Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Complexity, moral reasoning, and attitudes toward capital punishment De Vries, Brian 1984

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

Item Metadata

Download

Media
831-UBC_1984_A8 D49.pdf [ 4.36MB ]
Metadata
JSON: 831-1.0096288.json
JSON-LD: 831-1.0096288-ld.json
RDF/XML (Pretty): 831-1.0096288-rdf.xml
RDF/JSON: 831-1.0096288-rdf.json
Turtle: 831-1.0096288-turtle.txt
N-Triples: 831-1.0096288-rdf-ntriples.txt
Original Record: 831-1.0096288-source.json
Full Text
831-1.0096288-fulltext.txt
Citation
831-1.0096288.ris

Full Text

COMPLEXITY, MORAL REASONING, AND ATTITUDES TOWARD CAPITAL PUNISHMENT by BRIAN DE VRIES B.A., The University of B r i t i s h Columbia, 1980 A THESIS SUBMITTED IN.PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1984 © Brian de Vries, 1984 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department Of P s y c h o l o g y  The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date 26 S e p t e m b e r 1984 DE-6 (3/81) i i ABSTRACT This study examined several interrelated issues in s o c i a l -developmental psychology: (a) the relationship between conceptual and integrative complexity; (b) the relationship between conceptual complexity and moral reasoning; (c) the relationship between attitudes toward c a p i t a l punishment and moral reasoning; (d) the relationship between attitudes toward c a p i t a l punishment and conceptual and integrative complexity; and (e) the relationship between the levels of moral reasoning that subjects use to substantiate their own, versus an opposing, position on a moral problem ( i . e . , c a p i t a l punishment). Participants were 72 university students (from f i r s t - y e a r to graduate school) who completed the Paragraph Completion Test (assessing conceptual complexity), wrote an issue composition on c a p i t a l punishment (assessing integrative complexity and moral reasoning), generated statements supporting and opposing c a p i t a l punishment (assessing moral reasoning), and responded to a c a p i t a l punishment attitude questionnaire. Results indicated that the two measures of complexity are comparable--but not equivalent--assessments of the cognitive structures in information processing, and that conceptual complexity and moral reasoning are moderately related. Integrative complexity and attitudes toward c a p i t a l punishment related in a c u r v i l i n e a r manner with extreme attitudes (both pro and con) characterizing conceptual si m p l i c i t y and moderate attitudes characterizing increasing complexity. Moral reasoning and attitudes toward c a p i t a l punishment related in a linear manner with opposition to c a p i t a l punishment increasing with moral stage. Furthermore, subjects did not always use higher moral reasoning to substantiate their own attitude or position on this issue, rather they used higher moral reasoning when opposing i t , regardless of their chosen po s i t i o n . The implications of these findings for the theories of moral reasoning and conceptual and integrative complexity are discussed and suggestions are advanced for the direction of future research. i v TABLE OF CONTENTS Abstract i i L i s t of Tables v i L i s t of Figures v i i Acknowledgements v i i i Introduction 1 Moral Reasoning 2 Conceptual and Integrative Complexity 9 Attitudes 15 Moral Reasoning and Conceptual Complexity 20 Moral Reasoning and Capital Punishment 23 Conceptual Complexity and Capital Punishment 29 Hypotheses 34 Method 36 Subjects 36 Measures 36 Paragraph Completion Test 36 Capital Punishment Issue Composition 37 Statements Supporting and Opposing Capital Punishment ....37 Capital Punishment Questionnaire 38 Scoring 38 Conceptual Complexity 38 Integrative Complexity 40 Moral Reasoning 42 Attitudes toward Capital Punishment 46 V Results 47 Normative Data 47 Conceptual and Integrative Complexity 48 Complexity and Attitudes toward Capital Punishment 51 Conceptual Complexity and Moral Reasoning 55 Moral Reasoning (Pro versus Con) and Attitudes toward Capital Punishment 61 Discussion 68 Conceptual and Integrative Complexity 68 Complexity and Attitudes toward Capital Punishment 71 Conceptual Complexity and Moral Reasoning 75 Moral Reasoning (Pro versus Con) and Attitudes toward Capital Punishment 78 Conclusion 83 References 86 Footnote 95 Appendix A - Test Materials 96 v i LIST OF TABLES 1. Median S p l i t of Scores on the PCT and Issue Composition ...50 2. Summary of the Analysis of Variance for the Mean Complexity Scores 52 3. Summary of the Analysis of Variance for the Mean Complexity Scores, Including Simple Effects and Trend Analysis 53 4. Summary of the Analysis of Variance for Total Weighted Average Scores 59 5. Median S p l i t of Scores on the PCT and Moral Reasoning 60 6. Summary of the Analysis of Variance for Pro and Con Weighted Average Scores, Including Simple Effects 63 7. Mean Weighted Average Scores for Pro and Con Positions as a Function of Attitude toward Capital Punishment 64 8. Percentage of Subjects at each Moral Stage Who Oppose Capital Punishment 67 LIST OF FIGURES v i i 1. Mean Complexity Scores on the Issue Composition as a Function of Attitude toward Capital Punishment 56 v i i i ACKNOWLEDGEMENTS I would l i k e to express my appreciation to many people for the roles they have played in f a c i l i t a t i n g the completion of this thesis. F i r s t , and foremost, I would l i k e to thank my supervisor and friend, Dr. Larry Walker. His unwavering academic and emotional support and guidance throughout t h i s project consistently demonstrated how invaluable a good supervisor and a good friend can be. I also wish to thank my committee members: Dr. Peter Suedfeld for his helpful suggestions, expert knowledge and his willingness to share i t ; and Dr. Bob Knox for his i n s i g h t f u l comments, careful analysis and valuable assistance. My gratitude is also extended to Susan Bluck for her assistance in the scoring of test materials. I would also l i k e to express my gratitude to my parents, family, and friends (especially Debbie, Mick, Alma, Bruce, and Pat) who bestowed upon me a seemingly unconditional confidence in my a b i l i t y at times when I doubted i t most. The graduate students of the Psychology Annex--a term now used to locate individuals in emotional, as opposed to physical, space--also provided me with unending encouragement and moral support, for which I w i l l always be g r a t e f u l . 1 Introduct ion The cognitive-developmental approach to morality (Kohlberg, 1973, 1976) posits stages of moral reasoning which are held to develop in an invariant and h i e r a r c h i c a l sequence, each stage representing a more complex structure. Complexity, in this sense, is the a b i l i t y to d i f f e r e n t i a t e (referring to the number of categories or kinds of information a person i s able to process in a given situation) and integrate (referring to the development of complex connections among these d i f f e r e n t i a t e d c h a r a c t e r i s t i c s ) . These processes are hallmarks of the information processing theory of conceptual complexity, as proposed by Schroder and others (Harvey, Hunt, & Schroder, 1961; Schroder, Driver, & Streufert, 1967), and as such form an interesting and empirically v e r i f i a b l e p a r a l l e l between these two theories. The issue of c a p i t a l punishment has been, and remains, a controversial topic and one that has been researched within the domain of attitudes (an information processing structure relevant to measures of complexity) as well as moral reasoning (as i t is included in Kohlberg's structured interview and scoring manual, Colby et a l . , in press). As such, i t forms the connective tissue between the above two theories and presents some interesting hypotheses to investigate. The following sections present brief theoretical and methodological accounts of the areas of moral reasoning, conceptual and integrative complexity, and attitudes in general, and an integration of these three areas with the hypotheses that such interactions generate. 2 Moral Reasoning Kohlberg (1969, 1973, 1976) has presented an approach to moral reasoning that can best be characterized as cognitive-developmental. Aligned with the general c h a r a c t e r i s t i c s of cognitive stages as i n i t i a l l y presented by Piaget (i960), Kohlberg believes that development proceeds through a series of q u a l i t a t i v e l y d i s t i n c t stages, that the structure of thought can be distinguished from i t s content, and that the developmental stages occur in an invariant, h i e r a r c h i c a l , and universal sequence. This approach is said to provide a v a l i d conceptual and methodological framework within which moral reasoning can be developed without imposition of s p e c i f i c values. That i s , higher stages of moral reasoning are predicated on the types of reasoning individuals provide as opposed to the values to which individuals adhere. Kohlberg claims that morality should not be defined in terms of conformity with the preva i l i n g group norm, for i t remains a philosophical rather than a behavioral concept; a person's morality cannot be assessed without knowing that person's point of view and intentions. Kohlberg maintains that behavior has an underlying structure; i t i s not an aggregate of disconnected responses triggered by external s t i m u l i . Nevertheless, Kohlberg (1978) has discussed the necessity and d e s i r a b i l i t y of going beyond "cognitions" to the realm of behavior and towards such an end, Blasi (1980) has written a review a r t i c l e describing the relevant research in this area. Before describing the Kohlbergian stage sequence, i t would be instructive to examine more f u l l y the nature and 3 c h a r a c t e r i s t i c s of cognitive-developmental stages (Walker, in press). F i r s t , there are d i s t i n c t q u a l i t a t i v e differences between dif f e r e n t stages of development; there is a logic underlying these q u a l i t a t i v e differences. That i s , the hierarchy follows a l o g i c a l sequence in which each successive stage builds upon and elaborates on the preceding one. Second, each individual follows the same invariant sequence of stages. Cultural factors and the nature of one's s o c i a l environment may cause the rate of development to vary or cause development to cease, but are held not to change the sequence of development. Third, each stage represents a structured whole. A response r e f l e c t s the way in which an individual organizes thoughts: the structure of his/her reasoning. Individuals should be r e l a t i v e l y consistent in their reasoning across varying contents and contexts. Fourth, the stages are seen as hi e r a r c h i c a l integrations in that each subsequent stage is both more di f f e r e n t i a t e d (there are more parts that are more complex) and more integrated (the parts are better organized and capable of being combined in new ways). Of c r i t i c a l importance to the stage notion of moral development is the demonstration that the stages r e a l l y represent an invariant sequence. Kohlberg has addressed this in his longitudinal study (Colby, Kohlberg, Gibbs, & Lieberman, 1983) reporting 6% regressions of stages and no stage-skipping, which is less than the level of measurement error. Walker (1982) has also investigated t h i s claim by attempting to induce experimentally regression and stage~skipping--both violations of an invariance claim. Subjects were exposed in a brief role-4 playing situation to one of the treatment conditions which entailed the presentation of reasoning statements at, below, or above the stage of reasoning demonstrated by the subjects during a pretest interview. Results supported the sequentiality claim as development was always to the next higher stage, regardless of stage to which subjects were exposed. Equally important to thi s stage notion of moral development is the demonstration that the logic of each stage forms a structured whole. In line with this assumption, Colby et a l . (1983) suggested that one would expect to find a high degree of internal consistency in the assigned stage scores. This internal consistency may be noted through a d i s t r i b u t i o n analysis, for each subject, of proportion of reasoning at each of the five stages, where one would expect that the majority of subjects would receive a l l of their scores at either a single stage or at two adjacent stages. In fact, t h i s i s what Colby et a l . (1983) have reported. The mean percentage of reasoning at individuals' modal stage was 67% for the three alternate forms of the interview combined and the mean percentage of reasoning at the two most frequently used stages (which were always adjacent) was 99% for the three forms combined. Further support for claims of structured whole comes from a factor analytic study also reported in Colby et a l . (1983). They found that no more than one interpretable factor emerged, leading them to conclude that moral judgment is a single, general domain. Empirical support may also be found for the cognitive-developmental h i e r a r c h i c a l claim of thi s moral stage theory. Walker, de Vries, and Bichard (in press) extended and improved 5 upon previous research (e.g., Rest, 1973) demonstrating that subjects prefer, but f a i l to understand, higher stage reasoning. Walker et a l . assessed evaluation and understanding of moral stage prototypic statements which had been equated for d i f f i c u l t y of language and found that subjects displayed a h i e r a r c h i c a l pattern of responding: understanding of stage prototypic statements was cumulative and higher stage statements were preferred over lower stage statements, i f subjects were capable of recognizing the difference. Kohlberg's theory postulates three basic levels of reasoning about moral issues within each of which are embedded two