher in certain personality structures than in others. It was also shown that "norms" of prejudice seen as varying over time at different rates for different concepts, occurred in their entire range (no prejudice - strong prejudice) only in two major types of personality functioning. It was argued that, i f , as i t seems, prejudice' does not occur in certain personality structures, then the whole issue of tolerance, etc. is in the entire domain of discussion between the two remaining personality types. In conclusion, the adequacy of the "free response" instrument was assessed in terms of i t ' s possible purposes. As a research and diagnostic tool i t received "high marks", but as a means to separate students into different levels and as a means to monitor their progress, i t received hardly any "marks" at a l l . It was argued that by taking bits and pieces from the theory and, armed with a very problematic measurement instrument, one would rush headlong into yet another episode of charlatanism. Several research questions were posited, the most important ones being questions about generalizability and developmental processes of system functioning. Along this way efforts should also be continued to develop an "objective" standardizable instrument preferably as a direct rather than an indirect measure of system functioning. The theory of belief system functioning was viewed as a valuable topic in teacher education. TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS v LIST OF TABLES . . . . v i i LIST OF FIGURES v i i i ACKNOWLEDGEMENTS ix I. INTRODUCTION 1 Nature and Purpose of Study A Brief Descriptive Outline of Some Important Aspects of Harvey's Work Description of Harvey's belief systems — Distribution of the systems from samples of defined populations — Description of the tests used by Harvey for the purpose of assigning people to the belief systems — A brief summary of some of the construct validations reported by Harvey. II. MODIFICATION OF "THIS I BELIEVE" QUESTIONNAIRE 14 The Rationale for the Modification Presentation of the Modified Instrument III. OBSERVATIONAL RESULTS FROM SAMPLE 25 Response Categorizations from the "Free Response" Section of the Modified Instrument Judgemental procedure (see also appendix A) — Key used for categorizations — Display of examples of response categorizations — Display of frequency distributions of complete sample and of various analyzed sub-groups of sample. vi Response Categorizations from the "Forced Choice" Section Key used — Display of frequencies within each of the 14 items — Display of a comparison of frequencies obtained for the total sample by both the "Free Response" and "Forced Choice" sections. IV. DISCUSSIONS 43 Examinations of Examples of "Free Response" Patterns as Indicators of System Functioning; Success and Failures External and Internal Group Comparisons of the Sample Under Study V. CONCLUSION: INTERPRETATIONS, AND POSSIBLE DIRECTIONS OF FURTHER RESEARCH 63 Questions of Adequacy of the "Free Response" Measure Implications from Group Comparisons WORKS CITED 74 APPENDIX A 76 Judgemental Strategy and Rationale for Strategy, Display of Two Agreement Indexes and Summary of Agreement Indexes in Percentage Forms APPENDIX B 82 A Complete Quotation of "System Response Patterns" From 0. J. Harvey's 'Scoring the "This I Believe" (TIB) Test', pp. 9-12 v i i LIST OF TABLES 1. Structural Main and Subvariables of Harvey's Belief Systems . . . . 5 2. Some Important Content Variables of Harvey's Belief Systems . . . . 6 3. Percent Distributions of Samples of Defined Populations 9 4. Dogmatism & Authoritarianism as Approximate Measure of Belief Systems 13 5. Key Used for Classifying Belief Systems from the "Free Response" Section (Part B) 25 6. Key Used for Rank-Ordering "Articulateness" from the "Free Response" Section (Part B) 26 7. Keys Used for Classifying Two Additional Responses from the "Free Response" Section (Part B) . 27 8. Frequency Distributions of Belief Systems from "Free Responses" of Total Sample (N = 122) and Various Subgroups of Sample 37 9. Key for System and Admixture Designations for the 14 Item Forced Choice Section _ 39 10. Frequencies of System Type Responses for Each of the 14 "Forced Choice" Items from Total Sample (N=122) . . . . . . .40 11. Number and Extent of Agreements on Scores from "Free Responses" and from "Forced Choices" from Total Sample (N=122) 42 12. Two Indexes of System Response Type Attraction for 14 "Forced Choice" Items 50 13. Relations of Indexes (I's) of System Response Type Attraction for 14 "Forced Choice" Items 51 14. Belief System Distribution of Sample as Estimated by the "Free Response" Measure Compared to the Estimated Distributions If It Had Been Drawn from a Population of Liberal Arts Students 55 15. Chi-Square Tests for Differences in Distributions of Belief Systems 56 16. Levels of "Articulateness" in Belief Systems and Admixtures . . . . 57 17. Levels of "Articulateness" in Belief Systems Combined With Dominant Admixtures 58 18. Test for Independence Between "Judgemental Confidence" and Levels of "Articulateness" 60 19. Sex Distributions in Belief Systems Combined with Dominant Admixtures 61 20. Grade Distributions in Belief Systems Combined With Dominant Admixtures 61 21. Number and Extent of Agreements on Categorizations of "Free Response" Between Judge I and Judge II." Total Sample N=122 78 22. Number and Extent of Agreements on Scores From "Free Responses" Between 1st Panel Adjustments and 2nd Panel Adjustments. Total Sample N=122 79 23. Indexes of Judgemental Agreements 81 v i i i LIST OF FIGURES 1. Frequency Distributions of Belief Systems and Dominant Admixtures as "Seen" from "Free Responses" and from 14 "Forced Choice" I terns 14 ACKNOWLEDGEMENTS I wish to thank Dr. Stephen F. Foster, Dr. J. Gordon Nelson, and Dr. Harold Ratzlaff for agreeing to serve on my committee, for reviewing the proposal for this thesis, and for much helpful advice. I am much indebted to Dr. H. Ratzlaff for reviewing s t a t i s t i c a l material in this paper. My special thanks go to Dr. S. F. Foster for his supervision, encouragement, time, and patience. My thanks are extended also to Professor 0. J. Harvey of the University of Colorado for giving me his kind permission to use and modify his test format (TIB-74) Copyright 1974. I would also like to thank Dr. Carolie J. Coates for presenting 0. J. Harvey's work with much enthusiasm in the summer of 1974 at this University. 1 I. INTRODUCTION Nature and Purpose of Study The primary purpose of this observational study is to evaluate the problems of measuring "belief systems" in adolescents. The study is based on Harvey's model of personality organization which allows inferences of differences in structures of personality from analyses of expressed beliefs. Specified differences in expressed beliefs as well as their assumed implica-tions are interchangeably referred to as "conceptual systems" or "belief systems". Theoretical and reported empirical investigations of the model w i l l be outlined in some detail in the section to follow. Problems of response categorizations w i l l be analyzed from responses to a two-part modified version of Harvey's measurement instrument which was administered to 122 Grade 9 and 10 high school students. One section of the modified instru-ment is a "free response" measure, the other, a "forced choice" measure. In the categorizations from the "free responses", judgemental d i f f i c u l t i e s w i l l be examined and the obtained categorizations from the "free response" measure w i l l then be compared to those from the "forced choice" section. Secondly, the paper w i l l compare the best attainable categorizations from the sample with Harvey's published frequencies of categorizations from college groups. This secondary purpose amounts to a goodness-of-fit test and invites some a p r i o r i hypothesizing that can be expressed as follows: inasmuch as Harvey's model describes developmental stages (i.e. belief system 1 is in some sense assumed to be a prior developmental stage to belief system 4), and inasmuch as developmental stages may be assumed to 2 be in greater flux i n grade 9's and grade 10's than in college freshmen, i t is a reasonable expectation that the comparison should show a significantly greater frequency of system l's and admixtures in this sample of high school students than the reported frequencies from a freshman population. Further-more, the expected results would tend to be pushed into the same directions from contaminations of f i c t i t i o u s l y higher categorizations of system l's and admixtures caused by judgemental misreadings of responses relatively lower in articulateness than those of college students. Thirdly, the study w i l l examine internal comparisons of subgroups of the sample at hand. As a corollary to the reasons advanced in the previous paragraph, one may reasonably expect a relatively greater frequency of system l's and admixtures in lower levels of articulateness than at higher levels. One may also reasonably expect a directionally similar difference in comparisons of grade 9's with grade 10's. Concerning other internal comparisons such as between response patterns of boys and g i r l s , neither theoretical nor common sense assumptions can easily be advanced for any further useful a p r i o r i statements. It must be noted at this time that the study sample is a non-systematic sample drawn from one school; i.e. , not a random sample of grade 9 and 10 students. For this reason, any generalizations from the observed outcomes to other populations or to the same population can be done only speculatively. In the section to follow, the reader w i l l find an overview of some aspects of Harvey's work that are considered essential to the understanding of the concerns of this paper. 3 A Brief Descriptive Outline of Some Important  Aspects of Harvey's Work 0. J. Harvey, David E. Hunt, and Harold M. Schroder have posited four major conceptual' systems to which people may be assigned in accordance with an analysis of their expressed beliefs. These beliefs have been characterized by Harvey as "deeply held attitudes or values".^ In contrast to recent efforts in attitudinal research to attach very specific meanings to concepts such as "attitudes" and "values", Harvey chooses to use these terms rather loosely - a choice probably indicative of the lack of an English equivalent for the German "Anschauung" which denotes a person's relatively consistent norm of thinking, feeling, and behaving towards many different objects and situations as long as these are perceived by the 2 person -as important to himself. More recently, Harvey has theorized that the belief systems w i l l function (i.e., express and behave) at their most consistent patterns at some optimal level of personal involvement. In spite of some clarifications through research i t is not quite clear what 3 this optimal level i s . Hence, for the time being, i t may be simpler to think of belief systems in terms of one of Harvey's earlier definitions: 0. J. Harvey, "Beliefs and Behavior: Some Implications for Education", The Science Teacher, 37 (December 1970) : 10. 2 0. J. Harvey, David E. Hunt, and Harold M. Schroder, Conceptual  Systems and Personality Organization (New York: John Wiley & Sons, Inc., 1961), p. 2. In this early work the authors explain their preference for the term "Anschauung" over "attitude". 3 One attempt to cl a r i f y the assumption for an optimum level of "ego - involvement" can be found in the study by Alma Grabow Miller and 0. J. Harvey, "Effects of Concreteness - Abstractness and Anxiety on Intellectual and Motor Performance", Journal of Consulting & C l i n i c a l  Psychology, 40 (1973): 444-451. 4 "A belief system represents a set of predispositions to perceive, feel toward, and respond to ego-involving stimuli and events in a consistent way. " The theoretical principles underlying Harvey's belief systems are derivatives from several generalizable theories. Somewhat simplistically one may visualize these theories as convergent onto a kind of life-force principle that seeks to maintain the integrity of any organism or system. It i s from this principle and from empirical observations that the belief systems have been construed and delineated as different sets of "definitional matrices" that people appear to adopt for interpreting the world around them and for maintaining and defending the "self".^ Harvey has also theorized that the kind of belief system that a person comes to represent results from different styles of upbringing.^ More immediately important than theoretical positions is the extent to which the systems of the model can be clearly and consistently differentiated. To this end one must examine the defining variables of the different systems with some care. The postulated belief systems have been defined in terms of (a) "structure" and (b) "content". By "structure" Harvey means the 4 0. J. Harvey, "Belief Systems and Education: Some Implications for Change", in The Affective Domain, ed. Jack Crawford (Wash., D.C.: Communication Service Corporation, 1970), p. 68. ^Harvey's theoretrical positions are outlined br i e f l y in O. J. Harvey "Conceptual Systems and Attitude Change", in Attitude, Ego-Involvement and  Change, ed. Carolyn W. Sherif and Muzafer Sherif (New York: John Wiley & Sons, Inc., 1967), pp. 201-205. This theory is explored in O.. J. Harvey and Catherine Felknor, "Parent - Child Relations as an Antecedent to Conceptual Functioning", in Early Experience and the Processes of Socialization, eds. G. A. Milton, R. A. Hoppe, and E. C. Simmel (New York: Academic Press, 1970), pp. 167-203. 5 organizational patterns of beliefs f a l l i n g into hierarchies that parallel only approximately the main dimension referred to as "concreteness-abstract-ness". The more concretely functioning person would tend to structure (express) his beliefs in terms of "absolute" rather than " r e l a t i v i s t i c " ; "sophisticated" (accepting readily ideals of the predominant culture) rather than "open", and other similar structural dimensions. The non-parallel patterns of these subvariables seem to arise from some qualitative difference in functioning of the systems which is not entirely controlled by the main dimension of "concreteness-abstractness". The pattern that would arise from Harvey's definitions in his scoring booklet for the "This 7 I.Believe" (TIB) test can be shown in table form. TABLE 1 STRUCTURAL MAIN AND SUBVARIABLES OF HARVEY'S BELIEF SYSTEMS Systems Structural - Dimension 1 2 3 4 concrete , . . .main . - abstract ( . . ) dimension a b c d absolutism - relativism a,b a,b c d cliche - originality a,b ,c a,b, c a,b,c d naivete - awareness a,b c a,b d cynicism - criticalness x a X d sophistication - openness a,b c a,b d defensiveness - co-operation b a c,d c,d SOURCE : 0. J. Harvey, 'Scoring the "This I Believe" (TIB) Test 1. Compiled from material on pp. 6-8. The following exposition relies heavily on 0. J. Harvey's 'Scoring the "This I Believe".(TIB) Test', manuscript on f i l e , 0. J. Harvey, University of Colorado, 1965. (Mimeographed.) 6 The rank order (a), (b), (c), (d) indicates how closely the list e d system functions with regard to the f i r s t term of the subvariable. That i s , i f a system is designated (a) i t functions most with the f i r s t term of the subvariable, i f i t is designated (d) i t functions least with the f i r s t term and, conversely, functions most with the last term. If two designations are given, the rank order position i s shared with another system; i f three designations are given, the rank order position is shared with three systems. An (x) means that Harvey has indicated no functioning at a l l along that list e d dimension. Superimposed within this "structure" of the belief systems is the "content", meaning any and a l l items about which a person has any strong ("egc—involving") feelings. Although Harvey does not e x p l i c i t l y say so, i t seems clear that the "content" dimension (presumably a discrete variable) i s never completely independent of the structural dimension (presumably a continuous variable). Harvey also indicates a different "content" focus for the systems. For this reason, i t seems justifiable to l i s t separately variables that are mainly "content" defined. TABLE 2 SOME IMPORTANT CONTENT VARIABLES OF HARVEY'S BELIEF SYSTEMS Systems "Content" Dimension 1 2 3 4 external causality internal causality a ,b (God) c,d a,b c,d (people) need for friends high — low a,b ,c a,b ,c a,b,c d rule orientation in problem solving . high low a,b a,b c,d c,d SOURCE: 0. J. Compiled from material Harvey, 'Scoring the on pp. 7-9. "This I Believe " (TIB) Test'. 7 The qualitative difference in the variables that are mainly "content" defined make any rank ordering almost impossible. The "need for friends" variable providing one of the most important guide posts for identification of the systems, especially system 3, may possibly serve as the best il l u s t r a t i o n . Harvey uses the following descriptive definition: "System 1, 2, and 3 persons tend to place a high value on friends. System l's typically say that the most important thing in l i f e i s to have a few good friends. System 2's value friendship highly, but generally indicate that such relationships are d i f f i c u l t to develop, are unstable affairs, and hurt people when they dissolve. System 3 persons tend to say that everyone needs lots of friends - that such relationship help one to grow and understand oneself. System 4's welcome friendship but do not have a strong need to surround them-selves with friends."