stages. The i n i t i a l study which led to the formulation of the stages of moral development was begun in the 1950's in suburban Chicago, in which groups of boys at d i f f e r e n t age levels were administered lengthy interviews about moral issues. (For example, should Heinz steal an overpriced drug to save his wife from dying? Should Heinz be punished for stealing?) They were subsequently reinterviewed every 3 or 4 years u n t i l they reached middle adulthood. The stages of moral reasoning that were derived from these interviews can be described as follows (complete stage descriptions are found in Kohlberg, 1981, and in the scoring manual, Colby et a l . , in press). The preconventional l e v e l is characterized by a be l i e f that what is right i s limited to following concrete rules backed by power and punishment. When not prohibited, right behavior i s that which serves one's own interests, or the interests of some other person. The reasons for upholding what i s right include considerations of avoiding punishment, deference to power, 6 serving s e l f - i n t e r e s t s , and/or exchanging favors. The so c i a l perspective here is predominantly centered around the self and/or physical dimensions or consequences of rules and behaviors. Stage 1 is an orientation of punishment and obedience in which the physical consequences of an action and the dictates of authority define what is considered right and wrong. Stage 2 is an orientation of individualism, instrumental purpose, and exchange where right is defined as serving one's own interests and desires and cooperative interaction is based on terms of simple exchange. Reasoning at these stages is most t y p i c a l l y found among children, some adolescents, and r e l a t i v e l y few adults. At the conventional l e v e l , the "right" i s that which conforms to the expectations concerning "good" behavior of the larger society or some smaller segment of i t . One i s concerned with upholding so c i e t a l rules, expectations, and roles. One should act properly in these regards, not out of concern for punishment or s e l f - i n t e r e s t , as the previous l e v e l , but because of an inner motivation to do what i s approved of and expected by society. The reasoner i s concerned with s o c i a l opinion, l o y a l t y , and receiving the approval of others. In terms of s o c i a l perspective, the highly egoistic orientation of the e a r l i e r l e v e l i s now subordinated to the views and opinions of the larger s o c i a l group of which one is a member. Stage 3 e n t a i l s mutual interpersonal expectations, relationships and conformity and is one in which the emphasis i s on good person stereotypes and a concern for approval. Stage 4 — s o c i a l systems and conscience maintenance--focuses on obeying the law and doing 7 one's duty in an e f f o r t to maintain s o c i a l order, and following rules which are needed for the larger society as a whole. Reasoning at these stages i s most c h a r a c t e r i s t i c of adults, but also found in some adolescents. At the postconventional or pr i n c i p l e d l e v e l , right i s defined by general or universal human ri g h t s , values, or pr i n c i p l e s which both the society and the individual are obligated to uphold. Rules and laws are perceived as necessary for the functioning of society and to guarantee j u s t i c e , but one must also recognize their a r b i t r a r y nature and see that their v a l i d i t y l i e s in their acceptance by the members of society. The orientation is not to rules in and of themselves, but to the pr i n c i p l e s and purposes behind them. The s o c i a l perspective of the p r i n c i p l e d l e v e l individual i s one which precedes society and attachments to i t . Social practice should be based upon morality, rather than morality deriving from s o c i a l practices. Stage 5 supports s o c i a l contract or u t i l i t y and individual rights, and what is considered right is to uphold basic rights, values and the mutually agreed upon contracts of a society, even when doing so might be in c o n f l i c t with certain concrete rules and laws of the soc i a l group. Stage 5 reasoning i s the designated morality of the U.S. Constitution based upon r a t i o n a l l y chosen p r i n c i p l e s of justice maximizing the welfare of the society. Stage 6 deals with universal e t h i c a l p r i n c i p l e s . It i s an e t h i c a l stance derived from philosophical considerations. There are no concrete rules for behavior and as such, one must be guided by autonomous self judgment. Reasoning at either of these two stages i s r e l a t i v e l y rare ( i . e . , less 8 than 10% of adults have attained Stage 5, Colby et a l . , 1983). These six stages exist in theory, but presently Kohlberg is p r a c t i c a l l y acknowledging the existence of fiv e stages. That i s , the present scoring manual allows for the stage c l a s s i f i c a t i o n of fiv e stages for current theorizing remains unclear as to whether Stage 6 is h i e r a r c h i c a l l y d i s t i n c t from that preceding i t , and there i s no evidence of Stage 6 in Kohlberg's longitudinal data (Colby et a l . , 1983). Kohlberg assesses moral development by way of a structured interview consisting of three hypothetical dilemmas which put two moral issues in c o n f l i c t . Subjects are read these dilemmas and a series of probing questions following each in an attempt to ascertain what should be done and why: to e f f e c t i v e l y draw out the individual's reasoning. The f i r s t dilemma deals with the issue of l i f e versus law; the second deals with conscience versus punishment; and the t h i r d deals with contract versus authority. Scoring these responses e n t a i l s matching what has been recorded with an existing c r i t e r i o n judgment. Several other measures exist for the assessment of moral reasoning, but the one most relevant to t h i s study i s that advanced by Gibbs and Widaman (1982). Rather than an interview, these authors have constructed a "Social Reflection Questionnaire": a questionnaire booklet containing two so c i a l problems with questions for subjects to answer. The so c i a l problems are the standard Kohlberg dilemmas and the questions involve choosing an appropriate response and then elaborating upon that response in their own words. The benefit of such an approach to the measurement of moral reasoning i s the potential 9 for group administration, indeed a time-saving device. Gibbs and Widaman (1982) report a co r r e l a t i o n of .85 between the two te s t s . Conceptual and Integrative Complexity The work of Kelly (1955) has been heralded as the foundation and the subsequent inspiration for the many theoretic a l and methodological developments of conceptual complexity (Streufert & Streufert, 1978). His seminal contributions in the area of personality theory, in particular his Role Construct Repertory Grid, have formed the theoretical basis for the measure of conceptual complexity. The basic unit, upon which Kelly has elaborated his theory, is the "construct": a bipolar dimension which is the result of an individual's process of construing or interpreting events. It is a cognitive dimension of s i m i l a r i t y and contrast. In i t s most rudimentary form, a construct is a measure of d i f f e r e n t i a t i o n whereby, given three objects or events, a person may organize or interpret the relationship among them such that two are similar and the t h i r d i s in contrast. The work of B i e r i and others ( B i e r i , 1969; B i e r i , Atkins, Briar, Leaman, M i l l e r , & Tripodi, 1966) is an extension of the t h e o r e t i c a l propositions of K e l l y . B i e r i , however, r e s t r i c t e d his notions of cognitive complexity to the s o c i a l realm and dealt with individuals' dimensional v e r s a t i l i t y in their social judgments ( B i e r i , 1968). Higher degrees of complexity come from greater numbers of available dimensions that could be organized in more v e r s a t i l e ways. V e r s a t i l i t y refers to an individual's a b i l i t y to d i f f e r e n t i a t e independent dimensions (not categories 1 0 or concepts) as well as an individual's capacity for a r t i c u l a t i o n , similar to discrimination. This l a t t e r aspect "refers to the process of making discriminations within a dimension rather than between dimensions, as in d i f f e r e n t i a t i o n " (Streufert & Streufert, 1978, p. 20). Further developments can also be attributed to Zajonc (1960) and Scott (1969), primarily in the assessment of conceptual structure by way of t r a i t -sorting techniques; but of greater relevance to th i s study i s the systems theory a r t i c u l a t e d by Harvey, Hunt, and Schroder (1961), and the subsequent refinements, elaborations, and extensions by Schroder, Driver, and Streufert (1967) and Suedfeld (1978). The focus of the Harvey et a l . (1961) work was on s t y l i s t i c modes of information processing as dimensions of personality organization. They suggested that, rather than viewing individual differences from the point of view of the content and d i r e c t i o n a l i t y of thought processes (such as attitudes and b e l i e f s ) , i t would be more p r o f i t a b l e to examine the way in which individuals combine and use information. To thi s extent, both content and structure are important in understanding the orientation of an individual to his/her envi ronment. Harvey et a l . (1961) proposed a stage sequence of development along a continuum of complexity of conceptual structure from concrete to abstract. Concrete structures are defined as minimal d i f f e r e n t i a t i o n (referring to the number of categories or kinds of information an individual i s able to process in a given situation) whereas abstract structures are 11 defined as maximal d i f f e r e n t i a t i o n and integration (referring to the complex connections made among these d i f f e r e n t i a t e d c h a r a c t e r i s t i c s ) . Progression along this continuum was seen to flow from reliance and dependence on external authority to rebe l l i o n and avoidance of external control to dependence once again, but t h i s time "upon a f i n i t e number of persons, not upon rules (as in systems [ i . e . , stages] one and two)" (Harvey et a l . , p-. 30) to independence "without negative or positive dependency on external c r i t e r i a or negative or positive control, of or by others" (p. 30). Development i s highly dependent on training and upbringing. Development may be arrested at any one of the four stages or at a point of t r a n s i t i o n between stages i f the proper conditions for progression are not met. In 1967, Schroder, Driver, and Streufert proposed a theory of information processing which refined and elaborated upon the work of Harvey et a l . (1961). Schroder et a l . (1967) saw the previous theory evolving into one in which structural and content variables were int e r r e l a t e d aspects of conceptual complexity. That i s , conceptual development was described in terms of content such as dependence on external authority, primarily in the realm of interpersonal relationships. Furthermore, Harvey et a l . (1961) spoke of the continuous dimension of complexity while also describing discontinuous developmental leaps in stages. The proposed theory, then, abandoned t h i s developmental orientation, which was said to lack empirical support, in favor of a content-free individual difference approach to information processing; one that was said to deal with "the nature of the relationship between a person 12 and the objects of his world" (Schroder et a l . , 1967, p. 9). In Schroder et a l . ' s approach, d i f f e r e n t i a t i o n and integration remain the integral processes mapped onto a continuum of si m p l i c i t y to complexity of information processing. Simplicity i s characterized by black and white, categorical thinking. C o n f l i c t , uncertainty, or ambiguity is viewed as unpleasant and i s warded off; unambiguous and quick resolutions to problems are sought as a manner of reducing incongruity or dissonance. Single or unidimensional and fixed rule structures underlie thought and there i s an i n a b i l i t y to consider another person's perspective. A moderate position on this continuum would be represented by simultaneous generation of alternate and di f f e r e n t perceptions of the same event demonstrating the a v a i l a b i l i t y of alternative rule structures. S i m i l a r i t i e s and differences are noted without considering their relationship, or expressing their interaction as a q u a l i f i c a t i o n as opposed to the emergence of a comparison rule. At the complex end of the continuum, c o n f l i c t i n g a lternatives are tolerated and, in fact, are seen as instruc t i v e in that they may lead to new ways of organizing information and viewing the world. Causal statements may be made stemming from the generation of functional relations between alt e r n a t i v e s . The generation of new solutions i s made possible by looking at antecedent and subsequent events; a greater "connectiveness" between alter n a t i v e s i s made possible by theorizing as to the existence of these reasons. The test most t y p i c a l l y employed to examine t h i s 1 3 information processing variable i s the Paragraph Completion Test (PCT) developed by Schroder et a l . (1967). This i s a semi-projective test designed with the b e l i e f that complexity can be measured in verbal materials. Subjects are presented with six sentence stems designed to tap various domains of the decision-making environment. They are asked to complete the sentence already begun for them and write an additional two or three sentences in the 2 minutes a l l o t t e d per stem. These completions are scored on a 7-point scale r e f l e c t i n g the continuum described above. Scorers are trained to look beyond the content and examine the underlying rule structure and the manner in which dimensions are created, examined, and combined. As mentioned before, the primary emphasis of the Schroder group was to demonstrate c h a r a c t e r i s t i c levels of complexity of par t i c u l a r i ndividuals. However, there was also a recognition of the interactive nature of environmental and d i s p o s i t i o n a l factors with information processing. That i s , research has shown that complexity levels vary not only among individuals, but also within the same individual as a function of the s i t u a t i o n . For example, i t has been shown (Schroder et a l . , 1967) that as environmental stress and threat increase, there is a corresponding decrease in levels of complexity of decision makers in that environment. It i s t h i s l a t t e r l i n e of research that has stimulated the work of Suedfeld and his colleagues. Rather than viewing conceptual complexity as a r e l a t i v e l y stable personality c h a r a c t e r i s t i c , Suedfeld investigates the cognitive aspect of information processing that interacts with the environment. As 14 such, the term "integrative complexity" (Suedfeld & Tetlock, 1979) has been adopted to distinguish the two orientations. Suedfeld's emphasis d i f f e r s from those previously discussed ( i . e . , Harvey et a l . , 1961; Schroder et a l . , 1967) in two other, more methodological ways: (a) The materials to be scored are drawn from archival sources (e.g., speeches, texts) rather than sentence completions. (b) Stemming from the f i r s t , hypotheses that formerly eluded investigation are now possible, through the analysis of communications of national and international decision makers and p o l i t i c a l figures in history, for example. The scoring of t h i s archival material is an adaptation of that prepared for the Paragraph Completion Test and has been found to be appropriate for a wide variety of connected verbal materials provided that they are of s u f f i c i e n t length (Suedfeld, 1978). It is one objective of t h i s study to attempt to validate this claim and provide an empirical demonstration of the relationship between materials scored under the rubric of integrative complexity and the PCT. A series of studies have been conducted by Suedfeld and his coinvestigators that have demonstrated the interaction between c h a r a c t e r i s t i c s of the environment and information processing in high l e v e l decision-makers. Suedfeld and Tetlock (1977) examined the relationship between communication complexity and c r i s i s resolution demonstrating that complexity levels in the communications of major decision-makers were s i g n i f i c a n t l y lower during c r i s e s that ended in war than during c r i s e s that were resolved peacefully. Similar findings resulted from another study by Suedfeld, Tetlock, and Ramirez (1977). 15 Attitudes The study of attitudes has long played an integral role in the endeavors of s o c i a l psychologists. In fact, Gordon A l l p o r t once said that "the concept of attitudes is probably the most d i s t i n c t i v e and indispensible concept in contemporary s o c i a l psychology" (1958, p. 36). As one comes to expect however, such a popular and important f i e l d of inquiry is not l e f t untouched by controversy. For example, controversy has arisen in regards to the d e f i n i t i o n of an attitude and how i t d i f f e r s from a be l i e f or an opinion. Moreover, the translation of an attitude (an internal process) into a behavior or something that is quantifiable externally is also not a universally accepted process (Zimbardo, Ebbesen, & Maslach, 1977). The proposed resolutions to these controversies have shaped the study of attitudes and present a springboard into a discussion of this topic. Attitudes have generally been considered as a form of mental readiness or internal predisposition that exert some general and consistent influence on a f a i r l y large class of evaluative responses. In such a sense then, attitudes are internal and private events whose existence may only be inferred through introspection or from some form of behavior (Zimbardo et a l . , 1977). The verbalization of such an attitude is what is commonly considered an opinion. A b e l i e f , on the other hand and in accordance with Kogan (1980), is said to represent a more s t r i c t l y cognitive appraisal. The difference l i e s in that the l a t t e r i s one that can hypothetically be checked for i t s accuracy. This i s not to suggest that b e l i e f s are necessarily 16 a f f e c t i v e l y neutral; i t is that b e l i e f s are best seen as stemming from a more consensually validated data base. Thus, one would expect less variance, in a given population, on a measure of b e l i e f s than one would on a measure of a t t i t u d e . Correspondingly, any instrument that confuses the two (belief and attitude) raises the issue of just what a score on that instrument s i g n i f i e s . Investigators in this area (e.g., Rosenberg & Hovland, 1960) have found i t helpful to conceptualize attitudes as having three components: a f f e c t , cognition, and behavior. The a f f e c t i v e component i s said to consist of a person's l i k i n g or emotional evaluation of some target object. The cognitive component consists of a person's b e l i e f s about, or factual knowledge of, the target object. The behavioral component involves the person's overt behavior toward the target object. This conceptualization has received a great deal of attention and u t i l i z a t i o n e s p e c i a l l y in the realm of attitude change. An inherent confusion from such a conceptualization, however, stems from the d i s t i n c t i o n s l a i d out above. It seems possible that one could draw p a r a l l e l s between the terms attitude, b e l i e f , and opinion and the components a f f e c t , cognition and behavior, respectively. That i s , an attitude "implies that one is favorably or unfavorably disposed toward the target object of the attitude" (Kogan, 1980, p. 301); a b e l i e f is the more s t r i c t l y cognitive appraisal; and, an opinion is the overt verbalization (behavioral expression) with respect to the target object. Such interaction of terminology can only serve to generate confusion in understanding t h i s area. 17 Perhaps a more useful and relevant d i s t i n c t i o n to make concerning the concept of an attitude is that proposed by Schroder et a l . (1967). They claim that attitudes are t y p i c a l l y described in terms of the magnitude and d i r e c t i o n of their contents ( i . e . , the degree of item endorsement). Cognitive st r u c t u r a l processes, however, should strongly a f f e c t the way in which content i s assimilated, organized, and expressed. That i s , from the foregoing discussion of conceptual complexity, "int e g r a t i v e l y complex structures d i f f e r e n t i a t e and integrate more complex information than do concrete structures. Attitudes are formed as a consequence of the d i f f e r e n t i a t i o n and integration of dimensions of information about a domain of st i m u l i " (Schroder et a l . , 1967, p. 126). The result of such a l i n e of argument i s the expectation that concrete attitudes are: (a) based on less information since less information tracking occurs associated with less d i f f e r e n t i a t i o n and integration; (b) more stable over time since less information i s sought and assimilated or accomodated; and are consequently (c) more c a t e g o r i c a l . On the other hand, complex attitudes: (a) perceive a broader range of information as being relevant; (b) tend to integrate other than only the sali e n t items of information or more than one dimension leading to a more f l e x i b l e , less extreme attitude; and are therefore (c) less categorical. Several d i f f e r e n t paper-and-penci1 tests have been developed to measure attitudes. Typically, these tests e n t a i l the presentation of a series of statements and subjects are requested to check those statements with which they agree 18 (Thurstone's Method of Equal-Appearing Intervals; Thurstone & Chane, 1929) or to indicate the extent to which they agree or disagree with each statement or item on a 5-point scale (Likert's Method of Summated Ratings; L i k e r t , 1932). Two other tests that have also been f a i r l y highly refined and extensively used are: Guttman's Scalogram in which the attitude i s measured by having the subject check a l l the statements on a unidimensional scale that are acceptable (Guttman, 1950), and Osgood's Semantic D i f f e r e n t i a l in which the procedure i s to have subjects judge a p a r t i c u l a r concept on a set of semantic scales which are verbal opposites separated by seven discriminable steps (Osgood, Suci, & Tannenbaum, 1957). The use of such scales is understandably widespread for they are simple to administer and easy to score. They can be given to large numbers of people for which a great deal of data may be c o l l e c t e d in a short space of time. "They also have the convenience of a neat continuum of scores that w i l l probably closely resemble a normal d i s t r i b u t i o n enabling the investigator to employ conventional s t a t i s t i c s for his or her analyses" (Kogan, 1982, p. 307). A serious conceptual problem, however, i s the methodological s i m p l i c i t y in the use of such scales. Attitude items or statements of the sort employed in the above scales necessarily force subjects to generalize about the stimulus class or event. An example of t h i s problem is found in the work of Kogan (1979). He administered his scale of Attitudes Toward Old People to a sample of elderly subjects with the expectation that the el d e r l y individuals would respond more consistently 19 because of their direct personal involvement in the target a t t i t u d i n a l domain. The response patterns were generally less consistent, for which Kogan (1982) offered the following explanat ion: Instead of the overgeneralized, low salient attitude l i k e l y to be t y p i c a l of younger samples, older persons w i l l have had personal experience with older people who f i t some of the items, but w i l l not have known older peers to engage in behaviors or hold opinions expressed in other items. In other words, the older respondent's view of his age peers is l i k e l y to be more d i f f e r e n t i a t e d than that held by younger persons, and such d i f f e r e n t i a t i o n cannot be accommodated by generalized attitude scales, and, in fact, is responsible for reducing the indices that make such scales psychometrically adequate. I raised the p o s s i b i l i t y that attitude scales may have prompted us to ask the wrong question. Instead of asking about the degree of p o s i t i v i t y or negativity in views held about older people, one might more pro f i t a b l y ask how d i f f e r e n t i a t e d is a respondent's view of older persons. (p. 308) Perhaps the answer l i e s in using no single or unitary measure, but rather a combination in which d i f f e r e n t i a t e d views may be displayed in conjunction with the e a s i l y administered item-endorsement technique. Perhaps, too, such a combination would address the content-structure d i s t i n c t i o n proposed by Schroder et a l . (1967). That i s , in providing individuals with 20 the opportunity to both describe their a t t i t u t d e - - i n their own words—and indicate their agreement or disagreement with relevant a t t i t u d i n a l statements, one could more eas i l y examine the structure of their thought on this issue and the content around which th i s structure i s formed. The three areas of moral reasoning, conceptual complexity and attitudes have been described separately in an attempt to promote c l a r i t y . This separation might be considered a r t i f i c i a l on both conceptual and empirical grounds, however, for there is a common thread connecting these topics. A discussion of these in t e r r e l a t i o n s h i p s is the purpose of the following sections; an examination of these int e r r e l a t i o n s h i p s is the purpose of this study. Moral Reasoning and Conceptual Complexity Kohlberg (1981) has stated that his stages of moral judgment development represent a cumulative hierarchy of cognitive complexity in that each succeeding stage in the hierarchy i s more cognitively d i f f e r e n t i a t e d and better integrated. Such references suggest that there should be a re l a t i o n s h i p between his stages and measures of conceptual complexity. The demonstrated relationship between increases in role-taking a b i l i t y and moral maturity (Walker, 1980; Selman, 1976) also suggests that there might be a relationship to conceptual complexity. As described by F l a v e l l (1968), in order to take on the role of another, one must be able to discriminate or d i f f e r e n t i a t e his or her role a t t r i b u t e s . The progression of the two theories seems to develop from rules and expectations as being perceived as external to the 21 s e l f , through to the self as i d e n t i f i e d or equated with the rules, to the self as d i f f e r e n t i a t e d from pre v a i l i n g conventions. Kohlberg (1976) c l e a r l y describes this in his description of the three levels of moral reasoning as do Harvey et a l . (1961) in their systems theory and Schroder et a l . (1967) in their analysis of ranges along the simplicity-complexity continuum. Furthermore, both theories emphasize form or structure as a basis rather than content. The focus is on the cognitive structure that underlies thought, with the structure becoming increasingly complex as one moves up the hierarchy or along the continuum. In moral reasoning, higher stages are seen as being more morally adequate than the lower stages to the extent that higher stages