^ Harvey's four belief systems w i l l now be presented in a much abbreviated form. System 1: Along the structural dimensions, system 1 persons are found at the lower end of the hierarchies. They tend to express their beliefs in simplistic terms, in either - or propositions, in clear-cut avoidance of uncertainty and acceptance of norms that should, ought, or must be followed. Their most prominent belief content focuses on religion, authority, tradition, people with influence and power, to a l l of which they express highly positive attitudes. System 1 people tend to be rule oriented and seem to have d i f f i c u l t i e s to admit another point of view. System 2: In a structural sense they tend to be very close to system 1 people. They share to only a slightly lesser degree the absolutism and evaluative organization of beliefs. System 2 tend to focus their belief on the same content as system 1 people. However, they tend to hold strongly negative attitudes to items to which system 1 people would tend to be highly positive. System 2 people distrust authority, Ibid., p. 8. 8 feel alienated and tend to have low self-esteem, but w i l l generally express sympathy towards the underdog and lone individual. System 2 people are the most cynical and defensive of a l l the postulated systems. System 3 people i n their structural organization tend to be more complex and open than either systems 1 or 2. In terms of content their focus is on people, friendship, neighborliness, harmony, and helpfulness. They seem to be almost totally animated by personal relationships, forming often most complex dependency patterns. System 3 people w i l l be highly positive to situations and plans that are perceived as beneficial to other people. Conversely, the only negative attitudes generally expressed are to aspects that may harm people. System 4 people occupy the uppermost end in the structural organiza-tion of beliefs. How definitely they seem to be there is somewhat graphically illustrated by the designations just presented in tabulated form from the definitions of Harvey. No other system has such a clear rank order position on a majority of defined dimensions. Their relatively highly abstract functioning may find expression in tolerance of inconsistencies and contradictions, and in some apparent paradoxes. System 4 people seem to have fewer pre-focused "content" items than any 9 of the other systems. Distribution of Systems in Samples of Defined Populations Distribution patterns found by 0. J. Harvey seem to suggest two trends. Ibid., pp. 9-12. In view of the importance for correct scoring procedures the reader is provided with a f u l l quotation of Harvey's characterization of "System Response Patterns" in Appendix B of this paper. 9 (A) The educational institutions (as defined below) tend to favour the recruitment of system 1 persons and to a lesser extent system 3 persons; and, conversely, disfavours system 2 people and to a much lesser extent, system 4 people. (B) There is an apparent acceleration of the selectivity trend (with the exception of system 3) as people advance to more senior positions in educational institutions. These two effects are tabulated below. TABLE 3 PERCENT DISTRIBUTIONS OF SAMPLES OF DEFINED POPULATIONS Percent distributions of samples of defined populations System 1 35% 45% 55% 75% 90% System 2 15% 5% NIL NIL NIL System 3 20% 25% 15% NIL NIL System 4 7% 5% 4% NIL NIL Sample: Sample: Sample: Sample: Sample: Several 1000 Undergrad. Practice Principals Superintendents Liberal Arts Ed. Majors, Te ache rs, of Colorado, of Colorado, Students, University of University Wyoming, Wyoming, Utah, University of Colorado of Colorado Utah, New New Mexico Colorado Mexico SOURCE: O. J. Harvey, "Beliefs and Behavior: Some Implications for Education", The Science Teacher,37 (December 1970): 13. A l l definitions of samples exactly as given by 0. J. Harvey. Note: Total percentages do not add up to 100 because some respondents can not be assigned to any of the major systems. Description of the Two Tests Used by 0. J. Harvey for the Purpose of Assigning People to the Belief Systems Harvey's most important test for the purpose of assigning people to one of the belief systems is the previously referred to sentence completion 10. test called "This I Believe" ( T I B ) . I t contains nine such items as, "This I Believe about lying". The subject must write at least two sentences for each item in the sequence of appearance in the (TIB) booklet. Each item must be completed within two minutes. The Test is scored by trained judges. A system designation is assigned to the respondent even i f one response is discrepant or f a i l s to discriminate. If there is enough variety in the responses the most predominant system is l i s t e d f i r s t , the next second. For example, 2_, 4. (Unfortunately, one does not know precisely what constitutes "enough".) Harvey says that rarely w i l l one require more than two designa-tions. Harvey reports an interjudge r e l i a b i l i t y for three or four judges as .9 or higher. Harvey also reports a high .80's test-retest r e l i a b i l i t y for one week and also for six month intervals."'""'" The second test obviates the use of judges. It is called "Con-ceptual System Test" (Form CST-71). This instrument relies mainly on the qualitative differentiation of the "content" variables and other variables from construct validations. For example item 35 reads: "Guilt results from violation of God's law." A high agreement score on this item would place the respondent towards system 1. Another example, item 26, reads: 12 "I enjoy making sacrifices for the sake of the happiness of others." i U 0 . J. Harvey, This I Believe Test (Form TIB-74), Copyright 1974, 0. J. Harvey. "^Harvey, "Belief Systems and Education", p. 73, and 0. J. Harvey, ed., "System Structure, F l e x i b i l i t y and Creativity", in Experience, Structure  and Adaptability (New York, Springer Publishing Company, 1966), p. 47. It i s assumed that these figures mean simple ratios of agreements to total responses. 0. J. Harvey and James K. Hoffmeister, Conceptual Systems Test Form CST-71, Copyright 1971, 0. J. Harvey and K. Hoffmeister. The example given derives from a copy of this 48 item test. 11 For this test six different factors have been analysed: supernaturalism or divine fate control, need for structure and order, need to help people, need 13 for people, interpersonal aggression, and general distrust. This test when used for system classification needs pre-established cut-off values for the mean scores on the factors. The actual classification is based on "a great deal of empirical f i t t i n g " . Harvey reports that the CST is as good for system 1, 2 and 3, as the TIB, but not as good for system 14 4 identification. The following w i l l give a selective summarization of a few of the assumed construct validations for the belief systems. Intelligence measures (WAIS) from three different samples were found, as expected, not to be a defining variable for the belief systems. Harvey reports that only on sub-tests measuring verbal intelligence and vocabulary did system 2's and 4's score higher than the other systems. Of these two, system 4's were not significantly higher than system 2's. Harvey notes, however, that in "actual vocabulary usage" both in TIB answers and i n the role playing a l l systems were equal to each other. Characterizing "creative" responses as essentially both novel and appropriate, Harvey concludes from two studies mentioned that "creativity" tends to be a defining variable along the lines of theoretical expectations; i.e., system 4's are much more "creative" than system l ' s . One study would 13 Harvey, "Belief Systems and Education", pp. 73-74. 14 Ibid., p. 74. Harvey, "System Structure, F l e x i b i l i t y and Creativity", pp. 48-64. With one noted exception the above reference i s the source for a l l the mentioned assumed validations. seem to support a systems 4 - 2 - 3 - 1 rank order from most to least "creative". The other study, however, puts system 3's closer to system 4's 16 than to system I s . Religious involvement appears to follow quite closely along the assumed directions with an overall systems 1 - 3 - 4 - 2 rank order from most to least "religious". Patterns of church attendance show the follow-ing trends: system l's attend church regularly - systems 3's occasionally - system 4's rarely - system 2's very rarely. This participation pattern tends to be paralleled in expressed beliefs about religion. System l's show the highest proportion of beliefs expressing that there is only one true religion and that everyone needs to worship. System 4's have a low involvement, while system 2's in addition to low involvement tend to express negative attitudes to religion. Surprisingly though, in a l l systems most respondents characterize themselves as "not very religious". Using two scales, one assuming to measure "authoritarianism" (identified as the "F-scale") the other, "dogmatism" (identified as "Rokeach's Dogmatism Scale"), Harvey reports a " f a i r l y " accurate means of actually assigning respondents to the belief systems according to positions in a 2 x 2 contingency table. However, in another place Harvey has indicated an apparently more accurate way of categorizing respondents by using the same measures for a 3 x 3 contingency table as shown below. Harvey, "System Structure, F l e x i b i l i t y and Creativity", p. 56-58. The relationships, however, to the construct are not as clear-cut as one might wish or as the reader might be led to assume. Harvey reports rank order of relative percentages from different systems. In other words, there were some system l's that scored high in the c e l l of high novelty and high appropriateness. 13 TABLE 4 DOGMATISM & AUTHORITARIANISM AS APPROXIMATE MEASURES OF BELIEF SYSTEMS DOGMATISM a H CO <: E-t a o a E h Lowe r Middle Higher Third Third Third Lower Third 4 2 Middle Third 3 Higher Third 1 SOURCE: Harvey, "Conceptual Systems and Attitude Change", in Attitude, Ego - Involvement and Change, eds. Carolyn W. Sherif and Muzafer Sherif (New York: John Wiley & Sons, Inc., 1967), p. 210. Harvey summarizes a number of other measures that are assumed to have strengthened the construct validity of the belief systems. Some of these measures (like the mentioned "creativity" scores) are not paper and pencil tests but observed differences of behavior in several performance tasks which Harvey discusses under the headings of "Cue Utilization", "Change of Set", and "Admission of Deviant Inputs into the Systems". 14 II. MODIFICATION OF "THIS I BELIEVE" QUESTIONNAIRE The Rationale for the Modification Rather large claims have been made for the implications of Harvey's model to problems in education (a point to which this paper w i l l return). In view of these claims a reliable and valid measurement instrument is an essential requirement. From the foregoing introduction i t w i l l have become apparent that the belief systems are a construct of social-psychological interaction. For such a model to be useful i t must remain sensitive to cultural and social differences not only in a particular sense but also in a generalizable one. For example, i t does not require much acumen to see that a belief item about the "American Flag" w i l l not likely be an "ego-involving" item for a Canadian high school student. It is rather more d i f f i c u l t , however, to foresee whether a substitution to "Canadian Flag" would be adequate. A very likely possibility i s that there are generalizable differences in cultural "imprints" from symbols. In this instance a good guess is that the symbol "flag" means much less to a Canadian than to an American. Unfortunately, without empirical evidence one cannot know how good one's guess i s . Nonetheless, just as with any other test which is to be used in a new situation, one can improve the face validity and content validity of this instrument. The "free response" (Part B) of the modified instrument reflects an attempt to make this improvement by a kind of cultural judgement. In view of the propinquity of the American and Canadian social-cultural setting the modification is a small one. One further change in Part B is the 15 reduction of items to six from nine in the original TIB. The reason for this step derives from Harvey's direction that the scoring w i l l be most reliably done when evaluating the "global" response pattern rather than each item separately. If this is indeed true, then this reduction may be a justifiable economy without a great penalty ( i f any) in the form of a reduced r e l i a b i l i t y . The inclusion of Part A, which constitutes a much more extreme change from the TIB than Part B just discussed, was prompted by two con-siderations. The foremost of these was a concern for practicality. If indeed Harvey's model is useful for education then one needs not only a reliable and valid way of assigning subjects to the belief systems but also one that is easily and quickly scored. The second consideration came from the reading of the construct validation summarized in the pre-ceding section. It w i l l be recalled that system 2's and 4's were found to perform better on subtests in verbal intelligence and vocabulary than the other systems. If intelligence is not a defining variable in any sense for the systems, then confounding influences from better verbal a b i l i t i e s (in spite of the assurance from Harvey that respondents of neither system show a general verbal superiority in the TIB answers) were seen as a possible threat to a reasonable confidence level in classifying the responses. To the extent that one can draw an analogy from the principle that "recognition is easier than r e c a l l " , i t was reasoned that i f respondents could be made to "recognize" rather than to "verbalize" their "correct" response, one would eliminate verbal superiority as a possible confounding influence. At this point one may ask, why not use the Conceptual System Test which has been described previously. The reader w i l l recall that the CST was empirically fit t e d to correspond as closely as possible to the TIB. It is 16 also reported as being not sensitive to system 4 classification. Using the CST as an alternate form would require a new empirical f i t every time the TIB calls for modification. Without doubt, any scale that is related to any other scale in any systematic way can be f i t t e d as a valid substitute measure. But i t was thought possible to construct a scale that would measure the belief systems more directly than the circumspectual approach of the CST. The outcome was a "forced choice" scale (Part A) from multiple choice items thought to centre on Harvey's postulates and probably as easily modifiable as the TIB. Also, i f this approach should indicate some success, one might in time strengthen this scale by the "horses-mouth-approach", i.e., by substituting actual responses from Part B, the slightly modified TIB form. In the following display of the complete test form the reader w i l l be given system indications as scored for the available blanks. With the exception of the f i r s t question, a l l questions as well as item positioning are meant to be randomly arranged. The test is in booklet form. Every question appears on a separate page, here indicated by a short horizontal line. Presentation of the Modified Instrument This I Believe Questionnaire Adopted from Form TIB-74 (Copyright 1974 0. J. Harvey) By special permission from Professor Harvey, University of Colorado, 1975. What this questionnaire is a l l about This questionnaire attempts to investigate the extent to which students might respond differently from adults to issues that concern most people at one time or another. Adult responses to these kinds of questions have already been compiled and are well documented — but we don't have at present any equivalent data on high school students. It is for this reason that we ask you to give this questionnaire your close attention. You may put your name on the following identification page. But i f you are more comfortable without identifying yourself, just leave your name off — but, please, make sure you supply the other requested information. Identification Page Name: ' _ _ _ (Given Name) Sex: Grade: Part (A) Forced Choice  Directions The next 14 pages w i l l each present 4 different statements about one particular belief issue. Imagine yourself as being forced to take sides. That i s , you must choose one statement as representing you. This means that sometimes you may be i n a real bind because a l l statements may seem equally reasonable or equally bad — or equally incomplete. Even in such a case take one position to which you can give your most support. Read silently (so as not to influence others) a l l 4 statements. Then mark your choice - only one - with a capital M. (M) stands for: Most like my own position. When the time given for one item is over, the administrator w i l l direct you to turn to the next page. This I believe about questionnaires like this (item #1) (3) It depends on their purpose.' If put to good use, they can help people to learn more about themselves and others. (if desired) ., , (Family Name) Age: . (Year) (Month) School: 18 (1) They are necessary for establishing new knowledge and for advancing social science for the benefit of a l l people. (2) They are usually a complete waste of time and do no good to anyone. This one won't be any different. (4) I don't quite know what this is a l l about! But i t looks like a painless break from the daily routine — and i t may have some useful-ness. This I believe about schools (Item #2) (2) Schools are part of the establishment, but without them you are dead. (4) There are some things that anyone could bitch about; but at least there are quite a few choices. (3) What really counts is that you meet a lot of people who often become good friends. (1) Schools are a must for becoming a successful citizen. This I believe about lying (Item #3) (1) Lying is the exact opposite of saying the truth, and always reveals a coward. Once a l i a r always a l i a r . There is no such thing as a white l i e . (4) Lying i s the opposite of saying the t r u t h — but not the exact opposite. That makes i t complicated. (2) Lying is a fact of l i f e — like i t or not! Most people can be conned into anything. (3) Lying can save face and can prevent hurting people.. In such a case lying i s preferable to the truth. 