are more cognitively d i f f e r e n t i a t e d or capable of handling more problems or moral c o n f l i c t s . The stages represent an increased d i f f e r e n t i a t i o n and h i e r a r c h i c a l integration of moral values. Given t h i s conceptual basis, i t i s surprising to note that r e l a t i v e l y few studies have addressed t h i s i n t e r r e l a t i o n a l issue. Two s h a l l be reviewed here. Sull i v a n , McCullough, and Stager (1970) explored the relationships between ego development (Loevinger, 1966), conceptual systems development (Harvey et a l . , 1961), and moral judgment development (Kohlberg, 1973) with a sample of adolescents. Of pa r t i c u l a r interest to t h i s discussion i s the relationship between the l a t t e r two areas of development—for which an o v e r a l l correlation of .62 was found. The authors commented that t h i s c orrelation c l e a r l y represents the 22 correspondence between these two measures, but that, even under ideal measurement conditions, one could not expect "that the ... dimensions would be perfectly related i f they are tapping somewhat similar yet d i f f e r e n t dimensions of personality development" (Sullivan et a l . , 1970, p. 410). That i s , they are not equivalent measures. Thus, there i s at least moderate support for the rel a t i o n s h i p between conceptual complexity and moral reasoning. A study by Falconer (1973) also lends q u a l i f i e d support to the hypothesized rel a t i o n s h i p between these two areas. Falconer used B i e r i ' s (1969) measure which defines complexity as "the tendency to construe s o c i a l behavior in a multi-dimensional way such that a more cognit i v e l y complex individual has available a more v e r s a t i l e system for perceiving the behavior of others than does a less cognitively complex person" (p. 46). Complexity, in th i s sense, is a d i f f e r e n t i a t i v e a b i l i t y embracing only half of the conceptualization as proposed by Harvey et a l . (1961) and Schroder et a l . (1967). Perhaps, t h i s , in part, explains the resultant moderate support generated. A s i g n i f i c a n t difference was found on the B i e r i measure of complexity between the lowest and highest stages of moral development evidenced in the sample. There were no differences found between adjacent stages of moral development; however, a nonsignificant trend was noted. This discussion leads to the second proposed objective of the present study: an investigation of the relationship between conceptual complexity and moral reasoning. The design of the present study improves upon the two studies c i t e d above for several reasons: (a) Complexity i s defined in the terms provided 23 by Schroder et a l . (1967) encompassing both d i f f e r e n t i a t i o n and integration (as opposed to B i e r i ' s method) and representing the cognitive a b i l i t y for information processing as opposed to a less validated cognitive developmental sequence (as proposed by Harvey et a l . , 1961). (b) Throughout the years, Kohlberg's theory of moral development has been reworked and refined resulting in a more stringent stage c r i t e r i a (Colby et a l . , 1983). In fact, C a r r o l l and Rest (1982) have reported a modest corr e l a t i o n of .39 between the present scoring system (now in apparently f i n a l form) and the i n i t i a l one. The greatest support for the theory and the measure has come using this present formulation. It i s for these two important reasons that th i s proposed study is seen as an improvement and extension of the work preceding i t . Moral Reasoning and Capital Punishment Capital punishment has been, and remains, a controversial issue intensely debated by both armchair and professional s o c i a l s c i e n t i s t s , theologians, and p o l i t i c i a n s around the world. A main focus of research into t h i s area has been of a factual nature; in p a r t i c u l a r , researchers have examined the deterrent e f f e c t , a question that largely remains unanswered. Representative research comes from S e l l i n (1959) and P h i l l i p s (1980). S e l l i n (1959) has examined a large mass of homicide data and concludes that the death penalty has largely f a i l e d as a deterrent. He claims that i t s use has exercised no influence on the extent of the fluctuating rates of c a p i t a l crimes. P h i l l i p s (1980), however, sees i t d i f f e r e n t l y . His s o c i o l o g i c a l , archival analysis suggests that the wrong metric 2 4 has been used in examining deterrent e f f e c t s ; observing yearly murder rates of c a p i t a l crimes in correspondence to the death penalty clouds or obscures the demonstrated weekly or monthly deterrent e f f e c t s . A d e f i n i t i v e answer has yet to be advanced. What seems apparent, however, i s that t h i s research on the factual aspects of c a p i t a l punishment has l e f t l i t t l e time or energy for researchers to pay attention to the moral-psychological issues (Kohlberg & Elfenbein, 1981). The research that has been done in the more a t t i t u d i n a l realm has often observed the correlation between attitudes toward c a p i t a l punishment and other personality measures such as the authoritarian personality (Vidmar & Ellsworth, 1974), a constellation of attitudes of prejudice, ethnocentrism, conservatism, and authority as i n i t i a l l y conceived by Adorno, Frenkel-Brunswick, Levinson, and Sanford (1950). As Kohlberg and Elfenbein (1981) point out, however, i t would probably be more f r u i t f u l to examine c a p i t a l punishment from the perspective of an interaction between an individual's moral assumptions (for the moral nature of a decision regarding c a p i t a l punishment seems i n t u i t i v e ) and attitudes. Such a conceptualization f i t s in well with the content-structure d i s t i n c t i o n in moral reasoning made by Kohlberg whereby the content of a stage i s the attitude or judgment formed and the structure i s the method of reasoning employed. "Whether or not a subject approves of c a p i t a l punishment, for example, i s a matter of content; the rationale he adopts i s a matter of structure" (Kohlberg & Elfenbein, 1981, p. 270). Without denying the existence of structure and content as 25 separable and separate concepts, Kohlberg and Elfenbein (1981) admitted that there is a " p r o b a b i l i s t i c tendency for certain stage structures to generate s p e c i f i c moral attitudes" (p. 270). This was addressed by examining the relationship of moral stage to opinion about c a p i t a l punishment using 30 longitudinal subjects. Kohlberg and Elfenbein (1981) predicted a l i n e a r relationship with opposition to c a p i t a l punishment increasing with stage of moral reasoning. Since the p r i n c i p l e d l e v e l represents a formal conception of justice as reversible role-taking (following from Rawls' theory of j u s t i c e , 1971), Kohlberg (1981) claimed that the upper l e v e l p r i n c i p l e d thinker assumes the position of having an equal (or unknown) chance of being anyone in a p a r t i c u l a r society and thus sets up a f a i r mental representation so that any p r i n c i p l e s agreed upon w i l l be just. For example, Kohlberg and Elfenbein (1981) say that any rational person would risk a penalty less harsh than death ( i f one assumed the position of ordinary c i t i z e n or victim) thus leading to the rejection of c a p i t a l punishment. They found that at Stage 3 and below, almost a l l of the subjects claimed that i t was right to give the death penalty; Stage 4 subjects were divided on t h i s issue; and Stage 5 subjects unanimously rejected the use of c a p i t a l punishment. Kohlberg and Elfenbein (1981) concluded that "on the whole, the early stages of development are characterized by moral approval of the death penalty, but as individuals progress in their moral development, their moral reservations about c a p i t a l punishment multiply and they gradually come to condemn i t as immoral" (p. 271). DeWolfe and Jackson (1984) supported t h i s finding in a 26 study which found p r i n c i p l e d moral reasoning (as determined by Rest's, 1979, Defining Issues Test, a questionnaire measure assessing preference for p r i n c i p l e d moral reasoning statements) to relate negatively to attitudes toward c a p i t a l punishment (r_ = -.41). That i s , preference for upper le v e l reasoning was associated with increasing opposition to c a p i t a l punishment. The conclusion of both of these studies i s congruent with other research suggesting that higher stages of moral reasoning are t y p i c a l l y associated with l i b e r a l attitudes (Candee, 1976; Fishkin, Keniston, & MacKinnon, 1973; Haan, Smith, & Block, 1968). This has become a somewhat controversial issue in i t s e l f , as i t has led others ( i . e . , S u l l i v a n , 1977) to c r i t i c i z e Kohlberg's theory of moral development on the grounds of a l i b e r a l i deological bias. In response, Kohlberg and Elfenbein (1981) point to the p r o b a b i l i s t i c (as opposed to deterministic) relationship between structure and content as a function of factual assumptions made by subjects connected to the values of their culture. Thus, they suggest that t h i s i s a c u l t u r a l phenomenon rather than a bias inherent in the theory. Any predominance of l i b e r a l i s m i s a s o c i e t a l function and not a theoret i c a l bias. It is from t h i s foregoing body of research that a t h i r d objective of t h i s study emerges: to examine the relationship between stage of moral reasoning and attitudes toward c a p i t a l punishment. Rather than r e p l i c a t i n g the work already described, the proposed study d i f f e r s in a least three fundamental ways, (a) Kohlberg derived the content of an individual's reasoning by way of the subject's response to the question: "Is i t ever right 27 to give c a p i t a l punishment?". However, i t would seem more quantita t i v e l y and q u a l i t a t i v e l y v a l i d to allow a range of responses, as opposed to only a dichotomy, in the assessment of an attitude. Hence, an attitude questionnaire permitting responses from strongly disagree to strongly agree in response to a series of statements surfaces as an appropriate measure. In conjunction with Kohlberg's method of assessment, the attitude towards c a p i t a l punishment is to be derived from Likert's method of summated ratings by way of a measure developed by Palys (1981). (b) Kohlberg and Elfenbein (1981) used an exclusively male sample in their study. The sample of this study w i l l comprise equal numbers of males and females, (c) Kohlberg and Elfenbein (1981) used a now outdated scoring system. This study uses the present scoring system (Colby et a l . , in press). Given the changes in stage descriptions, i t would be interesting to note the relationship between the present conceptualization of moral reasoning and attitudes toward c a p i t a l punishment. The derivation of the fourth objective of t h i s study is probably best understood in r e l a t i o n to the foregoing discussion. As mentioned e a r l i e r , the manner in which moral reasoning i s assessed comes from presenting subjects with a situation in which two moral values are in c o n f l i c t , and then questioning them as to what should or should not be done in this s i t u a t i o n . Thus, a choice is always involved. For example, in response to the question, "Is i t ever right to give c a p i t a l punishment?", a subject i s faced with the option of responding in the affirmative or the negative. This is their content 28 evaluation. Their moral stage, however, is derived from their elaboration of this response supplying the reasons to the questions why or why not. Str u c t u r a l l y , reasons for either answer may be i d e n t i c a l , but with respect to the content, the two responses may be in complete disagreement. The disagreement i s on the issue, not on the reasoning. The scoring of a moral reasoning protocol e n t a i l s matching the subject's responses to c r i t e r i o n judgments provided in the scoring manual. The protocol stage score is derived by summing across stage-scored responses for both sides of the issue and weighting the favored or chosen position more heavily (by a ra t i o of 3:2). The underlying assumption here i s that better reasoning i s provided for the chosen position. What has become apparent, in reviewing the relevant l i t e r a t u r e in this area, is that there has been no systematic investigation of the differences in le v e l of reasoning between chosen and nonchosen positions within individuals and within moral dilemmas. In fact, the closest approximation to such a task has been the study by Nisan and Kohlberg (1982) who examined stage scores across issues, but not by chosen versus nonchosen position. Thus, i t i s an intent of this study to examine any differences in the l e v e l of moral reasoning of subjects as they consecutively support and oppose the use of c a p i t a l punishment. The cognitive-developmental notion of structured whole suggests that reasoning on both sides of the issue w i l l be at comparable or equivalent stages. This appears somewhat inconsistent with the aforementioned weighting procedures and the underlying rationale. This study w i l l 29 attempt to deal with this incongruity in Kohlberg's theory. Conceptual Complexity and Capital Punishment As outlined above, Schroder et a l . (1967) have proposed a most useful d i s t i n c t i o n between the content and structure of an attitude. An attitude structure refers to the conceptual processes that an individual employs in organizing and processing information about a par t i c u l a r range of s t i m u l i . The content is t y p i c a l l y , a description of an attitude in terms of i t s magnitude and di r e c t i o n relevant to a p a r t i c u l a r range of sti m u l i . As such, simple attitudes are based on a narrow range of highly s a l i e n t information. Information that i s descrepant or that does not f i t the ex i s t i n g attitude i s only minimally perceived or u t i l i z e d . Hence, attitudes that are s t r u c t u r a l l y simple are expected to be more categorical. The more abstract or complex the attitude, however, the broader the range of information that i s perceived as relevant. As such, the potential for inclusion of discrepant units of information i s higher, t y p i c a l l y resulting in a less categoric a l , and subsequently, less extreme a t t i t u d e . At least p a r t i a l substantiation of the above theoret i c a l positions comes from a number of experiments in which the F Scale (as advanced by Adorno et a l . , 1950) has been correlated with complexity in the range of -.25 to -.55 (as reported in Schroder et a l . , 1967). The occurrence of such correlations stems from the overrepresentation of cognitively simple persons among high F scorers. Further support for t h i s position exists in the work of L i n v i l l e and others ( L i n v i l l e , 1982; L i n v i l l e & Jones, 1980). 