19 This I believe about equal opportunity for both sexes (Item #4) (2) Females w i l l never have a real chance — the establishment is f u l l of male chauvinists who never w i l l give women an equal break. (1) A l l people must be given the same basic chance in l i f e . But men are men and women are women. Therefore, i t is best i f people stay with roles for which they are most suited by nature. (3) If this i s to be achieved, i t must be done carefully so that friendly relationships do not suffer. Women's l i b . i s only good i f i t does not hurt people in the long run. (4) Equal opportunity, yes! To do a l l things that are practical and enjoyable. But remember, equality does not mean sameness. For example, I don't expect many girls would want a job in the coal mines. This I believe about friends (Item #5) (1} Friends are very important in l i f e . Friendship i s necessary for developing a well-adjusted personality. People depend on each other and friendship provides this helping hand. (2) Friends are important — as long as you can trust them — and of this you can never be quite sure. Friends can be and often are one of the great disappointments in l i f e . (4) Friends are great! You have to understand their strong and weak points. I don't depend on my friends very much to do things that really interest me. (3) I like to do things together with my friends. Life without friends would be a real drag! The main thing, however, i s that people need and deserve help from each other. This is what true friendship is a l l about. 20 This I believe about alcohol (Item #6) (3) It can be both, harmless or harmful. It depends on the type of people and the occasion. As long as people learn to drink in moderation, they w i l l not harm themselves or others. (1) It can be a social disease, because many people have a tendency to become alcoholics. Therefore, some s t r i c t controls on the use of alcohol are required. (2) There always have been drunks. No one can do anything about the drink-ing problem anyway. If people want to become alcoholics that should be their business. (4) I guess people could do without i t . A real alcoholic is a terrible sight; but sooner or later most people w i l l learn how alcohol affects them. Then they have to make a decision — hopefully the right one. This I believe about my teachers (Item #7) (2) Most teachers are doing what they are doing because they get paid for i t . No big deal! (3) Teachers are people. You have to know them before you can say anything. Regardless of anything else, most teachers help students quite a b i t . (4) Teachers? I can't put them into one bag — they differ a lot. Maybe this i s a copout on my part. (1) Most teachers try. their best to make students into knowledgeable and law-abiding citizens. This I believe about teamwork (Item #8) (2) Most people hope that the others w i l l make a l l the effort; so, often nothing gets done unless somebody plays the sucker. 21 (4) Its success depends on the strength and weaknesses of the people involved. Sometimes you are better off to do the job yourself. (1) It i s good and necessary for development of character — especially in sports. The opportunity to pool resources for the common good should never be passed up. (3) It may not always succeed. But i f i t develops new bonds of friend-ship, everybody can gain something. Even the weakest person has something to offer. This I believe about fashions (Item #9) (1) Fashions are not as important as neatness. Nothing is as revolting as a g i r l who is sloppily dressed — regardless of fashion. (4) Fashions come and go. If you wait long enough you are always back in. I have not enough time to worry about them very much. (2) People are judged by what they wear. Fashions are invented to get money out of people who are scared to be called old-fashioned. (3) Fashions are fine as long as they allow people to express their own personality. This I believe about the church (Item #10) (2) It tries i t s best to hold on to i t s power and to keep people ignorant. It has polluted a l l true religion. (4) It doesn't bother me that there are many people for whom.the church doesn't mean much one way or the other. (1) If i t preaches the true gospel, a lot of people could be saved from sin of which we a l l are guilty. (3) The question should be: How much effort does the church make in 22 providing help and comfort to those who need i t most. My rating depends on the answer to this question. This I believe about my future (Item #11) (2) It a l l depends on how I w i l l make out. You need good connections; otherwise, you can't get anywhere. This is true now more than ever before. (4) I hope that I do not have-to worry too much about money or health, so that I can do things that really interest me. (1) It a l l depends, of course, on the economic situation. If industry and science keep up with the increasing demand and i f I do as well as is expected of me, my future should be secure. (3) There are many po s s i b i l i t i e s . But I hope to find a f i e l d in which I don't have to get a l l worked up about competing with others.. I would much sooner find a future that involves work with people. This I believe about foul language (Item #12) (3) It depends on the circumstances. But i t should never be used i f i t hurts people who mean something to you. (1) Even i f many people use i t , inside themselves they know i t i s f i l t h y and i t should be punished. (2) Since everything else has been corrupted, why worry about foul language? (4) Sometimes i t allows you to get r i d of your frustrations. But i f i t is used a l l the time i t gets to be a bore — i t is actually only noise in the air that cannot do real harm. This I believe about prejudice against East Indians (Item #13) (4) What I know about this problem is mostly second hand. I have no really good explanation what causes i t or how to cure i t — in myself and in others. (2) Most people only pretend that they are not racists. Everyone has to look out for himself in this dog-eat-dog world. Right now the East Indians are the real underdogs. (3) It has to do with their different religion and strange way of dressing. On a personal level most people are ok. So, one has to know them on an individual basis. (1) Prejudice and racism are bad and should be condemned. As long as the East Indians respect our laws and don't interfere with religion, there should be no prejudice. This I believe about our country (Item #14) (3) Whatever else may be said, Canada is known for i t s contribution to achieve harmony among people. We have not many enemies but a lot of friends. (2) It always used to be one of the best places in the world. Everybody with eyes to see knows what is happening right now, though. But there is no use talking about i t . (4) It is small in population and large in space. It is home to many different people — this makes i t an interesting country to grow up and live i n . (1) It is probably s t i l l the best and most respected placed on earth. We have more opportunities for people than most other countries. Canada stands as a symbol for freedom and peace. 24 Part B Free Response The following pages w i l l give you an opportunity to respond freely to a given issue. In this section you are simply requested to state your position — as i t f i r s t comes to your mind and as forcefully as you can. Do not worry too much about spelling or neatness. You w i l l have 2 minutes for each question. At the end of 2 minutes you w i l l be asked to turn to the next belief issue. Put as much of your own true opinion into i t as you possibly can. Try to write at least two sentences about each topic. This I believe about our school: This I believe about gi r l s going into technical occupations: This I believe about myself: This I believe about more immigrants coming into Canada This I believe about my own generation: This I believe about my best friends: 25 III. OBSERVATIONAL RESULTS FROM SAMPLE Response Categorizations from the "Free Response"  Section of the Modified Instrument Because the "free response" categorizations represent the anchor point for this study they w i l l be presented f i r s t . The strategy used as well as several measures of interjudgemental and intrajudgemental agreements have been carefully noted and are given a detailed exposition in Appendix A of this paper. Here, i t may be sufficient to indicate that assigning of belief systems was taxing the a b i l i t i e s of the judges much more severely than any of the other additional judgemental tasks. In comparison, the rank ordering of "articulateness" of respondents on a scale from 1-6, for example, appeared a very minor problem. Here, disagreement was never greater than one rank. Moreover, disagreements occurred much more rarely. The keys used for categorizations are presented in tabulated form below. TABLE 5 KEY USED FOR CLASSIFYING BELIEF SYSTEMS FROM THE "FREE RESPONSE" SECTION (PART B) 1, Dominant 1 1s 1-2, 1-3, 1-4, 1-X 2, Dominant 2 2-1, 2-3, s 2-4, 2-X X's 3, Dominant 3 1s 3-1, 3-2, 3-4, 3-X 4, Dominant 4 4-1, 4-2, s 4-3, 4-X The key in Table 5 represents a small departure from Harvey's key. The f u l l X designation means either that the judges could not agree on a designation 26 even a f t e r discussing the response or that they agreed that i t was impossible to categorize that p a r t i c u l a r response with even the s l i g h t e s t degree of confidence. IX, 2X, etc. means that the f i r s t named response type was i d e n t i f i a b l e - but not overly dominant because i t was d i f f u s e d by other discernable response types. Thus, for example, Harvey's designations of a 2-3-1 would collapse i n the present study to a 2-X. I t was understood that the X designation was to be used as l i t t l e as possi b l e . 1, 2, etc. designa-tions mean -the response type i s d e f i n i t e l y dominant - even i f other tendencies are vaguely discernable. 1-2, 1-3, etc. designations mean that both response types are i d e n t i f i a b l e - the f i r s t one stronger or equal to the second - again even i f other tendencies are vaguely discernable. TABLE 6 KEY USED FOR RANK-ORDERING "ARTICULATENESS" FROM THE "FREE RESPONSE" SECTION (PART B) 1-2 Below Average 3-4 Average 5-6 Above Average expressed without much, i f any, elaboration often only one sentence or a sentence fragment - thoughts are usually two sentences, expressions, although rather simple, are usually quite c l e a r . Thoughts tend to be expressed with some elaboration. usually two or more sentences. Fluency and expressiveness -seems high f o r grade l e v e l - thoughts tend - vocabulary r e l a t i v e l y r e p e t i t i v e and to be w e l l elaborated. r e s t r i c t e d NOTE: "Articulateness", ignores to some extent the mechanics of l i t e r a c y such as s p e l l i n g , punctuation, and grammatical correctness, e s p e c i a l l y where "expressiveness" i s greater t h a n - l i t e r a c y . 27 TABLE 7 KEYS USED FOR CLASSIFYING TWO ADDITIONAL RESPONSES FROM THE "FREE RESPONSE" SECTION (PART B) Degree of Acceptance of Immigrants From the "Free Response" item "This I believe about more immigrants coming into Canada." Strong Rejection Open rejection, direct reference to a particular group or race - negative commentary General Rejection Usually oblique reference to 1 others 1, also negative commentary Conditional Acceptance Complete Acceptance Acceptance only with Acceptance some or many ' i f s ' - without any or with some expressed reservations disapproval Degree of Acceptance of Girls From the "Free Response" item "This I believe about girls going into technical occupations." Rejection Total rejection and/ or openly expressed disapproval. Conditional Acceptance Expressed conditions such as strength, mental abi l i t y , etc. must be met - must equal that of men. Complete Acceptance Total acceptance; only condition - sometimes expressed is 'liking' of such an occupation. What follows is a minimally edited transcription of categorized examples from the "free response" section. Corrections of spelling and grammatical structure were restricted to those instances that seemed to obscure a ready access to the meaning. Some swear words were abbreviated in accordance with the convention of polite usage. For reference purposes a l l examples are prefixed with a capital letter and identified with the belief systems i t was judged to represent. The belief questions are indicated in abbreviated form. A l l examples are from the present sample of 122 grade 9 and 10 students from a high school i s a middle-class d i s t r i c t . 28 (System 1 - Articulateness: 5, Boy, Grade 10) Our school: Girls: Myself: Immigrants: Own generation: Best friends: Its o.k. but i t ' s not as good as a Junior High. I'd much rather prefer a s p l i t Junior & Senior High because I hate having so many people in one school But I certainly learned alot more up here they really push you & the teachers are quite good. I think i t s stupid, every since womens Lib movement they (women) think they should get away from the children leave the home & compete for higher employ-ment. They're just not made for i t . One of these days they'll learn. I'm the greatest that ever lived. I personally think since the Japs raided Pearl Harbour more and more chinks and Japs have been invading Canada. Vancouver has the largest amount of slant eyes than in any other part of Canada, geez the Japanese have practically overpopulated their own home cities so now their coming over here and taking a l l the jobs, and making more babies than ever. I think they stink. I don't know. I'm questions. I don't know where I'd be with out them. (System 1 - Articulateness: 3, Boy Grade 10) Our School: Girls: Myself: Immigrants: Own generation: Our school is okay because we don't let tor many foreigners in. The school is in a good residential area so the people aren't scums. If the girls can handle i t i t ' s okay with me but in a lot of jobs a man is more suitable. I'm a average guy a with predujice but average intelligence. I think they should get r i d of the one's in here and don't l e t any more stupid Pun Jabs and Chinese into the country. Our generation is with i t while the older generation is old fashioned like we w i l l be in a couple of years. Best friends: I think my best frinds are nice guys. They're responsible and honest and their not immigrants. 29 (System 1 - Articulateness: 4, Boy, Grade 10) Our school: Girls: Myself: Immigrants: Own generation; Best friends: That in our world today we need school to get a good job to look after a family. A few people who drop out get lucky but very rarely. I don't mind girls going to technical occupations as long as they can do the job. There is no reason why they shouldn't. I believe that I am not prejudice and do not consider myself a male chauvanist pig. I think that there should be a ^ certain no. of visa's given out a year because we have enough people on welfare right now without needing any more. I believe that my own generation is a good one. There are always some bad applies in a barre l l . I believe that my best friends consider me as equal as them. Everybody has there own opions of l i f e and that is what I feel. D: (System 1 - Articulateness: 2, G i r l , Grade 9) Our school: Gi r l s : Myself: Immigrants: Own generation: It i s good and the teachers are good. I think i f g i r l s wish to do that why not. I am a lucky person to be livi n g where I am and to go to a school, and not have to think about were am I going to get my next meal. I think Canada should think about the people we already have? It is good and we are lucky to have a l l the things we have. Best friends: I don't have a best friend only good friend. (System 1 - Articulateness: 5, G i r l , Grade 10) Our school: Our school i s really a good one. There are a lot of nice people and good teachers. It is clean and i t is not a problem to come to school, 'cause most of the people like i t here. Girls: I think that i t ' s great. If they can handle the job then why not? I am not a woman's l i b freak but i f that is what they want to do and i f they can do i t , well go right ahead. 30 Myself: I come from a good family. I enjoy sports, which takes up most of my time. I am not with the type of kids that drink or smoke. I don't think that i t is right. I enjoy myself and family and I have really great friends. Immigrants: Well knowing that we can only have a certain amount of people and that we are reaching our limit I think that we shouldn't have any more. Most of the people that are coming in are getting jobs and our own people can't. I don't believe that we should accept immigrants for jobs unless our own people have them f i r s t . Own generation: Best friends. Quite of lot of the people now are into drugs and drinking of course a lot • aren't. I feel that our generation is not controlled as s t r i c t l y as our parents which is good but i t isn't good either. Our generation is easier than our parents because, we didn't have to worry about wars; not being able to get food etc. but we have had a few problems. My best friends are really good people. They really do go out of their way to help others as well as them-selves. We are very close, and do a great deal of things together. I think that I am really lucky and I couldn't have any nicer friends. (System 2, Articulateness: 2, Boy, Grade 9) Our school: Gi r l s : Myself: Immigrants: Own generation: Best friend: I think this school is to small for a l l the people who go here. P.S. This school is f !!! For those g i r l s going into technical schools is good. P.S. But the schools they come for are f !!! Far-out Immigrants coming into Canada is good, i t makes Canada a better place to live in. My generation is far-out and better then the f. old foukes generation. I think my best friend is f !!! (System 2, Articulateness: 5, Boy, Grade 10) Our school: People in our school try very hard to upper-class and "cool". Really, however, a l l the back-stabbing and office p o l i t i c s that goes on though just shows our school as a viscous micro-cosm of the outside world. Girls : If they are qualified to do so, yes, i t is fine. However, i f they are seeking equality, but really strive to be dominant, I'm against i t . 31 Myself: Immigrants; Own generation: Best friends: I'm influenced in many ways by my friends. However, I .try to be as individual as possible and develop as a person. Right now, I'm lost in l i f e with so many personal choices evolving. Fine, but we should consider our job markets and other economic factors as well. We should also consider our English and French national heritage before being overwhelmed by in Eastern society within Canada. We are lost and confused as a whole. Our radical a c t i v i t i v i t i e s , though, only reflect our desires to be independent as previous generations have. We're going to f i t into the same establishmentarian mold as they do. My best friends, although few, are different. Many impress me as schemers. There are only a few in which I place an absolute trust. However, there i s always somethings I never t e l l them and they never t e l l me. (System 3, Articulateness: 6, G i r l , Grade 10) Our school: Girls: Myself: Immigrants: I think school i s a really important place to meet people and make friends. It is also good in the education point of view and i t helps kids prepare for the future. Our school is very cliquey (sp.?) but most of the people are in the group of friends that they want to be in so they enjoy i t . I really love xxxxx and would never change schools. I feel that i f a g i r l wants to go into a technical occupation i t is her decision and I'd give her credit for i t . I wouldn't want to be that g i r l , but I'm not going to put her down for doing what she wants to do. My friends are the most important thing in my l i f e and I couldn't live without them. I do care about my future, the world, etc. but I ' l l always put friendship before anything. In order to be happy I could do without everything but friends. I believe i t is fine for people to immigrate into Canada but they should be prepared. It isn't f a i r to just plop them in our country because they would be lost. But i f they are given (or have arranged for) a place to li v e , and they w i l l be able to cope, they sure can come - as long as i t doesn't affect us in the wrong way. There should, however be a limit 'cause soon they could outnumber Canadians (I mean people born here). Own generation: I dont'd think we are a 'bad generation' like alot of people have shown. We are responsible and have a 32 great advantage because we were born into a sophis-ticated world. We like to have fun, be rowdy, but most kids do have 'respect' for elders, property, etc. Best friends: My best friends are really important to me. I can t e l l them everything and they are understanding and truthful. They have pulled me through alot of times that my parents couldn't have helped with and I honestly couldn't live without them. I: (System 3, Articulateness: 5, Boy, Grade 10) Our school: Our school has a lot of opportunities. One has many choses to choose in what he wants to do. There is too much compitition in this school. A person should just be able to get involved. There are many good teachers in the school but some of them are either to slack, or s t r i c t . I wish that some teachers would act more like human beings. Girls: I feel that i t is good for a g i r l to go in for tech-nical occupations. In many respects we need a womens view point. This is only i f they try their best and have a lot of interest in i t . Myself: I want to accomplish something in l i f e that may help people. I try to treat everybody f a i r l y but I don't always do. I may chosen occupation I think that any office job would bore me. I would like a job out in the f i e l d away from the city. I believe that I would like to work close to nature. Immigrants: I think that immigrants should come to this country. They bring with them their culture and religions which we can learn from. But this should a l l be done in moderation. Own generation: My generation has many faults in i t . I believe that every generation has something to add to socity. So I believe that our generation has some good in i t . One is that feelings are express openly there is no more what other people think but this i s s t i l l prominent in my school. Best friends: My best friends should stand by me in time of trouble. There should be an understanding between us. Each of us should respect each others comments and interest. J: (System 4, Articulateness: 5, Boy, Grade 10) Our school: Our school i s pretty good, although we really don't need to stay here as long as we do. Work is stretched out in Grades 8 & 9, but in Grade 10 up they pack more in. 33 Girls: Myself: Immigrants: - i f they want to, why not? (Isn't that a slogan). If females who are well suited to certain technical fields, wish to have a technical occupation, then they should be incouraged as much as males are. It's hard to think objectively about yourself. I am academically inclined but lazy, and I have a poor social l i f e but enjoy talking to people. I enjoy people by themselves, not in gangs. Canada should welcome immigrants of any race, creed, or colour as long as our own economic situation can with-stand them (which i t can right now). We are a rich country and are obligated to help relieve other countries.... Own generation: Best friends: My own generation is the opposite of the last. Most of us agree (politically) with our parents. In short, we are lazy and spoiled and cannot live happily with-out comfort (therefore drugs, etc...). My best friends are great people; they are loyal, kind, generous etc.... Maybe I expect too much from friends. If that's the case, that's why I think I don't have many, (friends). K: (System 4, Articulateness: 6, G i r l , Grade 10) Our school: Its an okay school as far as they go. One probably isn't better than the other. Though in sports compitition perhaps certain areas of kids are better athletes than others because of diet, background and training. School is just a place to gain more s k i l l s and knowledge to prepare you for the future. Other things as making friends and social l i f e just f a l l in naturally. Girls: If a g i r l wants to go into a technical occupation i t s up to her. If though that is what she chooses there should be no social pressures to prevent her and companies should be willing to train her for the job of her choice. You (everyone of us) are entitled to a job or s k i l l we choose to train in providing there is room and work w i l l be available when the training ends. Myself: This i s rather an open question so heres how I ' l l answer. I want to live how I want and make choices for myself. Everything I do i f I'm interested I try as hard as I can. Perhaps I would be called a perfectionist. But that doesn't mean neat. Thats about a l l I have to say about myself. Immigrants: Everyone is born on to this earth completely equal to 34 each other. So i f someone wants to come to Canada they should be allowed but the situation changes slightly i f there is a job shortage, etc. The immigrant should try to f i t in as best as they can to our way but this is sometimes hard. It should also work vice versa. Own generation: Our generation is unique I think compared to anything I've heard of. Because of excessive dope and booze we are going to have a problem with ourselves as we get older. The trouble with the law enforcement w i l l be increased. I don't no how i t w i l l end or what exactly w i l l happen. One good thing from our generation is accepting each other better. This is important I think (feel). Best friends: I don't think best friends are totally important i t s best to try to get along with everyone as best we can. That w i l l make everything a b i t better. If that worked we'd have less problems. L: Systems 1-2, Articulateness: 3, Boy, Grade 9 Our school: I believe that our school is a l i t t l e lacky in some aspects of education that some teachers teach. Girls: I believe that even i f g i r l s are qualified to do technical work, they shouldn't be. A womans job is in the home, with the kids, etc. If a g i r l wants to work she should do maids work, cook or be a secretary or something like that. Follow tradition, leave mens work to the men. Myself: I believe that I am an intelligent, respectable, person. Maybe a b i t of an introvert, but aside from that I think I w i l l be a good citizen. Immigrants: I believe that there should be no more immigrants coming into Canada because our population is getting infested with i l l i t e r a t e s that cannot even speak english and i f they continue to come i n , they w i l l ruin our society. Own generation: I believe my own generation is very capable of chang-ing the world for better or worse. Best friends: I don't have any!! M: Systems 1-3, Articulateness: 5, G i r l , Grade 10 Our school: I really like our school alot. I wouldn't leave for anything. One thing though i t i s very cliquey (sp.) I feel sorry for people who are not in a clique. I guess that is the way with most schools. I think the best people go to xxxxx, and I'm proud to say this i s my school. 35 Girls A l i t t l e while ago this would not be considered proper. But I am a l l for i t , personally I can't see myself, but i f that's what some gi r l s want to do that is their buisness. Myself: Immigrants My friends i s what my l i f e i s a l l about. I enjoy other people, and that is one of the reasons I come to school as you could t e l l by my marks. If immigrants are willing to live the way Canadians do, why not. But I think there should be a limit because pretty soon there are going to be more Chinese people and Hindues than us. Own generation: Best friends: I feel my own generation are good people. We were to born in a space age and a time of changes. Sure there are bad people such as grease gangs or whatever. But I'm sure that there always have been. I think the image of the rebellious teenager has been exagerated. My best friends are very special to me. They are people that I can talk to about anything and visa versa. They are always there to help and without best friends I would be lost. N: Systems. 1-4, Articulateness: 4, Boy, Grade 9 Our school: School i s ok and you can learn quite a lot from i t , but you could possibly learn more without going to school. Girls Myself: Immigrants Own generation: Best friends: I think girls should have as eaqual opportunities a men, but some jobs aren't just suited for g i r l s . I believe in myself, and I know I can do anything I really want to. That is a l l that matters. I don't mind immigrants coming in , but i f too many come, there won't be any more room for us. I think that no more immigrants should be allowed in after the population reaches a certain point. It's alright, at least they're better than the last generation. Friends are a very important part of l i f e , and i t i s important that everyone has good friends. If you don't have any friends, you're in trouble. System X, Articulateness: 2, G i r l , Grade 10 Our school: Our school, like most these days, is based on competition (in sports, academics). I think this should be eliminated from the school. Girls: Myself: Immigrants: It's fine. There is no reason why they shouldn't enter such a f i e l d . Women are perfectly capable people. I don't believe much about myself. Fine, provided i t is on a limited and controlled basis. Own generation: Could be less corrupt. Don't push themselves into maturity. Best friends: They're pretty great people. System X, Articulateness: 2, G i r l , Grade 9 Our school: There is alot of nice people. But not much school s p i r i t . G i r l s : Myself: Immigrants : I think both sexes should do whatever they are comfortable in who cares i f they are male/female. I can't f i t i t a l l i n ! Too much to say. I don't care i f they come in or not. I'm not predjudice. Own generation: They are naive. They don't care about anything they are dumb. Best friends: My best friend I can trust but I do not like being with her/him a l l the time. Sometimes unaware. Table 8 on the page following w i l l display the frequency distribu-tions of the complete sample and of various analysed subgroups as defined by the keys in Tables 5 to 7. The abbreviations used derive from the complete headings used in Tables 5 to 7 and, therefore, seem self-explanatory The next subsection deals with response categorizations from the "forced choice" part. The "key" used for classifying systems w i l l be explain ed and then presented. This key, while based on probabilities of either one response type or two combined response types occurring by chance alone, was designed to allow a direct comparison with the categorizations obtained from the "free responses". The subsection w i l l then display the response type frequencies obtained within each of the 14 "forced choice" items and compare categorizations obtained by the "free response" and "forced choice" methods. .TABLE 8 •FREQUENCY DISTRIBUTIONS OF BELIEF SYSTEMS FROM "FREE RESPONSES" •OF TOTAL SAMPLE (N = 122) AND VARIOUS SUBGROUPS OF SAMPLE* - G • Description B i Dominant l's Dominant 2's Dominant 3's Dominant 4's o r i Y s 1 s T N 1 1-2 1-3 1-4 1-X 2 2-1 2-3 2 -4 2 -X 3 3-1 3-2 3 -4 3 -X 4 4-1 4--2 4-3 4-X X's Total Sample 73 48 121 50 7 8 6 5 9 5 - - 3 6 4 - 1 - 8 3 1 1 - 4 Boys, Grade 9 38 - 38 18 3 1 4 2 2 3 - - - - 2 - - - - 1 1 - - 1 Boys, Grade 10 35 - 35 17 4 - - 1 4 1 - - - 1 1 - 1 - 5 - - - - -A l l Boys 73 - 73 35 7 1 4 3 6 4 - - - 1 3 - 1 - 5 1 1 - - 1 Girls, Grade 9 - 20 20 3 - 4 2 1 1 1 - - 1 2 1 - - - - 2 - 1 - 1 Girls, Grade 10 - 28 28 12 - 3 - 1 2 - - - 2 3 - - - - 3 - - - - 2 A l l Girls - 48 48 15 - 7 2 2 3 1 - - 3 5 1 - - - 3 2 - 1 - 3 Articul.: 1-2 17 9 26 17 - - 1 2 3 1 2 Articul.: 3-4 37 9 46 22 5 4 1 1 4 3 - - 1 - 3 - - - - 1 - - - 1 Articul.: 1-4 54 18 72 39 5 4 2 3 7 4 - - 1 - 3 - - - - 1 - - - 3 Articul.: 5-6 19 30 49 11 2 4 4 2 2 1 - - 2 6 1 - 1 - 8 2 1 1 - 1 St. Re j . . of Imm. 11 4 15 9 2 1 Gen. Re j . of Imm. 19 13 32 20 2 1 1 1 3 1 - - 1 - - - - - 1 1 - - - -Cond.Acc.of Imm. 34 23 57 17 3 6 4 2 3 3 - - - 4 3 - 1 - 6 - 1 1 - 3 Compl.Acc.of Imm. 7 6 13 3 - - 1 2 1 - - - 1 1 1 - - 1 1 - - - 1 Rej. of Girls 8 - 8 3 2 - - - 2 1 Cond.Acc.of Girls 34 21 55 27 3 3 3 2 3 1 - - 2 2 2 - 1 - 4 - 1 - - 1 Compl.Acc.of Girls 24 25 49 16 1 5 3 2 3 2 - - - 4 2 - - - 4 3 — 1 — 3 * One person failed to respond in the "Free Response" section (Part B) of the instrument. Also, subtotals do not always add up to the expected number because on occasion a respondent failed to answer a particular item. Response Categorizations from the "Forced Choice"  Section of the Modified Instrument The key used for classifying systems from the "forced choices" is shown in Table 9 below as applied to system 1 and admixture 1-2 identi-fications. A re-arrangement of the response type totals into a different sequence allows identification' of the other systems. For example, the 1-2 system designation for a 9, 5, 0, 0 arrangement becomes a 4-3 for a 0, 0, 5, 9 re-arrangement. The rules adapted for system designations are based on probability considerations. They are as follows: (a) A system or an admixture designation w i l l be given as long as either the summed probabilities for chance occurrences for at least n A's of one category or at least n i A's and r\2 B's of two combined categories does not exceed 5%. Thus a score of (5) l's and (4) 2's does no longer qualify for either a system or an admixture designa-tion. Hence i t gets an X designation. (b) As long as the accumulated probabilities for two combined terms are at least 10 times lower than the accumulated probabilities for the highest term alone, an admixture designation w i l l be used rather than a system designation. Thus a score of (9) l's and (5) 2's is designated as admixture 1-2, while a score of (9) l's and (3) 2's retains a system 1 designation. (c) When the relationship in rule b no longer holds, an admixture becomes a 1-X rather than a 1-2. Thus a score of (6) l's and (3) 2's s t i l l qualifies for an admixture designation (Rule a) but no longer for a 1-2 designation. 39 TABLE 9 KEY FOR SYSTEM AND ADMIXTURE DESIGNATIONS FOR THE 14 ITEM FORCED CHOICE SECTION Response Percent Percent Type of of Totals 2 p.AOB 1 2 3 4 Designations .00 .00 A 2 H 1 .03 .00 10 4 0 0 1-2 .03 .00 10 3 - - 1 .03 .01 10 2 - - 1 .22 .00 9 5 0 0 1-2 .22 .01 9 4 - - 1-2 .22 .04 9 3 - - 1 .22 .11 9 2 - - 1 1.03 .00 8 6 0 0 1-2 1.03 .02 8 5 - - 1-2 1.03 .09 8 4 - - 1-2 1.03 .30 8 3 - - 1 1.03 .63 8 2 2 2 1 3.83 .00 7 7 0 0 1-2 3. 83 .01 7 6 - - 1-2 3. 83 .13 7 5 - - 1-2 3. 83 . 39 7 4 - - 1-X 3.83 1.35 7 3 ' - - 1-X 11.17 .03 6 6 - - 1-2 11.17 .67 6 5 - - 1-2 11.17 1.06 6 4 - - 1-X 11.17 4.00 6 3 - - 1-X 25.85 1.45 5 5 - - 1-2 25.85 6.14 5 4 - - X 25.85 13.09 5 3 - - X A = 4 — — X NOTE: p.A may be found in any table of summed binominal probability functions, p.AfiB have been derived from summations of exact probabilities for n]_Ann2B .„ , „ 14! l . n i + n , ,1,14-(nn+n?), [P niAnnoB = - -——; , -r) 'i {^ ) 1 1 ]. x z n i ! n 2! (14-n1-n2) 4y v2 y J Thus for any summed probability value for an A-B categorization in the table one reads at least of A and at least n 2 of B. The ceiling of a l l summations is n-j_+n2 = 14. Table 10 below w i l l display the frequency distributions of system type responses obtained within each of the 14 forced choice items. 40 TABLE 10 FREQUENCIES OF SYSTEM TYPE RESPONSES FOR EACH OF THE 14 "FORCED CHOICE" ITEMS FROM TOTAL SAMPLE (N=122) System Type Responses Type 1 Type 2 Type 3 Type 4 T.N. 1 22 2 69 34 117 2 43 27 29 19 118 3 14 21 48 33 116 4 40 4 17 57 118 5 38 6 59 12 115 6 12 6 60 40 118 § 7 15 12 65 26 118 2 8 18 3 71 26 118 ITE: 9 21 21 61 15 118 10 19 14 43 40 116 11 12 16 31 58 117 12 9 9 40 60 118 13 33 22 32 30 117 14 47 7 18 44 116 T.N. 343 170 643 484 1640 T.