30 L i n v i l l e (in press) has proposed a model focusing on the link between cognitive factors and a f f e c t i v e factors in information processing. The cognitive factor refers to the complexity of the knowledge structure used in thinking about a p a r t i c u l a r domain, whereas the a f f e c t i v e factor refers to the extremity of responses to stimuli or information in that domain. L i n v i l l e has noted that the majority of psychological theories about so c i a l evaluation have tended to emphasize--if not overemphasize--the u n i d i r e c t i o n a l effects and biases. For example, L i n v i l l e (in press) has stated that "greater extremity does not refer to a consistent tendency to rate stimuli more extremely in only one d i r e c t i o n , but rather to rate stimuli more extremely in both d i r e c t i o n s , either more p o s i t i v e l y or more negatively depending on the f a v o r a b i l i t y of information about the stimulus" (p. 5). Thus, the basic premise of the model i s as follows: The less complex a person's representation of a given domain, the more extreme w i l l be the person's affect regarding stimuli in that domain. In other words, i f the representation i s simple, that i s , a person's thinking i s in terms of fewer nonredundant features or aspects, affect w i l l be r e l a t i v e l y extreme. When the representation i s more complex, affect w i l l be more moderate. ( L i n v i l l e , in press, p. 5) A research program conducted by L i n v i l l e ( L i n v i l l e & Jones, 1980; L i n v i l l e , 1982) has provided empirical support for her complexity-extremity hypothesis. L i n v i l l e (1982), for example, correlated an individual-difference measure of complexity (by 31 way of a t r a i t - s o r t i n g technique, after Scott, 1962, 1969) with extremity of evaluation for the target group of older males. College-age males participated in this study in which i t was shown that greater complexity regarding older males related to less extreme evaluations of the same. The complexity and evaluative extremity scores were s i g n i f i c a n t l y negatively correlated (r_. = -.65). Moreover, cognitively simple subjects were both more favorable toward the favorable older male stimulus and more unfavorable toward the unfavorable older male stimulus. Similar results have also been reported using race and sex as ingroup-outgroup variables. L i n v i l l e and Jones (1980) tested the complexity-extremity hypothesis by d i r e c t l y manipulating complexity through task instructions which induced subjects to adopt a simple or complex orientation toward a set of s t i m u l i . The subject was asked to think about a stimulus (a strong or weak essay) in terms of several features (six features for the high complexity condition) prior to making a f a v o r a b i l i t y rating. As predicted, those in the low complexity condition rated the strong essay more extremely favorable and rated the weak essay more extremely negative compared to subjects in the high complexity condition who were more moderate. Similar results were achieved in a second study where subjects tasted and evaluated fiv e different types of chocolate-chip cookies ranging in quality given either two or six features to consider. These results demonstrate the existence of a relationship between complexity of thought or cognitive structures and styles of evaluation. In p a r t i c u l a r , more s i m p l i s t i c cognitive 32 structures relate to more extreme evaluations and more complex cognitive structures relate to more moderate evaluations. Complexity, however, in the terms described by L i n v i l l e (1982), is the assessment of discrimination, and not integration. Although L i n v i l l e likens her measure of complexity to that of Schroder et a l . (1967), i t actually embraces only half of this conceptualization in i t s f a i l u r e to assess integrative a b i l i t y . Tetlock (1983b, 1984) has provided results exploring the t o t a l realm of complexity in i t s r e l a t i o n to p o l i t i c a l ideology. Tetlock (1983b) assessed the integrative complexity of speeches given by United States senators who were known to have l i b e r a l , moderate, or extremely conservative voting records. It was assumed that right-wing orientations were generally more dogmatic and intolerant of ambiguity than left-wing or moderate orientations. The results add credence to such a position: the policy statements of extremely conservative senators were less integratively complex than were those of moderate or l i b e r a l senators (between whom there was no s i g n i f i c a n t d ifference). More d i r e c t l y relevant to t h i s study, however, is Tetlock's (1984) analysis of cognitive s t y l e and p o l i t i c a l b e l i e f systems in the B r i t i s h House of Commons. In a sample of 89 parliamentarians, Tetlock' was able to c l a s s i f y individuals as adhering to one of the following four p o l i t i c a l orientations: extreme s o c i a l i s t , moderate s o c i a l i s t , moderate conservative, and extreme conservative. Following Rokeach (1973), moderate s o c i a l i s t s are seen to value freedom and equality more highly than extreme s o c i a l i s t s (who value equality more highly than freedom) or moderate or extreme conservatives (who value freedom 33 more than e q u a l i t y ) . Tetlock (1984) suggested that adherents to ideologies that value both freedom and equality highly are under greater pressure to consider policy issues in more integratively complex terms than are advocates of ideologies that place a greater weight on only one value. He refers to t h i s as the value pluralism model. Results supported t h i s model in that both extreme groups had similar low leve l s of integrative complexity and s i g n i f i c a n t l y lower levels than moderate s o c i a l i s t s and moderate conservatives. A f i n a l objective of t h i s study, then, that stems from the foregoing research, i s to address the relationship between complexity and evaluation in the important and controversial a t t i t u d i n a l domain of c a p i t a l punishment. Like L i n v i l l e (1982), this study w i l l examine complexity in re l a t i o n to ratings of a t t i t u d i n a l statements, but d i f f e r s in the measure of complexity u t i l i z e d . Like Tetlock (1984), t h i s study w i l l examine integrative complexity, but also includes conceptual complexity and the examination of these two measures in rel a t i o n to c a p i t a l punishment with more of a moral, as opposed to p o l i t i c a l , analysis. Thus, the present study w i l l explore the t o t a l realm of complexity as defined by Schroder and others, who would predict a c u r v i l i n e a r relationship between attitudes and complexity. That i s , the more extreme the evaluation of c a p i t a l punishment (whether i t be extremely in favor of, or in opposition to, c a p i t a l punishment) the less complex w i l l be the structure of that att i t u d e . A more moderate attitude would be associated with a more complex structure. 34 .1 Hypotheses In summary, the hypotheses to be addressed in th i s study are: Hypothesis I. It is expected that the writings of individuals describing their attitude towards c a p i t a l punishment (integrative complexity) w i l l be related, in terms of complexity, to the completions of sentence stems in the Paragraph Completion Test (conceptual complexity). Both of these measures are assessing the structural complexity of thought ( a l b e i t , in d i f f e r e n t domains) and, as such, a moderate co r r e l a t i o n ( i . e . , accounting for less than 50% of the variance) is predicted. Hypothesis I I . Following from the theoretical descriptions of Schroder et a l . (1967) and Streufert and Streufert (1978), as well as the work of L i n v i l l e (1982) and Tetlock (1984), i t is expected that extreme attitudes (both pro and con) w i l l be related to a more simple structure and more moderate attitudes w i l l be related to a more complex structure. Hypothesis I I I . It i s expected that there w i l l be a s i g n i f i c a n t , positive r e l a t i o n s h i p between the measures of moral reasoning and conceptual complexity, both assessing the structure of thought. Hypothesis IV. Following from the work of Kohlberg and Elfenbein (1981) and DeWolfe and Jackson (1984), a linear relationship between attitudes toward c a p i t a l punishment and stage of moral reasoning i s expected. In p a r t i c u l a r , i t i s hypothesized that opposition to c a p i t a l punishment w i l l increase with stage of moral reasoning. 35 Hypothesis V. Although never e x p l i c i t l y addressed by Kohlberg, i t i s implied that reasoning about the chosen position i s , in some way, more advanced and better formulated than reasoning about the nonchosen position. It is therefore hypothesized that there w i l l be a difference between reasoning of the chosen position and reasoning of the nonchosen position. S p e c i f i c a l l y , i t is expected that the chosen position w i l l be represented by higher levels of moral reasoning than the nonchosen po s i t i o n . 36 Method Subjects Participants were 36 male and 36 female students who were drawn in equal numbers from three educational levels (introductory and senior undergraduate, and graduate) in an attempt to generate a range on several of the measures (e.g., conceptual complexity, moral development). S p e c i f i c a l l y , graduate students in the departments of law, philosophy, theology, and medicine were recruited under the assumption that such f i e l d s offer their students the opportunity to discuss philosophical and moral issues, perhaps promoting advanced understanding and reasoning of such issues. Introductory and senior undergraduate students were recruited from f i r s t - and third-year psychology courses, respectively. The mean age of the introductory students was 18.9 years, with a range of 17 to 28 years. The mean age of the senior undergraduates was 24.7 years, with a range of 18 to 45 "years. The graduate population had a mean age of. 28.3 years, with a range of 22 to 35 years. Measures Participants were administered, either i n d i v i d u a l l y or in groups, the following battery of tests in a single session of approximately 1 hour. (The complete battery of tests can be found in Appendix A.) 1. The Paragraph Completion Test (PCT). This test consisted of six sentence stems tapping d i f f e r e n t domains of the decision-making environment (Schroder et a l . , 1967). Participants were requested to complete the sentence which was 37 already begun for them, and then to write at least one more sentence in the 2 minutes that were provided for each completion. This measure was comprised of the following stems: (a) Rules . . . (b) When I don't know what to do . . . (c) When I am c r i t i c i z e d . . . (d) Policemen . . . (e) When a friend acts d i f f e r e n t l y towards me . . . (f) Confusion . . . . These completions were scored for the le v e l of conceptual complexity. 2. The Capital Punishment Issue Composition. In this task, participants were requested to write a r e l a t i v e l y brief composition ( i . e . , 5 to 10 paragraphs) discussing the issue of ca p i t a l punishment as they perceived i t . They were instructed to present their opinion and attitude toward c a p i t a l punishment and to include the discussion of any factors considered relevant. This task was scored both for level s of integrative complexity and moral reasoning. Moreover, individuals' position on the issue of c a p i t a l punishment ( i . e . , either pro or con) was also determined from this measure. This procedure i s not unlike that used by Kohlberg and Elfenbein (1981) in which attitude was assessed by response to the question, "Is i t ever right to use ca p i t a l punishment?". 3. Statements supporting and opposing c a p i t a l punishment. This task required subjects to generate detailed reasons to both support and oppose the use of c a p i t a l punishment, counterbalanced for order of presentation. These statements, 38 along with the preceding issue composition, were scored for moral development, to more f u l l y assess reasoning on both sides of the issue. This follows from Colby et a l . (in press) who base an individual's moral reasoning p r o f i l e on scores from both the chosen and nonchosen position on the issue ( i . e . , the preferred and nonpreferred pos i t i o n ) , both e x p l i c i t l y addressed in their structured interview. 4. The Capital Punishment Questionnaire (CPQ). This Likert-type questionnaire (Palys, 1981) provided 24 statements of opinion regarding c a p i t a l punishment (12 in a pro-capital punishment d i r e c t i o n and 12 in a a n t i - c a p i t a l punishment d i r e c t i o n ) . Subjects were asked to indicate the extent to which they agreed or disagreed with each statement on a 7-point scale ranging from -3 (strongly disagree) through 0 (neutral, no opinion) to +3 (strongly agree). Scoring Conceptual complexity. The assessment of conceptual complexity came from the PCT, the most frequently used and most thoroughly validated of the tests designed to tap an individual's c h a r a c t e r i s t i c mode or style of information processing (Porter & Suedfeld, 1981). The paragraph completions were scored on a 7-point scale for l e v e l of conceptual complexity as defined in terms of d i f f e r e n t i a t i o n and integration (previously described). The scoring procedure follows d i r e c t l y from the manual provided by Schroder et a l . (1967), as well as subsequent publications from Princeton University ("Princeton Manual Guidelines", undated) and the University of B r i t i s h Columbia ("UBC Guidelines for the Scoring 39 of Conceptual and Integrative Complexity", undated). Schroder et a l . (1967) have described a number of gradations or scores that f i t onto the 7-point scale defining the simplicity-complexity continuum of information processing. A score of 1 r e f l e c t s minimal or no d i f f e r e n t i a t i o n and no integration; a score of 3 r e f l e c t s moderate to high d i f f e r e n t i a t i o n with no real integration; a score of 5 r e f l e c t s moderate to high d i f f e r e n t i a t i o n and moderate to high integration; and a score of 7 r e f l e c t s maximal d i f f e r e n t i a t i o n and integration. Scores of 2, 4, and 6 can be seen to represent t r a n s i t i o n a l points between adjacent l e v e l s . It should be noted that this system of scoring focuses on the underlying cognitive structure of thought patterns and not on the content, and is therefore not biased for or against any p a r t i c u l a r p o l i t i c a l philosophy (Suedfeld & Rank, 1976). Completions were deemed unscorable i f only one sentence had been written or i f the sentences were incoherent, descriptive (as opposed to evaluative), c l i c h e s , quotations, or s a t i r e . According to these c r i t e r i a , only 9 of the 432 completions were found to be unscorable. A l l other completions were appropriate for the scoring procedures outlined above. For each subject, a mean score of the PCT stems was calculated and this was the measure employed in subsequent analyses. The PCT completions were b l i n d l y scored by stem, and not by subject, to prevent bias. In addition, a random sample of 96 completions (the 6 PCT stems for each of 16 subjects) was scored by another trained scorer to assess r e l i a b i l i t y . Interjudge r e l i a b i l i t y was found to be r = .79. 40 Integrative complexity. The issue composition task was included as a method of assessing integrative complexity, a term coined by Suedfeld and Tetlock (1977) as a means of distinguishing between the more t r a d i t i o n a l personality c h a r a c t e r i s t i c variable of conceptual complexity and t h i s more novel approach to information processing. Clearly, the paramaters of the issue composition d i f f e r from that of the PCT in that the former i s both longer, void of time r e s t r i c t i o n s , and on a s p e c i f i c issue. However, the issue composition also d i f f e r s from material t y p i c a l l y scored under the rubric of integrative complexity in that the composition i s not from an archival source and therefore material for these subjects is limited to that which was written for purposes of t h i s study. This difference, however, is more of a s h i f t in emphasis or a difference in source rather than a difference in the underlying theory of complexity or the basic scoring system. That i s , materials from both archival sources and the PCT are s i m i l a r l y scored along the 7-point simplicity-complexity continuum as a function of d i f f e r e n t i a t i o n and integration. Further differences warrant discussion at t h i s point. Archival materials have t y p i c a l l y been speeches and diplomatic communications by leading decision makers (Suedfeld & Rank, 1976; Suedfeld & Tetlock, 1977) or personal correspondence by eminent public figures (Porter & Suedfeld, 1981; Suedfeld, in press) which were written over time and in a variety of settings and s i t u a t i o n s . The material used in this analysis came from university students writing about c a p i t a l punishment at a single point in time. Differences between these populations and the 41 parameters within which they were studied appear worthy of mention. F i r s t , i t could be reasonably suggested that high-l e v e l p o l i t i c a l policy-makers, philosophers, and authors d i f f e r in some s i g n i f i c a n t way from university students. One would expect differences in performance between subjects on an individual difference variable, however, without jeopardizing the v a l i d i t y of the measure. Second, archival studies have t y p i c a l l y addressed complexity as a variable over time and situations, whereas this study examined i t at a single point in time. This once again is more representative of a s h i f t in focus than i t i s of an inappropriate application of the scoring system to a p a r t i c u l a r body of material. The main emphasis of the Schroder group was on the c h a r a c t e r i s t i c complexity levels of p a r t i c u l a r individuals, but some of their work has also addressed fluctuations and variations of conceptual l e v e l within individuals as a function of the situation/environment (Schroder et a l . , 1967). It i s , therefore, appropriate to examine complexity from a between- and/or within-subjects perspective. This study concentrated on the former. Thirdly, i t could be suggested that differences l i e in the motivation of the individuals who were supplying the data. That i s , participants in the study were requested to write on a s p e c i f i c topic and perhaps i t could be said that they were less motivated than those individuals selected for study in archival analyses writing on more in t e r n a l l y generated issues (with the i m p l i c i t assumption that motivation a f f e c t s complexity). This l i n e of argument weakens, however, when one considers that l e t t e r s and speeches are t y p i c a l l y intended for certain audiences and often 42 in response to other l e t t e r s , speeches, or actions. With th i s in mind, however, the instructions for the issue composition were kept in t e n t i o n a l l y i n e x p l i c i t and open-ended ("Discuss the issue of c a p i t a l punishment"), so as not to favor any pa r t i c u l a r position, not e l i c i t any p a r t i c u l a r structure. To summarize t h i s discussion, the important point i s that, although the scoring system for the PCT was designed for the material of the test, i t has been found to be appropriate for a wide variety of connected verbal materials with the provision that they are of s u f f i c i e n t length (Suedfeld, 1978). The basic scoring unit in integrative complexity is the paragraph (Suedfeld, 1978), defined as two or more sentences. "S u f f i c i e n t length" for t h i s material i s considered to be 5 to 10 paragraphs. Subjects generated an average of 6.3 paragraphs in their composition on c a p i t a l punishment, with a range of from 2 to 12. The guidelines and r e s t r i c t i o n s for the scoring of the PCT, as outlined above,, are equally applicable here. In accordance with these, only 2 of 453 paragraphs were c l a s s i f i e d as unscorable. The paragraphs of the issue composition were bl i n d l y scored by paragraph and not by subject, one randomly chosen paragraph at a time. For each subject, a mean score was calculated and th i s was the measure employed in subsequent analyses. Interjudge r e l i a b i l i t y was assessed by as second independent rater who scored a random sample of 92 paragraphs (the issue compositions of 16 subjects). Interjudge r e l i a b i l i t y was found to be r = .80. Moral reasoning. Moral reasoning was assessed by the 43 combined scoring of the issue composition and the statements of support for, and opposition to, c a p i t a l punishment. In and of i t s e l f , the issue composition was i n s u f f i c i e n t for the derivation of a c h a r a c t e r i s t i c score of moral reasoning, since i t does not e x p l i c i t l y address both sides of the issue (the chosen and nonchosen p o s i t i o n s ) . When scored in combination with the support and opposition statements, reasoning on both sides of the issue could be obtained and a s u f f i c i e n t amount of reasoning could be provided for r e l i a b l e scoring. The measurement of moral reasoning comes from a system advanced by Lawrence Kohlberg using the Standard Form Scoring Manual. After a decade of revisions, i t i s now in f i n a l form (Colby et a l . , in press). The scoring of a t y p i c a l moral judgment interview e n t a i l s breaking the interview into discrete judgments which are matched to c r i t e r i o n judgments (conceptual counterparts and prototypes found in the scoring manual) and assigned scores on the basis of these matches. For an interview judgment to be scorable, i t must provide reasoning which i s considered v a l i d by the subject ( i . e . , not discounted) and which is p r e s criptive in nature ( i . e . , what should be done, not what would be done). Each judgment meeting the c r i t e r i a described above was assigned a stage score. These scored judgments were then separated—by subject—on the basis of the side of the issue they addressed (judgments supporting c a p i t a l punishment and judgments opposing c a p i t a l punishment). This allowed both positions to be expressed in terms of a weighted average score (WAS) and a global stage score (GSS). The WAS i s a weighted 44 average of stage usage and is calculated by multiplying the percent usage of each stage by the number of that stage and summing the products (range i s 100 to 500). Thus, two separate WASs were derived for each subject, one expressing the le v e l of reasoning in support of c a p i t a l punishment (pro) and a second expressing the l e v e l of reasoning in opposition to c a p i t a l punishment (con). The GSS i s a more q u a l i t a t i v e measure of moral reasoning. GSSs were assigned on the basis of the pattern of percent usage and took the form of pure and mixed stage scores. A pure stage score was assigned i f one stage had greater than 25% of the scored responses ( i . e . , Stages 1,2, 3, 4, 5). A mixed stage score was assigned i f two stages each had greater than 25% of the scored responses ( i . e . , Stages 1/2, 2/3, 3/4, 4/5). Two separate GSSs were derived for each subject, one expressing the le v e l of reasoning in support of c a p i t a l punishment (pro) and a second expressing the le v e l of reasoning in opposition to c a p i t a l punishment (con). To describe the o v e r a l l l e v e l of moral development, a t o t a l weighted average score and a t o t a l global stage score were also generated for each subject. These composite scores encompassed a l l reasoning (both pro and con) but weighted the chosen position more heavily than the nonchosen po s i t i o n . The weights applied in t h i s procedure are 3 for the chosen position, 2 for the nonchosen, and 1 for a guess score (a score assigned when interview judgments do not meet a l l of the c r i t i c a l i n d i c a t o r s ) . These weights are those recommended by Colby et a l . (in press) and r e f l e c t the assumption that better reasoning i s used to support a chosen position. 45 Thus, a t o t a l of six measures of moral reasoning were derived for each subject: a t o t a l GSS and a t o t a l WAS (encompassing both pro and con reasoning), a WAS and a GSS for reasoning supporting c a p i t a l punishment, and a WAS and a GSS for reasoning opposing c a p i t a l punishment. It i s important to address at t h i s point a variation in procedure concerning the material scored for moral reasoning in th i s study and the structured interview t y p i c a l l y administered in the assessment of moral reasoning. It would be reasonable to suggest that the reasoning statements about c a p i t a l punishment are actually a subsample of the reasoning statements generated in an interview (comprised of three dilemmas). It i s not unlike examining the statements from one dilemma as compared to the statements from an entire interview. Kohlberg and Elfenbein (1981) scored the responses of their subjects on the c a p i t a l punishment issue separately from the rest of their interviews and found that subjects' stage of reasoning about the death penalty did not d i f f e r greatly from their o v e r a l l stage score. In fact, they reported that in only 10% of the cases did subjects' GSS d i f f e r from their stage of reasoning about c a p i t a l punishment, and the difference was never greater than one stage. This provides support for the u t i l i z a t i o n of one dilemma or issue in t h i s study. The scoring of the moral reasoning materials was done bli n d l y ( i . e . , without knowledge of subjects' scores on the other measures). Interrater r e l i a b i l i t y was established by a second independent rater who b l i n d l y scored the materials of 16 randomly selected subjects. There was 75% agreement in GSS 46 scoring ( i . e . , both pure and mixed stages) and r e l i a b i l i t y on the WASs was r = .89. This i s consistent with previous research in t h i s area (Colby et a l . , 1983; Walker, in press). Attitudes toward c a p i t a l punishment. Attitudes toward c a p i t a l punishment were assessed by the Capital Punishment Questionnaire (CPQ). The CPQ was developed by Palys (1981) out of a skepticism regarding "for-or-against" formats of questions t y p i c a l l y asked in public opinion p o l l s , which were perceived as a p o t e n t i a l l y false dichotomy. Two forms of the questionnaire have been developed, each of which has 24 items. The correlation between the two forms i s .95. Form A was a r b i t r a r i l y selected for use in t h i s study. Scores on the questionnaire involve summing the ratings (after L i k e r t , 1932) and can t h e o r e t i c a l l y range from -72 (extremely a n t i - c a p i t a l punishment) through 0 (neutral) to +72 (extremely pro-capital punishment). I n i t i a l tests of the CPQ, reported by Palys (1981), generated a range from -66 to +62, with a o v e r a l l mean of 7.42 and a standard deviation of 32.62. For each subject, a t o t a l score on the CPQ was calculated. Furthermore, to f a c i l i t a t e analyses (e.g., trend analyses, analyses of variance), subjects were formed into groups, such that roughly 20% of the subjects ( i . e . , groups of 14 or 15) f e l l into one of five q u i n t i l e s . (Means for the fiv e groups were -51.4, -24.7, 4.7, 20.6, and 41.8.) 