% 20.9 10.4 39.2 29.5 100 NOTE: Four respondents spoiled the "Forced Choice" section of the instrument. In some instances respondents of the remaining 118 failed to respond to a particular item, hence not a l l totals of system type responses add up to 118. A comparison of frequency distributions of systems and admixtures obtained from the total sample by either the "free response" or the "forced choice" section i s provided in the grouped bar graph of Figure 1.below. Table 11 which follows shows in detail the number and extent of agreements on scores obtained by either section. 41 CO w H u s H D Oi "Free Responses" "Forced Choices" Dominant 2's Fig. 1. Frequency distributions of belief systems and dominant admixtures as "seen" from "Free Responses" (.lined columns) and from 14 "Forced Choice" items (dark columns). For admixture types read from top down; thus 1 means 1-4. 4 Table 11 on the next page reads as follows: (1) Fu l l agreements are shown in the narrow zigzag band running from top l e f t to bottom right. (2) "First term" agreements (such as 1-2 and 1; or 1-2 and 1-4) are shown in the double T-lined wider zigzag band. (3) Agreement between "second and f i r s t terms" (such as 1-2 and 2; or 1-2 and 2-4) f a l l within the narrow bands of crosses. (4) Complete "reversed term" agreements f a l l into the intersection of the crosses. (5) Agreements on "second term" alone have been circled. Also, wherever a score f a l l s into any " f i r s t term" or "second term" X position at least a minimal part agreement exists. TABLE 11 NUMBER AND EXTENT OF AGREEMENTS ON SCORES FROM "FREE RESPONSES" AND FROM "FORCED CHOICES" FROM TOTAL SAMPLE (N=122) SYSTEM DESIGNATIONS FROM "FREE RESPONSES" Dominant l's 1 1 1 1 2 3 4 X Dominant 2's 2 2 2 2 1 3 4 X Dominant 3's 3 3 3 3 1 2 X Dominant 4's 4 4 4 4 1 2 3 X X •rH -o Q rH CN i—I ro i—i i—l X 1 1 1 5 1 o o o o o O O O c w £3 CN o Q CN CN rH CN ro CM <tf CN X o o o o o o D o o -H -g ro o . Q ro ro rH co CN co <tf ro X 6 1 9 1 © o D © 2 1 1 O o 2 1 o o a rH ^ CN <tf ro ^ x o 1 1 © o 1 o o © o 2! 1 x. 2 1 2 2 1. 48 7 8 5 4 9 5 6 4 3 1 1 NOTE: The reduction to 117" observations was caused by spoiled responses previously indicated. For admixture types read from top down; thus 1 means 1-4. 4 43 IV. DISCUSSION Examinations of Examples of "Free Response"  Patterns as Indicators of System  Funtioning; Success and Failures The examples of response categorizations from the "free response" presented in the previous section are intended to reveal both the sensi-t i v i t y and, paradoxically, also the shortcomings of the "free response" measure. In order to appreciate the problems of the judgemental task, one needs to know a few additional complexities of the belief systems model that were not mentioned in the introduction of this paper."'" As originally conceived, the belief systems are in a sense a chart of developing human personality. This chart shows "nodal" poles of "arrestations" and also "transitional" stages along a progression to the most complex potential 2 functioning of system 4. A further complication of the model is that i t accommodates inadequately simultaneous functioning of different inter-"""Harvey, Hunt, and Schroder, Conceptual Systems and Personality  Organization, chap. 6, pp. 158-203 gives a very detailed account of these complexities. In order to il l u s t r a t e a point adequately, i t seems unavoidable from time to time to refer to some additional complexities of the model. 2 Ibid., p. 66. The authors make clear that they do not attach any teleological meaning to the word "function". Thus, i f a system functions with greater complexity than another system i t does not necessarily have a different or higher purpose. If there is a purpose to any kind of "functioning" at a l l i t is the aforementioned principle of maintenance of the "self" which i s , however, the seed (or axiom) from which the whole theory of belief systems derives. pretation matrixes (belief systems) for different concepts." It is especially this last complication that w i l l register i t s e l f in incon-sistencies of responses. In order to gauge the most dominant or central aspects of the conceptual functioning, a judge must somehow absorb the global impression from a l l responses and constantly modify the impression 4 with some specific cues from one or the other response. So far as the experience with the sample shows, the "free response" measure appears at some instances remarkably sensitive and at other instances remarkably problematic. Dif f i c u l t i e s derive from at least three inter-related causes: (1) lack of judgemental expertise, (2) lack of expli c i t -ness in responses, and (3) lack of sufficient consistency in response patterns. This discussion w i l l proceed f i r s t with the apparent successes of the measure. Any free response measure (the present one is perhaps best character-ized as a guided-projective test) has the advantage that i t does not impose limits or constraints on possible responses, for this reason i t w i l l always be potentially more revealing than pre-scaled measures. A second advantage of this particular test is that i t calls for "opinions" rather than for "facts". People, and perhaps especially adolescents, seem to be much more ready to give an opinion about something than to account for some JIbid., p. 77-78, p. 111-112, and p. 170-171. Solutions proposed to the problem of "generalization" of functioning are probably just as tentative as they were 15 years ago.. 4 Harvey, "System Structure, F l e x i b i l i t y and Creativity", p. 47, to quote: "This global approach has been found to produce a higher r e l i a b i l i t y and validity than single item analysis, presumably because a context or yardstick i s provided against which a single response can more stably and accurately be interpreted." factual information."^ When examining many responses, these advantages are quite apparent in the extent to which several dimensions of the whole range of the belief systems seem to be represented. Since system 1 responses are represented in relatively large numbers and since they constitute the base position from a developmental point of view, the previous section displays as many as five different examples of these. Example A (p. 28) besides exhibiting most of the postulated signs of system 1 functioning — i.e., a black and white evaluativeness-avoidance of uncertainty, etc. — betrays very clearly a sudden narrowing of the subject's interpretive matrix, an increased "closedness" when a concept becomes more "central" i.e., more ego-involving. Thus, in the response patterns about school the respondent is relatively more "open" than in responses about immigrants and g i r l s . One can infer from the theory that school is a relatively "peripheral" concept to this respondent while "immigrants" is a particularly "threatening" concept, and, hence, causes a displacement of functioning towards increased "closedness".^ One may also note the great defensiveness in particular to the response about "myself". Within the theory this exaggerated proclamation about the "self" could be interpreted as a form of "avoidance". The administrators of this test reported several indications that the students enjoyed doing the test. In fact, some students who had not been selected for the test did ask whether they too would be allowed "to do one of these books". 6Since some of the Harvey, Hunt, and Schroder terminology not formerly introduced is almost self-explanatory this paper w i l l indicate usage of this terminology by quotation marks but w i l l not supply definitions. 46 In example B (p. 28) one can infer a much more homogeneously functioning system 1. The closedness is spread over a l l responses. One must note the extreme response generalization to almost a l l responses reflecting a particularly strong "ingroup-outgroup" feeling. Even in the response to "myself" the respondent makes an oblique reference to his concern with "others". Within the terminology of the theory this a l l -pervading almost involuntary response pattern would be characterized as "ambiguous". Examples C and D (p. 29) represent less extreme and more typical system 1 response patterns. On f i r s t sight the pleasantness and level of articulateness displayed in the response patterns of example E (pp. 29-30) might be mistaken for a more abstract functioning, but the reader should note the overall simplistic points of view. One must infer from the theoretical postulations a functioning characteristic of a secure belong-ing to the "right" class of people. The few problems perceived a l l seem solvable by doing the things that are "proper" and avoiding those that are not. Inferring again from the theory, one can assume that this system 1 functioning is a relatively "open" system within the points of "arrestation" either because a l l the stimuli from the questions are "peripheral" or because the respondent is so securely embedded in his environment that none of these questions could ever constitute a "refutation", i.e., a "threat". System 2 functioning, according to theory, is a necessary develop-mental progression towards more complex functioning, but as a stage of arrestation i t constitutes almost an inversion of system 1 functioning. Thus, the complexity of an object-subject relationship can be almost identical to system 1 functioning, but the direction of this relationship is usually inverted in the same sense that a love-hate relationship may be viewed as an inversion. 47 Hence, in example F (p. 30) system 2 functioning shows, i f any-thing, even greater "closedness" and even greater "ambiguity" than system 1. functioning in example B. However, in example F, one can infer from the generalized curse extending to four of the responses an a l l pervasive negativision. Here other negativism seems to follow the simple rule that the perceived direction of the responses of generalized "others" must be opposed on principle. This rule would explain the positive response to "immigrants". Also of note is the "avoidance" in the response to "myself". The excessively closed and generalized s;ystem 2 funtioning portrayed here 7 may well be a pre-indication of some form of developing psychopathology. Example G (pp. 30-31) in contrast to the previous one indicates a far more "open" system 2 functioning. In spite of a presumably higher level of intelligence indicated throughout the responses, one must note the all-pervasive distrust of others evident in every response, and also the fear of possible loss of individuality. From the overall response pattern one may infer an "indirect" system 2 functioning originating, according to theory, from neglect and indifference rather than from un-g r e l i a b i l i t y in the process of upbringing. System 3 functioning in example H (pp. 31-32) can be inferred from the very definite response focus on "people". This stage has been characterized in the original formulation of the theory as "mutuality". It may be helpful to think of this stage as a fence-sitting posture. This disposition finds expression in a peculiar s k i l l of anticipating an Harvey, Hunt, and Schroder, Conceptual Systems and Personality  Organization. See chap. 9, pp. 272-325 for an accommodation of the theoretical concepts of the belief systems to psychopathology. g Ibid., see p. 180-181 for "Indirect-Direct Variation within System 48 opposing view point which is always directly or indirectly accommodated by modifying or diminishing the "own" opinion. This "fence-sitting"" posture can be seen in a l l but two responses. The only two responses that carry a high degree of conviction are "myself" and "best friends". Both express an absolute, almost desperate, need for friends. These patterns are very consistent with the theoretical position that "refutation" of a system 3 functioning would be experienced by rejection and aloneness; hence, the great concern to remain inoffensive. In terms of the theory this is a form of "neutralization". Example I (p. 32), also judged as representative of system 3 functioning, is similar to the previous one. One should note the expressed fear of competition. This fear would be consistent with the theoretical assumption that any measurement with an absolute standard constitutes a potential rejection, hence a potential "refutation". Examples J and K (pp. 32-34) , judged to be representative of System 4 functioning, display a relative lack of evaluativeness but without the fence-sitting posture characteristic of system 3 functioning. There is also a noteworthy lack of self-consciousness, an enviably- relaxed exposition of thought, an awareness of multiple solutions to problems — a l l these features are quite consistent with the theoretical postulates of system 4 expression. This paper has now demonstrated in some detail that belief systems as postulated are, in fact, readily inferable from certain "free responses". Not only does i t appear possible to categorize these response patterns with relatively high judgemental confidence (see Appendix A), but also, to deduce certain theoretical positions within narrower bounds of the model. These theoretical points were selectively alluded to in the hope that they might demonstrate the degree of sensitivity of this instrument. With "admixtures" or, in a developmental sense, "transitional stages", judgemental confidence, as far as the resources of this study are 9 concerned, was immediately reduced (one may again refer to Appendix A). Discounting for a moment the problem of a possible lack of sufficient judgemental expertise for this particular study, one must, nevertheless, in examining the many response patterns rated as admixtures, acknowledge intrinsic problems. To i l l u s t r a t e , in example M (pp. 34-35) one finds an overall evaluativeness, a definite ingroup-outgroup feeling characteristic of a relatively "open" system 1 functioning. One must also note a strong people orientation but without strong indications of "manipulativeness" or dependency patterns. It would seem that the -3 rating, here occupying the second position, is more questionable than the 1- rating. The reader may find i t illuminating, or amusing, to apply his own ratings against the categorizations given to examples L (p. 34) and M (p. 34-35). Whenever an X enters into an admixture, one is faced with even more diffuse response patterns (in this sample there remained eight admixtures with an X as second term); but where an X appears alone, categorizations become meaningless because responses appear either truly contradictory or un revealing. From this sample of 122 respondents, four remained with an X rat ing. Two of these are transcribed as examples 0 (pp. 35-36) and P (p. 36). Once again the reader is invited to examine these for his own satisfaction. From the theory of development of system functions, one should find "transitional stages", i.e., designations of the types 1-2, 2-3, 3-4, (order may not be very important); but i t is not immediately clear how the theory accommodates "admixtures", i.e., designations, of the types 1-4, 2-4, 1-3. "Admixtures" present problems of "generalization" (see Footnote 3 p. 44) or present problems of some form of "regression". See also Harvey, Hunt, and Schroder, Conceptual Systems and Personality Organization pp. 282-283. 50 This paper w i l l now discuss the observations from the "forced choice" section of the instrument. Failings of the "Forced Choice" Section Even a cursory inspection of the observations from the "forced choice" section w i l l show that this measure is no ready substitute for the "free response" measure. For example, Fig. 1 (p. 41) shows at a glance the totally different patterns of the frequency distribution obtained by TABLE 12 TWO INDEXES OF SYSTEM RESPONSE TYPE ATTRACTION FOR 14 "FORCED CHOICE" ITEMS ITEM ATTRACTION FOR SYSTEM TYPE RESPONSE T.N. l's 2's 3's 4's (IF) Average expected from "Free Responses" 117 76 17 11 13 (10) Average observed from "Forced Choices" 117 24 12 46 35 NOTE: A l l comparisons were adjusted to the 117 observations avail-able from the "Forced Choice" section. either measure. Even more convincing is the lack of agreement when viewing the two measures interrelatedly as in Table 11 (p. 42); here, one finds only five complete agreements from 117 observations. Nevertheless, looking at these failings a l i t t l e more closely w i l l be instructive. By using data from Table 8 (p.,37) and Table 10 (p. 40), one can create some convenient indexes which w i l l make an inspection of the system type responses obtained for the 14 "forced choice" items somewhat more meaningful. 1 0 These indexes, shown in Table 13, may require some explanations. Index (IF) has been obtained By defining a "system type response" as a "score of attraction power" for the item (the item is the one that scores), i t becomes meaningful 51 by collapsing a l l System l's and dominant System 1 admixtures into l ' s ; the same was done for 2's, 3's, and 4's (shown under "ALL RESPONSES" in Table 8, p. 37). Index (IF), in spite of i t s precarious position due to judgemental uncertainties, provides a possible "baseline" against which index (10), the average Observed Item Attraction (computed from Table 10, p. 40), may be compared. This writer takes the position that index (IF) is a valid pro-visional baseline for a possible improvement and adjustment of the forced choice section. This point w i l l be elaborated in the concluding section; for now, i t w i l l be sufficient to discuss the possible causes for the apparent failings of the "forced choice" section. The index relationships in Table 13 below show the necessary factors required for bringing the observed average attraction power of the system type responses up to the TABLE 13 RELATIONS OF INDEXES (l's) OF SYSTEM RESPONSE TYPE ATTRACTION FOR ~14 ."FORCED CHOICE" ITEMS l's SYSTEM RESPONSE TYPE 2's 3's 4's (IF) = 3.1(10) (10) = .32(IF) (IF) (10) = 1.5(10) = .67(IF) (IF) = .26(10) (10) = 3.8(IF) (IF) = .4(10) (10) = 2. 5 (IF) baseline (IF's). The inverse relationships are also shown. For example, i t w i l l be seen that the average attraction of the 14 items for response type l's would have to be 3.1 times higher than observed in order to reach to use the concept of averages. For example, thus defined, the sum of attraction powers for system 1 type responses for a l l 14 "forced choice" items equals 343; hence, the average attraction power of the instrument for type 1 responses equals 343 * 14 = 24. This concept of average attractiveness w i l l allow a comparison with the average tendencies used in writing the test items. the expected value of (IF) at the baseline. A question arises from this data, why does the "forced choice" section attract only about 1/3 as many type l ' s , 2/3 as many type 2's, but 4 times more 3's and 2.5 times more 4's than would be expected from the (IF's). For a possible answer one must re-examine the "central tendency" employed in writing the items. One of the problems anticipated before writing the items was the contamination effect from social desirability; hence, the tendency for the response types to be equal in length and equal in apparent reasonableness. This tendency, in turn, usually dictated engaging the belief system type responses in a "peripheral" rather than in a "central" manner. Pursuing the line of reasoning above, the results shown in Table 11 (p. 42) take on a new perspective. One can see that 3's and 4's have very reasonable agreements on both measures, while most of the l's and 2's from the "free response" measure have been "drawn" also into the 3's and 4's by the "forced choice" measure.^ Because of their relatively high levels of "differentiation", 3's and 4's, w i l l (a) always recognize their "true" position even when their conceptual position is engaged only "peripherally"; and (b) have no incentive of any kind to "deny" that position; hence, the overall high agreements. One the other hand, the l's and 2's, because of their relatively low levels of "differentiation", may (a) in many instances f a i l to recognize their "true" position, and (b) having done so, take the remaining cues for social desirability and, hence, take the 3 or 4 option. Moreover, those l's and 2's that do not f a i l to Relatively small disagreements such as 3-1 and 1-3 or 3 and 3-1 are probably caused by the interaction of (a) judgemental error and (b) shortness of the "forced choice" measure. The high number of X and X admixtures produced by the "forced choice" measure would likely disappear by lengthening the measure to, say, 20 items. 53 "see" their "true" position w i l l most certainly also "see" the remaining cues for social desirability and may then very well opt in this "non-threatening" (i.e., "peripheral" concept engagement) for any "easy l i e " . While the above argument does explain the central tendencies of distortions for the "forced choice" measure, i t does not account for the rather large variations in "powers of attractiveness" for system type responses among items. (One could show the magnitude of these fluctuations by computing some form of variance index.) A content item analysis may explain why some items have, for example, almost no "power" to attract a system 2 type response. These fluctuations are readily seen in Table 10, p. 40. A content item analysis should show to what extent the central tendencies of item construction have not been adhered to. One may ill u s t r a t e this process with two items. Item #2 on pp. 17-18 has almost no attraction for type 2 responses. Instead, i t has a higher attraction for type 3 responses than the already high average of the measurement instrument taken as a whole. The reasons for these discrepancies are most likely a too exaggerated "negativism" for the type 2 choice and, conversely, a too pronounced "reasonableness" for the type 3 choice. Item #12 on p. 22 may serve as a second example. For this item, as Table 10 reveals, the power of attraction for a type 1 response i s only about 1/4, but for a type 4 response, almost 2 times the average attraction of the instrument as a whole. Here, the system 1 type response may well be culturally passe, i.e., a connection of " f i l t h y language" and "punishment" seem out of date to a l l but a few respondents. In contrast, the system 4 type response, by being the longest and having the "wisdomish" l i t t l e afterthought, appears sophisticated and, hence, unusally reasonable. In respect to item variance of "power of attractiveness" type 2 responses have the greatest fluctuation. (Table 10, p. 40.) There are 54 five items for which the power of attraction for type 2 responses f a l l s to less than half the average. This great unevenness would, of course, tend to weaken the power of the instrument to "recognize" system 2's, for the keying of the responses was tied to probability rules (see Key - Table 9, p. 39). In summary: this discussion has attempted to explain the central tendencies for the "forced choice" instruments to (a) correctly recognize 3's and 4's and, (b) falsely draw l's and 2's into type 3 and 4 categoriza-tions. It has also attempted to show that a content item analysis w i l l reveal reasons for exceptions to these central tendencies and, lastly, has pointed out one additional reason for the instrument's i n a b i l i t y to recognize system 2's. In the section to follow,the paper w i l l examine the obtained fre-quency.distribution of belief systems for this sample with a reported frequency distribution of a large sample from a different population. There-after, the paper w i l l compare different frequency distributions found among subgroups of the present sample. External and Internal Group Comparisons  of the Sample Under Study From groups for which Harvey has reported frequency distributions of belief systems, the most appropriate for an external comparison of this study is his reported sample of "several 1000 l i b e r a l arts students" (see p. 9 this paper). The best attainable estimate of the true frequency distribution of belief systems for this sample i s , as was argued, the one obtained with the "free response" measure. It was hypothesized that ..the present sample would show significantly more l's and admixtures than Harvey's sample. The rational for this hypothesis rests in part on the relatively large size of Harvey's sample which, surely, allows one to accept 55 the reported frequency distributions as a very good estimator of the distribution in his defined population of li b e r a l arts students. To the extent that this sample bears an acceptable resemblance to a true random sample, the differences hypothesized between this sample and Harvey's should prevail. Table 14- below shows the distribution of this sample and the expected distributions for the sample i f i t were drawn from the defined population of lib e r a l arts students. TABLE 14 BELIEF SYSTEM DISTRIBUTIONS OF SAMPLE AS ESTIMATED BY THE "FREE RESPONSE" MEASURE COMPARED TO THE ESTIMATED DISTRIBUTIONS IF IT HAD BEEN DRAWN FROM A POPULATION OF LIBERAL ARTS STUDENTS BELIEF SYSTEMS l's 2's 3's 4's ADMIXTURES TOTALS "Observed" 50 9 6 8 48 121 "Expected" (rounded to whole number) 42 18 24 9 28 121 NOTE: Figures for "expected" are computed from entries in Table 3 (p. 9). Figures for "observed", derived from Table 8 (p. 37). An inspection of Table 14 shows that system l's and admixtures are indeed more frequent than could be expected had the sample come from the population of li b e r a l arts students. But since there is a nearly equal frequency for the 4's i t can be seen that the "drain" into l's and ad-2 mixtures has come at the "expense" of 2's and especially 3's. X computa-tions show the following: (I) the sample is significantly different even at the last entry p = .001 2 in the table for chi-square values df = 4. (x = 33.92) 56 (II) by collapsing categories, other than the one tested into one c e l l one finds in Table 15 below (for a l l Ho: frequencies equal, and Hj:: frequencies not equal) whether or not these differences are s t a t i s t i c a l l y significant at a levels .001 and .05. TABLE 15 CHI-SQUARE TESTS FOR DIFFERENCES IN DISTRIBUTIONS OF BELIEF SYSTEMS df = 1 BELIEF SYSTEM DIFFERENCE SIGNIFICANT? = .001 = .05 System l's 2 X. =2.33 No No (one-tailed test) System 2's 7C = 5.29 No Yes System 3' s X.2 = 7.84 Yes Yes Admixtures Yes Yes 2 X. = 18.59 (one-tailed test) The above findings may be more easily interpreted in the light of differences of subgroups in this sample. These differences w i l l be pre-sented below. "Articulateness" (as defined in Table 6, p. 26) was assumed to be lower for system l's and admixtures. For a comparison of "articulateness" between observed and expected levels this paper w i l l assume that levels of "articulateness" are distributed normally; further, that average ratings (3-4) w i l l f a l l approximately within 1/2 standard deviations to either side of the median and, thus, account for about 40% of a l l ratings. Ratings 57 of Xl-2) would account for 30% and ratings of (5-6) for the remaining 30%. This d i v i s i o n i s suggested to provide a reasonable way of e s t a b l i s h i n g expected values f o r d i s t r i b u t i o n s of l e v e l s of "a r t i c u l a t e n e s s " within any of the categorized b e l i e f systems. Table 16 below compares observed (OBS) and expected (EXP) l e v e l s within the b e l i e f systems and admixtures. TABLE 16 LEVELS OF "ARTICULATENESS" IN BELIEF SYSTEMS AND ADMIXTURES System l ' s System 2's System 3's System 4's Admixtures Total N's OBS EXP OBS EXP OBS EXP OBS EXP OBS EXP OBS EXP 'ENESS 5-6 High 11 15 2 2.7 6 1.8 8 2.4 22 14.4 49 36. 3 t-i < "ARTICU 3-4 Avg. 22 20 4 3.6 0 2.4 0 3.2 20 19.2 46 48.4 En O EVELS 1-2 Low 17 15 3 2.7 0 1.8 0 2.4 6 14.4 26 36.3 Totals 50 50 9 9 6 6 8 8 48 48 121 121 Table 17 compares observed and expected l e v e l s of a r t i c u l a t e n s s for dominant l ' s , 2's, 3's and 4's ( i . e . , admixtures were collapsed into t h e i r dominant positions as indicated by t h e i r f i r s t term). As can be seen from the data i n Tables 16 and 17, "articulateness" for system l ' s as w e l l as f o r dominant l ' s , while s l i g h t l y underrepresented at the low end, follows very nearly the expected d i s t r i b u t i o n . The same can be s a i d f o r system 2's and for dominant 2's. In sharp contrast, the "articulateness" for system 3's and 4's f a l l s e n t i r e l y into the high end, l e v e l s (5-6). Even for the dominant 3's and 4's (Table 17) t h i s trend 58 prevails with the exception of three cases of average "articulateness" for the 3's and one case of average articulateness for the 4's. By inspecting TABLE 17 LEVELS OF "ARTICULATENESS" IN BELIEF SYSTEMS COMBINED WITH DOMINANT ADMIXTURES .Dominant l's .Dominant 2's .Dominant 3's Dominant 4's Total N' S JLATENESS" OBS EXP OBS EXP OBS EXP OBS EXP OBS EXP JLATENESS" 5-6 High 23 22.8 5 5.1 8 3.3 12 3.9 48 35.1 F "ART I CI 3-4 Avg. 33 30.4 8 6.8 3 4.4 1 5.2 45 46.8 O oo 1-2 Low 20 22.8 4 5.1 0 3.3 0 3.9 24 35.1 t> w Totals 76 76 17 17 11 11 13 13 117 117 NOTE: Four complete admixtures (X's) were removed from this comparison. the raw data, one may f a i r l y safely conclude that the system l's and 2's as well as the dominant l's and dominant 2's have a distribution that follows expected frequencies of "articulateness" as here defined. In spite of relatively small numbers, i t is also safe to conclude that system 3's and 4's as well as dominant 3's and dominant 4's are at significantly higher levels of "articulateness" than expected. For the few instances where the data in Table 16 does not show very clear-cut positions, "X. computations have been performed. These are summarized below: (a) System l ' s : The slight skew into lower levels of "articulateness" is not significant ( X 2 = 1.54, df = 2, a = .01). 59 (b) Admixtures: The displacement into higher than expected level of "articulateness" i s not significant at or = .01, but is significant at cc = .05 ( - X 2 = 8.91, df = 2) . (c) The total sample: The skew into higher levels of "articulateness" than expected is not significant at of = .01, but is significant at or = .05 (-x.2 = 7.4, df = 2) . The questions arises, to what degree was the judgemental categorization of responses dependent on levels of "articulateness". In order to answer this question each categorization was given a score of "articulateness" as well as a score of "judgemental confidence". This data is shown in Table 18 (p. 60). Contingency data in Table 19 shows observed sex distributions within combined belief systems and dominant admixtures with the expected values of proportions derived from the total division of boys and gi r l s in the sample. Similarly, Table 20 shows observed and expected grade-level distributions. The data is presented as sufficient evidence for stating that neither sex nor grade level is significantly differently proportioned within any of the belief systems combined with dominant admixtures than could be expected from the..total proportions of the sample. Before interpretation of the findings in the next section the results may be briefly summarized as follows: Comparisons of the present sample with the distribution patterns as reported by Harvey, show greater numbers of system l's and admixtures. These differences follow expectations but are only significant for the admixtures. System 2's and 3's are significantly "underrepresented". But for system 2's the difference is not significant at Of = .001. System 4's are essentially equally represented. The extent of "underrepresentation" of the 2's and 3's in this sample were unexpected. 60 TABLE 18 TEST FOR INDEPENDENCE BETWEEN "JUDGEMENTAL CONFIDENCE" AND LEVELS OF "ARTICULATENESS" Low (0-1) JUDGEMENTAL CO! Med. (2) JFIDENCE High (3) Totals OBS EXP OBS EXP OBS EXP OBS EXP 5-6 = High oo oo 14 12.96 17 17.41 18 18.63 49 49 w En S B 3-4 00 rt! j Avg. £ B 12 12.17 19 16.35 15 17.49 46 46 W H Pi < 1-2 Low 6 6.88 7 9.24 13 9.88 26 26 Totals 32 32 43 43 46 46 121 121 NOTE: Data for "Judgemental Confidence" derives from recorded agreement levels between Judge I and Judge II (see Appendix A), "Low" means either no agreement or only partial agreement, "Med." means 1st term or reversed agreement (as 1 and 1-2; and 1-2 and 2-1), and "High" means f u l l agreement. For Ho: Variables are independent Hi: Variables are dependent ?c2 = 2.54, df = 4, a = .01 These figures indicate acceptance for Ho - which in turn means that there i s no dependency between these variables, i.e., "Judgemental Con-fidence" was not influenced by levels of "articulateness". A corollary, which seems more important than the direct result, i s that judgements were not influenced by levels of "articulateness". Comparisons of subgroups show that 3's and 4's are significantly stronger represented in high levels of "articulateness" than l's and 2's. One should note here that the comparisons are not relative to the total observed proportions of levels of "articulateness" but relative to the expectation of normal distributions of levels of "articulateness" within each system. This procedure allows the conclusion that l's and 2's, while 61 lower than 3's and 4's in representation of higher levels of "articulateness", are not significantly differently distributed from expected values of levels of "articulateness". For the total sample the observed levels of "articulateness" are skewed into the higher end (significant at ot = .05, but not at oc = .001) . TABLE 19 SEX DISTRIBUTIONS IN BELIEF SYSTEMS COMBINED WITH DOMINANT ADMIXTURES Dominc OBS mt l's EXP Domini OBS mt 2's EXP Dominc OBS mt 3' s EXP Dominc OBS mt 4's EXP Tot OBS als EXP CO o X « 50 46.77 10 10.46 5 6.77 7 8.00 72 72 w — CO w r H • H U 26 29.23 7 6.54 6 4.23 6 5.00 45 45 Totals 76 76 17 17 11 11 13 13 117 117 TABLE 20 GRADE DISTRIBUTIONS IN BELIEF SYSTEMS COMBINED WITH DOMINANT ADMIXTURES Dominant 1's Dominant 2's Dominant 3's Dominant 4's Totals OBS EXP OBS EXP OBS EXP OBS EXP OBS EXP CO co -Q 38 36.38 8 8.14 5 5.26 5 6.22 56 56 e> _m Ci rH 38 39.62 9 8.86 6 5.74 8 6.78 61 61 Totals 76 76 17 17 11 11 13 13 117 117 "Levels of Articulateness" do not seem to have any bearing on the "levels of judgemental confidence" and hence, by implication, no bearing on the belief system categorization of "free responses". Finally, neither sex nor grade levels show any notably dispro-portional representation within any of the belief systems combined with dominant admixtures. 63 V. CONCLUSION: INTERPRETATIONS, AND POSSIBLE DIRECTIONS OF FURTHER RESEARCH Questions of Adequacy of the  "Free Response" Measure A rephrasing of the primary purpose of this study w i l l help to make the following interpretations of outcomes appear in a natural sequence. The question asked: is i t possible in a "one-shot measure" to see enough of the specified variables from a personality model reflected in responses from adolescents."'" In more simple terms this is like asking when viewing a picture: "can I see enough of what I am supposed to see". In the discussion section of this paper the question has been answered with a qualified "yes". What remains now is to explain what precisely i s "enough" — and the answer to this depends, in the opinion of this writer, entirely on the purpose for which the instrument is meant to be used. Two conflicting purposes are now presented for estimating the adequacy of the "free response" measure. One purpose might be further research into personality structures and theory formulation or modifications. In this context, the "free response measure" would be used most likely in conjunction with other measurement instruments and observations of behavior. This process is usually referred to as construct validation, but in fact, would more humbly be described as construct differentiation:or cl a r i f i c a t i o n . It is very This question may, of course, have been answered by others in a much more complete manner than this study could hope to do. Nevertheless, Harvey gives only scant information on the effectiveness of the TIB with high school students. (See Harvey, "Beliefs and Behavior: Some Implica-tions for Education", p. 73.) 