47 Results Normative Data Participants were drawn in equal numbers from three educational levels (introductory and senior undergraduate, and graduate) to generate a range on the various measures. To examine these ranges, four analyses of variance were conducted on the mean complexity scores of the PCT and the issue composition, the t o t a l WAS of moral development, and the t o t a l score on the CPQ, a l l for educational l e v e l . Except for the last analysis, a l l were s i g n i f i c a n t and means were in the predicted d i r e c t i o n ( i . e . , graduate greater than senior undergraduate greater than introductory undergraduate). The ANOVA for mean scores on the PCT for the educational le v e l variable, F(2, 69) = 18.786, £ < .0001, revealed between-group means of 1.49, 1.71, and 2.23 for the introductory, senior, and graduate students, respectively, with a range of 1.0 to 3.5. The ANOVA for mean scores on the issue composition for these three groups was also s i g n i f i c a n t , F(2, 69) = 3.87, p_ < .03, with corresponding group means of 1.91, 1.97, and. 2.30, and a range of 1.17 to 3.50. These means are t h e o r e t i c a l l y consistent with others reported in the l i t e r a t u r e . For example, Tetlock (1983a) has reported means of integrative complexity ranging from 1.83 to 2.61 for groups of undergraduates and mean scores in the range of 2.17 to 3.20 for presidents (Tetlock, 1981), supreme court judges, and parliamentarians (Tetlock, 1984). Examining t o t a l WASs for educational group yielded s i m i l a r l y s i g n i f i c a n t r e s u l t s , F(2, 69) = 14.99, p_ < .001, with 48 means of 323.79, 342.50, and 379.96, and a range of 263 to 440 (or, expressed in GSSs, Stage 2/3 to 4/5). As mentioned above, the ANOVA of t o t a l scores on the CPQ for educational groups was not s i g n i f i c a n t , p_ = .19. This, however, i s not surprising since there i s no theoretical reason to expect differences on this attitude measure as a function of educational l e v e l . A point worthy of mention at t h i s time is that these differences across educational levels for some of the variables may be seen to represent a necessary "confound" in some of the relationships under investigation. However, the primary objective in the selection of subjects on the basis of educational l e v e l was to generate ranges in performance. Having accomplished t h i s and because educational l e v e l was not of theoreti c a l interest, educational l e v e l was excluded from subsequent analyses, consistent with previous research (e.g., Walker et a l . , in press). Conceptual and Integrative Complexity Hypothesis I addressed the r e l a t i o n s h i p between conceptual complexity and integrative complexity. It i s assumed that since these two measures are assessing somewhat similar, yet d i f f e r e n t , dimensions of the decision-making environment ( i . e . , general as opposed to s p e c i f i c decision-making processes) a moderately strong relationship w i l l be found between them. As a test of t h i s relationship, a Pearson product-moment cor r e l a t i o n was calculated betv/een the mean scores on the PCT and the issue composition y i e l d i n g , as expected, a s i g n i f i c a n t moderate 1 c o e f f i c i e n t of .33, p < .002. This rel a t i o n s h i p can be i l l u s t r a t e d through an examination 49 of the r e l a t i v e levels of complexity on both measures. A median s p l i t of the scores on the PCT and the issue composition is summarized in Table 1. Although further analyses might be somewhat redundant, i t i s interesting to note that a contingency analysis also revealed a s i g n i f i c a n t pattern, x 2(1, N = 72) 4.50, p_ < .03, indicating that 63.8% of the subjects were c l a s s i f i e d as either low or high on both measures. The above analyses demonstrate the similar nature of the two measures; i t is equally important, however, to examine the dimension(s) on which they d i f f e r . The contingency analysis c l e a r l y displays the relationship between the PCT and the issue composition in i t s analysis of r e l a t i v e l e v e l s ; however, i t obscures any picture of their absolute differences. To address th i s issue, a 2(Sex) x 2(Complexity measures: PCT/issue composition) ANOVA was conducted with repeated measures on the last factor, with mean complexity scores as the dependent variable. An issue to be addressed here i s whether or not i t i s appropriate to use the complexity scores from the two measures to represent the two levels of the dependent variable in this analysis. Although a moderate relationship between the two measures was described above, indicating some differences, they are c l e a r l y measures of the same variable. Their scoring procedures are id e n t i c a l ( i . e . , a str u c t u r a l analysis of written materials of s u f f i c i e n t length) and both assess the same processes ( i . e . , the structure underlying thought). This suggests that i t i s defensible to compare the scores yielded by these two measures; that i s , to examine these measures as two lev e l s of a single factor. The resultant ANOVA summary table i s 50 Table 1 Median S p l i t of Scores on the PCT and Issue Composition Issue Composition PCT Low High Low 23 13 High 13 23 51 presented in Table 2. The only s i g n i f i c a n t effect revealed by t h i s analysis was for the complexity factor, F_(1, 70) = 11.231, p_ < .001. The mean score on the issue composition was s i g n i f i c a n t l y higher than that on the PCT (2.06 vs. 1.81). Although t h i s difference of .25 is s i g n i f i c a n t , i t i s not sizable, given that i t represents only one-half of a scorable i n t e r v a l on the complexity scale. This finding suggests that both measures may be assessing the same processes, but one i s allowing for a greater expression. Complexity and Attitudes toward Capital Punishment Hypothesis II addressed the relationship between attitudes toward c a p i t a l punishment and complexity (as a generic term including both conceptual and integrative complexity). S p e c i f i c a l l y , a c u r v i l i n e a r relationship was hypothesized between the scores on the Capital Punishment Questionnaire and the mean scores on the measures of complexity. To test t h i s , a 2(Sex) x 5(CPQ q u i n t i l e groups) x 2(Complexity measures: PCT/issue composition) ANOVA was calculated with repeated measures on the last factor, and with mean complexity scores as the dependent variable. The resultant ANOVA summary table is presented as Table 3. Sex was not s i g n i f i c a n t either as a main effect or in interaction with other factors. Sign i f i c a n t main effects were revealed for the other two factors, q u a l i f i e d by an interaction between them, F(4, 62) = 11.917, p_ < .001. To is o l a t e the locus of t h i s interaction, a test of simple main e f f e c t s of CPQ groups was performed on both measures of complexity. Results of the 52 Table 2 Summary of the Analysis of Variance for the Mean Complexity Scores Source of variance df MS F Sex 1 0. 1 54 0.387 Error (between) 70 0.397 Complexity measures 1 2.253 11 .231* Sex x Complexity measures 1 0.009 0.045 Error (within) 70 0.201 *p < .001. 53 Table 3 Summary of the Analysis of Variance for the Mean Complexity  Scores, Including Simple E f f e c t s and Trend Analysis Source of variance df MS F Sex 1 0.134 0 .363 CPQ groups 4 1.010 2 .745* Sex x CPQ groups 4 0.209 0 .567 Error (between) 62 0.368 Complexity measures 1 1 .483 1 1 .767** Sex x Complexity measures 1 0.119 0 . 941 CPQ groups x Complexity measures 4 1 .502 11 .917** Simple main effects CPQ groups for PCT 4 0.331 1 .217 Error (within) 67 0.272 CPQ groups for issue composition 4 2. 1 06 1 0 .060** Trend analysis Linear function 1 0.019 0 .089 Deviation from linear 3 2.802 1 3 . 387** Quadratic function 1 6.920 33 .057** Deviation from quadratic 2 0.723 3 .456* (cont'd) 54 Table 3 (cont'd) Source of v a r i a n c e df MS F Cubic f u n c t i o n 1 0. .611 2, .921 Dev i a t ion from cu b i c 1 0. .805 3, .847 E r r o r ( w i t h i n ) 67 0. .209 Sex x CPQ groups x Complexity measures 4 0, . 1 37 1 .090 E r r o r ( w i t h i n ) 62 0, . 1 26 *p_ < .05. **2 < .001 . 55 test on the PCT were not s i g n i f i c a n t , indicating that the hypothesis was not supported for this measure of complexity. Results of the test on the issue composition were highly s i g n i f i c a n t however, F(4, 67) = 10.06, p_ < .0001. Further examination of this effect was performed by way of a trend analysis as a means of exploring the relationship between th i s measure of complexity and scores on the attitude questionnaire. Trend analyses examine various functions ( l i n e a r , quadratic, cubic, quartic) which best describe the relat i o n s h i p between variables. In this case, the quadratic function accounted for the greatest proportion of variance, F( 1 , 67) = 33.057, p_ < .0001, f i t t i n g what can best be described as a c u r v i l i n e a r pattern, with 7j = .61. Figure 1 displays this relationship. Residual variance was i n s u f f i c i e n t to provide s i g n i f i c a n t F-ratios for any other function. Figure 1 c l e a r l y i l l u s t r a t e s the c u r v i l i n e a r nature of the rel a t i o n s h i p between the issue composition as a measure of integrative complexity and the c a p i t a l punishment questionnaire. More extreme scores in either d i r e c t i o n on the CPQ correspond to more s i m p l i s t i c organization or structure of thought about c a p i t a l punishment, whereas more moderate scores are characterized by a greater complexity of presentation of ideas. Conceptual Complexity and Moral Reasoning Hypothesis III addressed the relationship between conceptual complexity and moral reasoning. The expectation was that there would be a posit i v e association between mean scores on the PCT and the t o t a l WAS. The PCT was the measure used in th i s case since i t i s the one t y p i c a l l y assumed to represent 56 Figure 1. Mean complexity scores on the issue composition as a function of attitude toward c a p i t a l punishment. 2.S A t t i t u d e T o w a r d C a p i t a l P u n i s h m e n t ( C P Q Groups) 58 c h a r a c t e r i s t i c levels of complexity. Moreover, the t o t a l WAS of moral reasoning i s a composite of scored responses from the issue composition as well as from the statements both supporting and opposing c a p i t a l punishment. It would therefore be unjustified--both conceptually and s t a t i s t i c a l l y — t o include the measure of integrative complexity (as assessed from the issue composition) in this analysis. The relationship between conceptual complexity and moral reasoning was examined by a series of analyses. F i r s t , a Pearson product-moment cor r e l a t i o n was calculated between mean scores on the PCT and the overall measure of moral reasoning (the t o t a l WAS). This yielded, as expected, a s i g n i f i c a n t c o e f f i c i e n t of + . 4 3 , p_ < . 0 0 1 , indicating correspondence between these two measures. This correspondence was further examined by conducting a median s p l i t on the PCT variable, then a 2(Sex) x 2(Complexity: high/low) ANOVA, with the t o t a l WAS as the dependent variable. The resultant ANOVA summary table i s presented in Table 4 . There were no main ef f e c t s nor interaction with sex, so i t was excluded from subsequent analyses. A main effect was found, however, for complexity, F( 1 , 6 8 ) = 1 1 . 4 3 9 , p_ < . 0 0 1 . The t o t a l WAS for high complexity subjects was 3 6 4 . 9 4 and the t o t a l WAS for low complexity subjects was 3 3 2 . 5 6 , a difference roughly equivalent to one-third of a stage. The essence of the above two results are perhaps most graphically depicted in Table 5 which presents a median s p l i t on both the PCT and t o t a l WAS. A contingency analysis revealed a si g n i f i c a n t pattern, X 2 ( 1 , N = 7 2 ) = 6 . 7 2 , p < . 