64 much like observing a phenomenon within a chamber of mirros that give a multitude of reflections, a l l of which are partially distorted. The human mind confronted with these distortions and searching for the true image, make provisional corrections on the reflections u n t i l a uniform picture emerges. In keeping with the above analogy, this writer would judge the "free response" measure to be a very good "mirror" to provide a provisional observation of very fascinating concepts of personality functioning. This paper has shown in some detail the often remarkable sensitivity of the instrument to rather specific positions in the theoretical framework. In addition, one must take note once again that extensive work done with the instrument has given i t some status as a research tool. Another primary purpose of the instrument might be i t s use as an "institutional" tool to "observe" existing personality structures and to observe "results" of "desirable" changes. Is the instrument good enough for such a purpose? The answer here is much more complicated than in the former situation. In the f i r s t place, one now makes vastly greater claims not only for the instrument but also for the theoretical positions which i t is supposed to reflect. Secondly, the whole question is clouded by the implicit or explicit impositions of value judgements. Precisely, one claim that would have to be made is that the measure can be used for establishing a reliable and valid baseline from which one can analyse a personality structure and from which one can observe change. At this point one is already in the double bind of certain inadequacies of the instrument as well as i t s underlying theories. Any "free response" measure is inextricably dependent on the level of judgemental expertise, hence, practically impossible to standardize. The only possible absolute 65 standard would be the "perfect judge", the one that has "divine insights". By necessity, the second best thing is to attempt to calibrate "imperfect" judges by some index of agreement. But the annoying problem really i s that even the most precisely computed numerical indexes of interjudgemental agreements are never really comparable and in.some extreme cases even totally misleading. The truth of this statement can be seen by conceptual-izing two teams of judges "hired" to judge the same kind of phenomenon. One team receives a very high agreement index, the other team a low one. It could well be that the high agreement index of the f i r s t team results from a shared misconception, a hidden constant error; and the low agreement of the second team may result from greater cognizance of conflicting variables in the observation. It i s , of course, a reasonable assumption that having a greater number of judges w i l l reduce the likelihood of extreme misjudge-ments. Unfortunately, this road often is not available especially when one needs a specialized level of expertise. Although i t is completely outside the scope of this paper to eval-uate c r i t i c a l l y the theory of conceptual systems, i t must be noted that weaknesses of the measurement instrument are reflections of weaknesses in the theory. In practical "institutional" applications one would be immediately confronted with these. One of these weaknesses seems the insistence that differences in system functioning are primarily generated by differences in training, another one is the as yet unsolved problem of "generalizability" mentioned previously. For example, how does one account for incongruities of joint functioning of at least two totally different levels such as may be seen in the conceptual functioning of a physisist on the fore-front of his f i e l d who w i l l insist at a different level with utmost conviction that "mother love is the only love". 66 Will one trust the average "institution" (in most instances, a school) to use this very problematic instrument to generate "reforms" and observe "changes" within the framework of a highly complex but also problematic theory? Surprisingly, these are precisely the stated aims of the authors of 2 the "belief systems". This writer would argue that on technical grounds alone the risks of engaging in professional charlatanism are simply too high at the present stage of confidence i n measuring and interpretation of conceptual systems. The f i r s t problem is to find a measurement instrument which is standardizable, unproblematic to administer, and easy to score. Preferably, this measure should retain some direct relationship to the TIB rather than an indirect relationship such as the CST. The reasons are that (a) adjust-ments to cultural differences can be made more easily and directly and (b) the "meaning" of the test is more easily communicable. (It seems rather problematic to explain to a parent: your son was placed in conceptual level III because he had less than, say, 4.23 on the "divine fate control" factor and more than, say, 5.2 on the "general pessimism" (etc.) factor.) This writer contends that continued research for establishing such a measure is a worthwhile challenge. There certainly are more creative approaches than the one taken to generate the present form. But perhaps the writer's analysis of the "failings" w i l l be perceived as correct and, in turn, help to generate an approach to correctly locate system 1 and 2 functionings. The interested reader is especially advised to consult Harvey, Hunt, and Schroder, Conceptual Systems and Personality Organization, chap. 10; also David E. Hunt and Edmund V. Sullivan, Between Psychology  and Education, (Hinsdale, 111.: The Dryden Press, 1974), chaps. 11 and 12. In the last work cited the authors have taken a somewhat simplified theoretical approach - possibly with the idea that teachers either cannot understand the f u l l complexity of the original formulations or cannot apply these concepts in practical situations. 67 Even in i t s present form the instrument may not be as "disastrous" as i t appears on f i r s t sight. In the discussion of the "forced choice" instrument i t was argued that the items could be improved by looking at extemes in the distribution of responses for indicators of directions for improvement of items through content analysis. It may, of course, prove to be impossible to write a "true Harvey item", but with sufficiently large numbers some form of weighted probabilities could possibly be used to p u l l the "false" l's and 2's out of the 3's and 4's categorizations. It must be noted at this time that the observed discrepancies of out-comes between the "free response" and "forced choice" measures permit inter-pretations quite different from the one advanced in this paper. For example, had this study included measures of behavioral observations, one could, conceivably, have found these to correspond closer to the "forced choice" section than to the "free response" measure. This possibility poses an interpretive dilemma that could only be resolved by observational studies of greater scope which would have to include behavioral measures for c r i t e r i a of validity. It is suggested, however, that within the resources of the present study the given interpretations are reasonable because they are based on the explicit understanding that the greater status accorded to Harvey's "free response" measure derives from reported and, presumably verifiable connections between the "free response" measure and behavior. Other problems that must be solved are inconsistencies or inade-quacies in the theory. The solution i s most certainly not a rejection but a refining process which probably w i l l take generations of painstaking work. Closely aligned with the above problems is the necessity to generate a perception level in people, an awareness, that these belief system differences really operate and that i t truly may be desirable to opt for a change into "higher" conceptual levels. 68 This last problem leads headlong into a discussion of value problems from which this writer intends to refrain. Nevertheless, since everyone is functioning within a cultural setting, a few concluding remarks on this topic seem to be i n order. Milton Yinger has observed that i t is d i f f i c u l t 3 to maintain that "personality" is an abstraction. This paper, in common with a l l consulted sources, contains in one place or another a reification of concepts, i.e., the concept becomes the person; for example, system 1 functioning becomes "a system 1 person". This "temptation" is probably aided by language restriction. It i s interesting to observe how such re-ifica t i o n w i l l open different traps for different cultural perceptions. One cultural norm may be to "accept" everything that exists as i t exists, another may be to "understand" everything that exists, a third to "modify", and a fourth to "improve". The last mode is most identifiable with the North American syndrome that Piaget reportedly has called the "American fallacy" but might less politely have called the "American maleficence", i f he meant the extreme case of cultural doctrine that "everything that exists must be made better, faster." If this cultural doctrine exerts too strong an influence on us, the complexity of the concepts of the belief systems may well assume simplistic reifications which w i l l not help but hinder any perceptible improvements in educational practice. The reader is asked to at least look at the proposals for "conceptual level matching" to "foster" system 4 functioning from this vantage point of value impositions. In no way should the criticism above be allowed to distract from a third, possibly very valid, use of the "free response" measure. In the J. Milton Yinger, "Research Implications of the Field View of Personality", American Journal of Sociology 69 (1963): 580. Reprinted in The Sociology of Personality, ed. Stephen P. Spitzer (New York: Van Nostrand Reinhold Company, 1969) pp. 168-188. 69 hands of experts this measure should serve extremely well for analytical purposes, such as identifications of certain forms of psychopathology. But such a use demands training and a level of expertise well beyond that of typical teachers or school counsellors. Implications from Group Comparisons Interpretations are restricted by (a) non-randomness of the present sample and (b) confidence levels in correctness of categorizations. This writer believes that in the comparison of the frequency distributions of this sample to frequency distributions found in "1000"s" of l i b e r a l arts students, i t was " f a i r " to treat Harvey's data virtually as a parameter of that defined population. (Technically this i s , of course, completely true only i f he had measured a l l l i b e r a l art students in the defined population.) Having made this allowance, the writer hypothesized in the intro-duction certain expected differences that the present sample would show to the "Harvey population". These differences were perceived as resulting from interactions of supposedly (a) lower levels of "articulateness" in adole-scents, (b) greater flux in adolescents functioning, and (c) lower stages of conceptual developments than those that are expected at the age level of Harvey's populations. Because i t was shown (Table 18, p. 60) that judge-ments seemed neither particularly hindered nor helped by levels of "articu-lateness", i t is suggested to accept these differences as genuine. The question now is whether one wants to view these differences entirely as peculiarities of the non-randomness of the sample. This writer believes that these differences may well represent differences of high school students in general, but also believes that these differences w i l l tend to disappear as students get older. In fact, one may hypothesize that the differences can be explained entirely by the high number of admixtures. 70 These could be viewed as oscillating or unstable stages of which a certain number w i l l settle in time into systems 2 and mainly 3, and in rare instances into stage 4 functioning. Why not into stage 1? Because concept differen-tiations are seen as already too advanced in admixtures for a "regression" to stage 1. Further investigations of distribution patterns of true random samples would be the means by which one could investigate the theoretical development aspects of the theory. Suppose, one draws many different random samples in different populations and cultures for which different patterns of childrearing are concurrently investigated. Suppose further that one measures system functioning with a standardized instrument that has the f l e x i b i l i t y to sort our "cultural norms" from individual differences. If one would find that the distribution patterns of system functioning remains essentially similar across a l l cultures and a l l patterns of childrearing, one would have to come to the conclusion that distributions of system functioning is "fixed", i.e., independent of childrearing. This, or contrary findings, would have enormously important implications for the theoretical assumptions that higher conceptual functioning can be "induced". Admittedly, one can only hope for a gradual c l a r i f i c a t i o n of this problem because the research task is also enormous. It is suggested that this question alone would be worthy of many research efforts. One internal comparison that showed a definite contradiction to what one would expect from Harvey's reported, validations, is the finding that levels of "articulateness" are much higher for system 3 and 4 function-ings than for system 1 and 2 functioning, not only in respect to each other but also in respect to the expectations that levels of "articulateness" would be distributed "normally" within each of the systems. This form of 71 comparison allows the qualification that while the distribution of levels of articulateness of system 1 and 2 functioning is not markedly different from what one would expect, system 3 and 4 functioning does not even seem to occur in "low" levels of "articulateness". One possible explanation is that Harvey's finding was more "impressionistic", and that a more "methodic" rating would show similar differences. If these differences were upheld they might prove to be co-defining variables of belief system functioning in the same sense that "intelligence" is seen as a co-defining variable of "creativity". It has often been said in studies of creativity that a certain level of " i n t e l l i -gence" is a necessary but not sufficient prerequisite for creative function-ing. Possibly a certain level of "symbolic wealth" may be a threshold below which system 3 and 4 functionings cannot occur. Clarification of these questions would seem to offer many interesting research problems. Not' discussed in the preceding section was an internal difference of response patterns in relation to shifting patterns of "norms" of pre-judice. This writer does not really know how to define this "norm" but i t is related conceptually to "public opinion". One is aware of these shifts without a clear quantitative index. For example, one is aware of what seems to be a much lower level of public tolerance towards immigrants in Canada than, say, 20 years ago. Similarly, one is aware of the fact that women are now accepted much more readily as equal to men (at least in a kind of stereotype fashion) than even five years ago. This awareness i s , for example, nicely expressed in the response to " g i r l s " in the transcription (M: p...35 this paper). Girls: "A l i t t l e while ago this would not be considered proper. But I am a l l for i t , personally I can't see myself, but i f that's what some girls want to do that i s their business." 72 These differences are clearly reflected in the overall response differences to "immigrants" and " g i r l s " . One can see from the data in Table 8, p. 37, that only eight times were " g i r l s " completely rejected (always by boys), but 47 times were "immigrants" completely rejected (15 times with strong ra c i a l overtones). It can be seen that not a single 3 or 4 categoriza-tion is associated with rejection of " g i r l s " and only two with rejections of "immigrants". This may mean that system 3 or 4 functioning is dependent neither on "norms" of prejudice nor on "changes" in these norms. Conversely, one must ask, is i t true that the entire range of prejudice (public opinion) is the domain of system l's and 2's functionings. These questions are posed here because they are in need of much clearer formulation, and they do have a bearing on the theory of belief systems as well as on problems of measure-ment of the systems. What could be the most relevant immediate contribution to education from the theory and findings of research with belief system functioning? This writer would simply propose to introduce these concepts in teacher training. If a l l aspiring teachers could be made aware that they enter a rather complez "business", we may in time progress from our depressingly simplistic and polarized cyclic "reforms" or "solutions" to educational problems. The analysis of the "faults" of institutionalized aims of education seem to this writer most aptly expressed in the concluding chapter on p. 342 or the authors of the belief systems: "Current educational practices reward stage 1 or 3 functionings by their emphasis either upon memorization and inflexible accretion of facts or upon successful interpersonal relationships ..." But certainly the teacher who is caught between his own limitations and his concerns for his day to day survival in an institutional setting, should not be expected to carry the banner of educational reforms with tools that are, at this moment, not precisely formulated and understood - let 73 alone communicable to the public. What must emerge f i r s t is a crystalized understanding of relationships and causations of system functionings and a standardized way of measuring these. "If" i t can be shown that system functioning is truly changeable then the public could be presented with clearly communicable choices - after a l l , the public must make this ultimate choice, not educational researchers. 74 WORKS CITED Cohen, J., "Weighted Kappa: Nominal Scale Agreement with Provisions for Scaled Disagreement or Partial Credit." Psychological  Bulletin, 70 (1968): 213-220. Cohen, J. , "A Coefficient of Agreement for Nominal Scales." Educational  and Psychological Measurement, 20 (1960) : 37-46. Harvey, 0. J., "Beliefs and Behavior: Some Implications for Education." The Science Teacher, 37 (December 1970): 10. Harvey, O. J.; Hunt, David E., and Schroder, Harold M. Conceptual Systems  and Personality Organization. New York: John Wiley & Sons, Inc., 1961. Harvey, 0. J., "Belief Systems and Education: Some Implications for Change", in The Affective Domain, pp. 67-96. Edited by Jack Crawford. Wash., D. C. : Communication Service Corporation, 1970. Harvey, O. J., "Conceptual Systems and Attitude Change", in Attitude, Ego- Involvement and Change, pp. 201-226. Edited by Carolyn W. Sherif and Muzafer Sherif. New York: John Wiley & Sons, Inc., 1967. Harvey, 0. J., and Felknor, Catherine, "Parent - Child Relations as an Antecedent to Conceptual Functioning", in Early Experience and the  Processes of Socialization, pp. 167-203. Edited by G. A. Milton, R. A. Hoppe, and E. C. Simmel. New York: Academic Press, 1970. Harvey, O. J., 'Scoring the "This I Believe" (TIB) Test', manuscript on f i l e , O. J. Harvey, University of Colorado, 1965. Harvey, O. J., This I Believe Test (Form TIB-74), Copyright 1974, O. J. Harvey. Harvey, O. J., "System Structure, F l e x i b i l i t y and Creativity", in Experience,  Structure and Adaptability, pp. 39-65. New York: Springer Publish-ing Company, 1966. Harvey, O. J., and Hoffmeister, James K., Conceptual Systems Test Form CST-71, Copyright 1971, O. J. Harvey and K. Hoffmeister. Hunt, David E., and Sullivan, Edmund V. Between Psychology and Education. Hinsdale, I l l i n o i s : The Dryden Press, 1974. Light, Richard J., "Measures of Response Agreement for Qualitative Data: Some Generalizations and Alternatives." Psychological Bulletin, 76 (1971): 365-377. 75 Miller, Alma Grabow, and Harvey, 0. J., "Effects of Concreteness - Abstract-ness and Anxiety on Intellectual and Motor Performance". Journal of  Consulting and C l i n i c a l Psychology, 40 (1973) : 444-451. Yinger, Milton J., "Research Implications of a Field View of Personality". American Journal of Sociology, 68 (1963) : 580-592. Reprinted in The Sociology of Personality. Edited by Stephan P. Spitzer. New York: Van Nostrand Reinhold Company, 1969, pp. 168-188. APPENDIX A Judgemental Strategy and Rationale for Strategy, Display, of Two Agreement Indexes and  Summary of Agreement Indexes  in Percentage Forms Mathematical models measuring agreement indexes in nominal scales assume a completely independent functioning of judges."*" This study did not have the resources to c a l l on experts to categorize the responses from the sample. Hence, a "training" process was incorporated with the aim to pro-duce the best possible judgemental categorizations under the given circumstances. The strategy adopted was as follows: both judges read the scoring booklet of Harvey, made independent categorizations from part of the sample, discussed and defended their respective categorizations, and then allowed the matter to rest. After two months both judges re-read Harvey's scoring booklet, discussed the key, and agreed on a slight modifica-tion of the key. Thereafter,'both judges categorized a l l responses independently of each other, then had a "panel" discussion in order to categorize disagreements. Between six and ten days later both judges once """Jacob Cohen, "A Coefficient of Agreement for Nominal Scales", Educational and Psychological Measurement, 20 (1960): p. 38. The follow-ing assumptions are specified. 1. The units (test protocols) are independent. 2. The categories of the nominal scale are independent, mutually exclusive, and exhaustive. 3. The judges operate independently. Assumption 2 probably is the most d i f f i c u l t to f u l f i l l in practice, especially when many partially overlapping variables (as in the present study) define a category. 77 again categorized a l l responses ( t r i a l 2) and, f i n a l l y , categorized dis-agreements. Each categorized response was now assigned a "confidence score" of either "high", "med.", or "low" in accordance with the key explained in Table 23. But the meaning of these "confidence scores" can most readily be assessed by looking at Table 21. The key for reading this table is the same as the key for Table 11 in the text: For convenience i t is reproduced here. (1) Ful l agreements are shown in the narrow zigzag band running from top l e f t to bottom right. (2) "First term" agreements (such as 1-2 and 1; or 1-2 and 1-4) are shown in the double-lined wider zigzag band. (3) Agree-ment between "second and f i r s t terms" (such as 1-2 and 2; or 1-2 and 2-4) f a l l within the narrow bands of crosses. (4) Complete "reversed term" agreements f a l l into the intersection of the crosses. (5) Agreements on "second term" alone have been circled. Also, wherever a score f a l l s into any !'first term" or "second term" X position at least a minimal part agree-ment exists. Table 22 shows the agreements on categorizations between the two 3 "panel" discussions. 2 Jacob Cohen, "Weighted Kappa: Nominal Scale of Agreement with Provisions for Scaled Disagreement or Partial Credit", Psychological Bulletin, 70 (1968): 213-220. In this article Jacob Cohen modifies and expands his earlier ideas on agreement measures. He gives several formulas for agreement indexes that are chance corrected and allow "partial credit" for "partial agreements". For a further in-depth study of these problems the reader i s referred to Richard J. Light, "Measures of Response Agreement for Qualitative Data: Some Generalizations and Alternatives", Psychological Bulletin, 76 (1971): 365-377. In the absence of some "rule of thumb" figures of what constitutes satisfactory levels of agreement these indexes may be quite d i f f i c u l t to interpret. 3 "Agreements" in this discussion are in Cohen's terminology only "associations". Thus for example, in Table 21 the 31 "associations" of judgemental positions would have to be devaluated by chance associations before one could properly speak of "agreements". In this example there are 31 system l ' s , but 52 o r l x 48 = 20.6 of the 31 would have to be considered "chance associations".. 78 TABLE 21 NUMBER AND EXTENT OF AGREEMENTS ON CATEGORIZATIONS OF "FREE RESPONSES" BETWEEN JUDGE I AND JUDGE II TOTAL SAMPLE N=122 JUDGE I (RUN 2) 1 Dominant lsts 1 1 1 2 3 4 1 X 2 Dominant 2nds 2 2 2 1 3 4 2 X 3 Dominant 3rds 3 3 3 1 2 4 3 X 4 Dominant 4ths 4 4 4 1 2 3 4 X X ! N rH 31 3 4 4 1 1 1 o 1 o 1 1 48 -P g H <N CO U l 4 1 1 2 2 1 10 C -P •g W rH 00 4 1 o 1 o j 6 O Q rH 2 1 o © 1 ! 5 rH X 4 1 0 o 0 1 . 7 C N 2 5 1 8 -p C CM rH rd w 1 3 o o o 4 C •H C CN CO 6 C N o 1 1 2 , o ' Q C N o ( o ( 1 1 C N X 1 3 1 1 0 o 3 ro 2 1 1 3 o 7 •p C n H Cu w 2 3 1 o 6 C t3 •H U rn (M 6 C O 0 0Q 00 ^ o o 1 1 co X 0 0 o 2 5 7 -P C rH CO w 1 o o o 1 2 c ,C -H +J ^ C N g O 0 Q ^ co o o 1 1 x 1 o 3 1 2 X 1 1 52 6 9 6 3 13 1 1 1 5 8 1 10 1 1 2 1 121 NOTE: The reduction to 121 observations was caused by one respondent fa i l i n g to f i l l in the "free response" section. 79 TABLE 22 NUMBER AND EXTENT OF AGREEMENTS ON CATEGORIZATIONS FROM "FREE RESPONSES" BETWEEN 1ST PANEL ADJUSTMENTS AND 2ND PANEL ADJUSTMENTS. TOTAL SAMPLE N=122 1ST PANEL ADJUSTMENTS Systems 1 & Dominant 1's 1 1 1 1 2 3 4 X Systems 2 & Dominant 2's 2 2 2 2 1 3 4 X Systems 3 & Dominant 3 1s 3 3 3 3 1 2 4 X Systems 4 Dominant 4 4 4 4 1 2 3 s 's 4 X X N r—1 CO 39 3 2 1 2 2 1 o 50 <z -r H rH t N r H 2 2 2 1 o 2 9 +J 1 1 1 o 1 2 1 o 1 8 4 J H H co 6 4 o o 1 1 6 > i o CO Q H X 4 o o 5 CN CO 7 1 1 o o 9 ca -CN CN '—1 CN 2 2 1 5 +J g g CN ro o o +1 - H ! N • * o ( o CO £5 >i o CO Q CN X 1 D 1 o o 1 3 ro CO 5 o 5 oS -ro ro r H ro 1 1. 1 o 1 4 •P g g ro CN o <D C +J - H ro ^ o 0 1 1 CO c > i 0 co Q ro X 0 o o CO o 5 1 1 7 03 -<J <J H 1 o o 1 1 3 •P 1 § •*CN- O 1 1 +J - H ro co B o o 1 1 > i 0 CO Q ^ X c 0 X 1 1 2 4 2_ 50 7 4 4 3 11 3 1 2 4 8 2 9 1 1 3 8 121 CO E-i 2 E-t co D § rH w 2 *C P 4 Q 2 CN NOTE: The reduction to 121 observations was caused by one respondent fa i l i n g to f i l l in the "free response" section. 80 The rationale for the strategy of alternating between independent and interdependent judgemental procedures can be explained by conceptualizing the "perfect judge". , A perfect judge would not make any variable errors or constant errors. A l l categorizations by a perfect judge would be indisput-able and, hence, a l l responses that he could not categorize would be due to lack of response information or to truly conflicting response informations. The point is that some of the judgemental inconsistencies in this study must be explained with other than variable error and constant error. The human judge w i l l tend to reach beyond the limitations of the perfect judge; s t i l l giving a "reading" when the perfect judge would simply say, "not enough information". As one can see in Table 22, even between the two panel discussions there are very few agreements on admixtures. In other words, the "free response measure" is probably truly incapable of yielding sufficiently clear information for categorizing admixtures in a much more stable manner than in this study. It was thought,. however, that the constant errors would be reduce-able through the interaction strategy just described. There are several particular d i f f i c u l t i e s in this judgemental task which can easily produce constant errors. For example, "expressed h o s t i l i t y " can very easily be mistaken for system 2 instead for system 1 functioning i f this expressed ho s t i l i t y i s not carefully viewed within other aspects of the response patterns. These errors would only be recognized through discussions of any disagreements.and renewed reflections on the specified variables. Unfortunately, there also comes a point when judges begin to share biases and "learning" becomes counter-productive. It is not easy to estimate when this point is reached. Variable errors caused by fatigue or momentary confusions are probably not much reduceable through learning. Judges can only make certain 81 to take as much care as possible and to allow themselves sufficient time. The judges for this study f e l t that they could do not many more than about 30 categorizations in one day - before experiencing extreme signs of fatigue. These errors show up in some rather gross discrepancies that probably un-fa i r l y reduce the agreement index, because "better" judges would not likely make those mistakes. In Table 21, for example, there are three disagree-ments about system 1 or system 4 functioning. These disagreements do not exist in the two panel discussions. This writer judges the figures in Table 21 as probably underestimating, and the figures in Table 22 as probably overestimating, the possible agreements of two "good" judges functioning independently. Table 23, for comparative purposes, summarizes agreement indexes of simple percentage counts that were attained in this study. TABLE 23 INDEXES OF JUDGEMENTAL AGREEMENTS ^ ^ INTRAJUDGEMENTAL INTERJUDGEMENTAL U U EH H Low Med. High (Trial 1) Low Med. High JI to JI 95 79 49 JI to JII 88 82 46 (Trial 2) JII to JII 83 66 35 JI to JII 92 89 39 Panel (1) to Panel (2) 100 81 54 JI (Trial 2) to Panel (1) 98 79 55 JII (Trial 2) to Panel (1) 94 71 43 JI (Trial 2) to Panel (2) 100 85 66 JII (Trial 2) to Panel (2) 97 81 55 NOTE: JI means Judge I and JII means Judge II. "Low" means here partial agreement which can range from second term with f i r s t term agreements (such as 4 and 1-4) to the lowest possible agreements, through use of designations (such as 4- and 1). "Med." means f i r s t term or reversed agreement (as 1 and 1-2; and 1-2 and .2-1) , and "High" means f u l l agreements. 82 APPENDIX B A Complete Quotation of "System Response Patterns"  From 0. J. Harvey's 'Scoring the "This I Believe"  (TIB) Test', pp. 9-12. System Response Patterns: System 1; Characterized, according to theoretical notions, by a strong need for structure, r i g i d adherence to rules, authorities and values which provide structure; and rejection of environmental inputs which are dissonant with the individual's organized modes of interpretation. TIB responses tend to be stated in a definite, hard-and-fast manner, showing l i t t l e doubt in the subject's mind about how he feels. The content shows adherence to norms and practices approved by society or prestige authorities; a negative reaction to rule-breaking; polarized evaluations. The reliance on authority is demonstrated by highly favorable attitudes toward religion, law, parents, friends. Other people must meet r i g i d standards of acceptability, operating in terms of the general behavioral principles of the subject. Religion is a highly consistent concern, serving as a base for the belief system in a l l aspects of l i f e in many cases. This referent tends to e l i c i t the most clear-cut System 1 responses of a l l the referents. System l's often demonstrate strong ingroup-outgroup feelings ex-pressed in intense h o s t i l i t y and negative feelings on some referents. In order to avoid confusion with System 2 responses that are similar because of their negativity i t is necessary to evaluate such responses in the total context of .all the responses. The overall impression of System l's is that of a person who has definite stands on every topic, states them evaluatively and unequivocally, and rejects things i f they do not meet his high standards of ideals of per-fection. The reader may feel that this subject is rather hostile toward his environment and other people, but there i s an underlying sense of stable acceptance of things as they should (ought, must, etc.) be by his standards. The words, "everything," " a l l , " "completely," "best," "worst," etc., are a l l words that indicate the extreme, clear-cut, definite aspects of existence as this person sees i t . Uncertainty is anathema to a System 1 and both content and structure in his responses demonstrate his drive to reject i t and find and maintain certainty in his environment. System 2: Characterized by terms highly similar to those of System 1 except for a reversal of certain central aspects of content. The structur-al aspects are similar and the responses indicate r i g i d i t y , simplicity, consistency and exclusivity. 83 This subject has the same drive for certainty as the System 1, but seems unable to rely on his world to find i t . Hence, he seems to obtain certainty by rejecting his world; as though negating i t provides his only source of certainty. The reader w i l l find a rejection of or hostile attitude toward authority referents, i d e a l i s t i c notions, most American standards and values, and most other people. Not a l l people are rejected, since this subject makes positive statements about the underdog, the loner, minority groups and individuality. Conversely, he makes negative statements about elements that might do harm to these people. There is a strong rejection of religion, people, friendship (because of i t s unrel i a b i l i t y ) , government, and, more subtly, ties and obligations and other freedom-retraining devices. This subject reacts negatively to these ideas, yet cannot ignore them. He tends to be factual and hard-nosed rather than i d e a l i s t i c about the world, requiring a similar need for structure observed in the System 1 person. Overall, the System 2 person appears extremely hostile and rejecting, concrete minded, non-analytic toward his environment, and a categorical acceptor or rejector in terms of pre-established negativity. System 3: The chief locus of satisfaction for a System 3 tends to be his relation with other people. His responses reflect the central importance of people and he accepts and voices the values of the people with whom he is in contact at the moment rather than i n i t i a t i n g behaviors or expressing beliefs that are contrary to the present group. TIB responses generally lack any expression of negative feelings. There-is a strong tendency to deal with the world through rather superficial aspects of i t , expressed in the use of cliches rather than directly. Relationships with people are e l i c i t e d even when TIB referents do not necessarily c a l l for them. Generally, the only negative reaction w i l l be to a referent indicating harm or injury to other people. The responses of these subjects are more complex, varied and abstract than Systems 1 and 2. These people are typically rather sophisticated in dealing with their world and do not demonstrate a hard-and-fast r i g i d i t y in responses to the referents. Though the repeated emphasis on interpersonal relations may at f i r s t appear to be a r i g i d response tendency, analysis shows a great deal of f l e x i b i l i t y and openness in the responses. Thus, i t is necessary to analyze the structural aspects of a System 3's responses, while remaining sensitive to the evidence of person-oriented content. The overall impression generated by the responses of System 3 persons is one of a positive attitude toward situations and ideals which are beneficial for people. System 4: Characterized by relative independence from the environ-ment, greater r e l i a b i l i t y on internally-derived stimulation, greater flex-i b i l i t y and openness, interest in (even seeking for) novelty, a relative lack of extreme evaluativeness or extreme acceptance-rejection behavior, the tendency to be aware of and to respond to referents in terms of multiple alternatives or interpretations. 84 TIB responses show a juxtaposition of diverse, often contrasting elements. There is a lack of one-way evaluativeness; certainty and definite-ness in commitment.to a single way of perceiving a situation are typically not evident in these subjects. The overall impression of a System 4 person is one of Complexity of thought and feeling. Depth of connotative implications rather than super-f i c i a l i t y in statements is most typical of these S's. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0094471/manifest

Comment

Related Items