0 1 , indicating 59 Table 4 Summary of the Analysis of Variance for Total Weighted Average Scores Source of variance df MS F Sex 1 744.851 0.463 Complexity 1 18410.684 11.439* Sex x Complexity 1 518.949 0.322 Error (between) 68 1609.456 *p_ < .001 . Table 5 Median S p l i t of Scores on the PCT and Moral Reasoning Moral Reasoning PCT Low High Low 24 12 High 12 24 61 that two-thirds of the subjects were c l a s s i f i e d as either low or high on both measures. A l l these analyses c l e a r l y support the relati o n s h i p between these two hi e r a r c h i c a l theories. Moral Reasoning (Pro versus Con) and Attitude towards Capital Punishment The f i n a l two hypotheses both addressed issues in moral reasoning and are therefore probably best examined together. Hypothesis IV predicted a posi t i v e , linear relationship between attitudes toward c a p i t a l punishment (as derived from the issue composition and the CPQ) and moral reasoning, with increasing opposition with higher l e v e l reasoning. Hypothesis V was included to examine differences in the reasoning of subjects in their support of, versus opposition to, c a p i t a l punishment--as a function of their chosen and nonchosen p o s i t i o n — w i t h the expectation that higher reasoning would be associated with the chosen position. To examine these hypotheses, a between-within design ANOVA was conducted with three independent variables: complexity (median s p l i t on the PCT), subjects' attitude toward c a p i t a l punishment (pro/con), and position on the issue of c a p i t a l punishment (pro/con). The median s p l i t on the PCT was included as a factor to examine any possible interactions. (Incidentally, by chance, there were equal numbers of subjects with pro and con attitudes toward c a p i t a l punishment.) The dependent variable in t h i s analysis was the WAS for both pro and con positions on the issue. The design, then, was a 2(Complexity: high/low) x 2(Subjects' atti t u d e : pro/con) x 2(Position: pro/con) ANOVA with repeated measures on the last 62 factor. The resultant ANOVA summary table i s presented in Table 6. A main eff e c t for complexity was revealed, as discussed in the previous section, with no interactions between th i s and any other factor, so the resultant pattern i s true for both high and low complexity subjects. The main effect for position was highly s i g n i f i c a n t , but q u a l i f i e d by an interaction between position and attitude, F( 1 , 68) = 4.716, p_ < .05. Recall that Hypothesis V predicted higher moral reasoning on the chosen position (either pro or con) than on the nonchosen position. However, the means, as presented in Table 7, indicate higher moral reasoning on the con position for both "pro" and "con" attitude subjects. Analyses of the simple main ef f e c t s of position for both "pro" and "con" attitude subjects were s i g n i f i c a n t , Fs ( 1 , 35) = 6.228, 2 < .02, and 27.021, 2 < -001, respectively. Thus, subjects, when j u s t i f y i n g opposition to c a p i t a l punishment, evidence higher moral reasoning than when they support i t . This seems especially true for the "con" attitude subjects. That i s , subjects do not always use higher moral reasoning to support their own position; in this case at least, they use higher moral reasoning to j u s t i f y opposition to c a p i t a l punishment (regardless of the chosen p o s i t i o n ) , contrary to Hypothesis V and implications drawn from Kohlberg's scoring system. In addressing the cognitive-developmental assumption of structured whole, the GSSs derived from pro and con reasoning were examined. Recall that GSSs take the form of pure and mixed stage scores, providing a 9-point scale ( i . e . , Stages 1, 1/2, 2, 63 Table 6 Summary of the Analysis of Variance for Pro and Con Weighted  Average Scores, Including Simple Main Effects Source of variance df MS F Complexity 1 28785. 543 9. ,301** Att i tude 1 7015. ,551 2. ,267 Complexity x Attitude 1 6453. ,328 2. ,085 Error (between) 68 3094. ,823 Position 1 34121 . ,098 29. ,096*** Complexity x Position 1 845. ,555 0, .721 Attitude x Position 1 5529. .996 4, .716* Simple main effects Position for pro attitude 1 6347. .250 6, .228* Error (within) 35 1019, .200 Position for con attitude 1 36137, .250 27, .021*** Error (within) 35 1 337 , .400 Complexity x Attitude x Position 1 1 906, .666 1 .626 Error (within) 68 1 172, .706 *p_ < .05. **p_ < .01. ***p_ < .001 . Table 7 Mean Weighted Average Scores for Pro and Con Positions as a  Function of Attitude toward Capital Punishment Attitude toward Capital Punishment Position Pro Con Pro 328.4 332.6 330.5 Con 347.2 377.4 362.3 337.8 355.0 65 2/3, 3, 3/4, 4, 4/5, 5). On th i s scale, 76.4% of the subjects used either the same or an adjacent stage in their reasoning of support of, and opposition to, c a p i t a l punishment. Surprisingly, 23.6% of the subjects evidenced differences of one stage or greater in their reasoning on the two sides of t h i s issue. That i s , almost one-quarter of the t o t a l subject population supplied reasoning in apparent v i o l a t i o n of the structured whole assumption. The . requirements of th i s assumption, however, have never formally been presented. It i s rather loosely assumed that individuals have cognitive structures which are consistent across i s s u e s — s u b j e c t s should be evidencing the same stage in their analysis of both positions. In this study, almost 25% of the subjects supplied reasoning on these two positions with at least a f u l l stage discrepancy. The above analysis examined the relationship between attitude and moral reasoning with attitude as a dichotomous variable (pro/con). An alternate approach would be to assess this relationship u t i l i z i n g a continuous measure of the attitude variable (such as the CPQ). As would be expected, the CPQ and the dichotomous measure of attitude were highly r e l a t e d — t h e c o e f f i c i e n t yielded by a p o i n t - b i s e r i a l c o r r e l a t i o n a l analysis was +.78, p < .001. To test the relationship between the CPQ and moral reasoning, a Pearson product-moment cor r e l a t i o n was calculated between t o t a l CPQ scores and t o t a l WASs, y i e l d i n g a s i g n i f i c a n t c o r r e l a t i o n of -.38, JD < .001. Thus, greater opposition to c a p i t a l punishment corresponds to higher levels of moral 66 reasoning. This finding i s i l l u s t r a t e d in Table 8 which displays the percentage of males and females at each moral stage who oppose c a p i t a l punishment. Also presented in this table are the findings of Kohlberg and Elfenbein (1981) who had an exclusively male sample. A similar pattern i s evidenced for both male and female subjects ( i . e . , greater opposition with higher stage), but females appear to be more opposed to c a p i t a l punishment. This difference was examined by conducting an ANOVA of the t o t a l CPQ scores with sex as the independent variable. A s i g n i f i c a n t e f f e c t was obtained, F( 1 , 70) = 4.354, p_ < .05, with females showing greater opposition to c a p i t a l punishment than males (means = -10.19 vs. 6.08, r e s p e c t i v e l y ) . Table 8 Percentage of Subjects at each Moral Stage Who Oppose Capital Punishment Sex Moral Stage Males Females Kohlberg & Elfenbein '( 1 981 ) - Males only 1 , 1/2 - - 0 2 - - 1 1 2/3 0 33.3 0 3 37.5 60.0 20 3/4 29.4 52.4 41 4 40.0 1 00.0 36 4/5 100.0 1 00.0 1 00 5 — — 100 Note. A dash indicates that no subjects were found at that stage. 68 Discussion The previous section presented the results of this study. In p a r t i c u l a r , a s i g n i f i c a n t , moderate co r r e l a t i o n was found between the comparable measures of conceptual and integrative complexity. Integrative complexity was found to relate in a c u r v i l i n e a r way to an attitude measure of c a p i t a l punishment; no s i g n i f i c a n t relationship was found between the attitude measure and conceptual complexity. Conceptual complexity and moral reasoning evidenced a moderately strong rel a t i o n s h i p . Moral reasoning and attitudes toward c a p i t a l punishment were found to relate l i n e a r l y , with opponents of c a p i t a l punishment substantiating their position with higher lev e l s of reasoning than proponents. Moreover, regardless of chosen position, subjects used higher levels of moral reasoning to oppose c a p i t a l punishment than they did to support i t . This present section more f u l l y discusses the implications of these findings. Conceptual and Integrative Complexity The Paragraph Completion Test i s a frequently used and well-validated measure of an individual's l e v e l of conceptual complexity. It has been shown to be a useful instrument, p a r t i c u l a r l y in personality research examining c h a r a c t e r i s t i c s and styles of thought in r e l a t i o n to other s t r u c t u r a l and d i s p o s i t i o n a l variables (Schroder et a l . , 1967). A productive offshoot of t h i s measure i s found in integrative complexity—the assessment of the structural complexity of thought from written and verbal materials and the subsequent stimulus of innovative research in human decision making as a function of environmental and s i t u a t i o n a l variables. Such an approach represents a real 69 advance in the soc i a l psychological study of human behavior outside the confines of the laboratory. It also has so c i o l o g i c a l , p o l i t i c a l and h i s t o r i c a l relevance in i t s analysis of s o c i a l l y s i g n i f i c a n t events. The basic premise of both of these measures is that in any written or verbal material on decision-making issues of s u f f i c i e n t length, individuals are providing manifestations of the structure underlying thought. Even though the thoughts may originate from a wide variety of sources (e.g., p o l i t i c a n s , authors, students) and the structure may be sensitive to forces in the environment (e.g., stress) the basic premise remains unaltered. This i s the conceptual base on which both theories are grounded. Less obvious in the l i t e r a t u r e , however, i s some form of empirical base or support demonstrating that the PCT and measures of integrative complexity are s u f f i c i e n t l y correspondent. This is not surprising given the " i n a c c e s s i b i l i t y " of subjects in studies of integrative complexity ( i . e . , h i s t o r i c a l and/or public figures) for the purpose of completing the PCT. This i s the rationale for the inclusion of the issue composition task in t h i s study: the examination of the rel a t i o n s h i p between conceptual and integrative complexity within subjects by the measures of the PCT and the composition on c a p i t a l punishment. The moderate correlation found (r = .33) adequately defines an area in which these two information processing measures overlap. What they share i s the way in which both assess or characterize the structural underpinnings of thought--the 70 degree(s) of d i f f e r e n t i a t i o n and integration. This is their conceptual and empirical interdependence or base. Clearly, however, this correlation i s less than maximal suggesting that these two measures are also accessing separate properties ( i . e . , areas in which there i s no overlap). This difference became apparent in the analysis which indicated that scores on the issue composition were s i g n i f i c a n t l y higher than those on the PCT. These higher scores of the issue composition may be a function of the characteristcs of the task. That i s , in completing the PCT, individuals are presented with six sentence stems, each of which is to be completed in an imposed time l i m i t of 2 minutes. The issue composition, on the other hand, involves the presentation of a topic which individuals are asked to discuss, without the same r i g i d time r e s t r i c t i o n s . This leads to an interesting empirical question: Do the constraints of the PCT a r t i f i c i a l l y l i m i t the level of complexity in which individuals respond? Perhaps th i s could be addressed by removing the constraints ( i . e . , time, number of topics, completion versus discussion) and having individuals respond to the stems as topics of discussion (and hence equivalent to the issue composition) as opposed to sentence beginnings. Conversely, perhaps the presentation of " c a p i t a l punishment" as a paragraph stem would render i t equivalent to the other stems comprising the PCT. Higher scores on the former and lower scores on the l a t t e r would support the PCT constraint hypothesis. Such an interpretation focuses exclusively on the finding 71 that the issue composition e l i c i t e d higher scores than the PCT. Certainly, t h i s is the major difference, but i t may well extend beyond t h i s . That i s , i f the score of every individual was consistently higher on the issue composition than on the PCT, the correlation between these two measures would be higher as well (as i t represents the variation of r e l a t i v e l e v e l s ) . Such was not the case. The moderate co r r e l a t i o n suggests that the issue composition may be more than just a better elaborated PCT stem; there may be q u a l i t a t i v e differences as well for even when corrected for attentuation, the co r r e l a t i o n s t i l l retains a moderate value (r = .42). That i s , the issue composition was on a topic of considerable controversy, and one that may be said to be more personally meaningful. The PCT s t r i v e s to tap issues relevant to a more generalized realm of decision making, a realm t y p i c a l l y comprised of more mundane and less controversial considerations. To this extent, the PCT may be seen to draw more spontaneous responses; the measure of integrative complexity may be seen to draw more deliberated responses. Perhaps herein l i e s a difference in q u a l i t y . The PCT and the measure of integrative complexity are s i b l i n g s from a common parentage. They share the same genetic inheritance, but d i f f e r in their appearance, orientation and intensity. Complexity and Attitudes toward Capital Punishment Schroder et a l . (1967) have proposed a d i s t i n c t i o n between st r u c t u r a l l y simple and s t r u c t u r a l l y complex attitudes. They suggested that simple attitudes are based on a narrow range of highly salient information and are consequently expected to be 72 more categorical and extreme. Complex attitudes, on the other hand, are supposedly based on a broader range of relevant information (encompassing both consonant and dissonant arguments) res u l t i n g in a less categorical and less extreme ( i . e . , more moderate) attitude. L i n v i l l e and colleagues ( L i n v i l l e , 1982, in press; L i n v i l l e & Jones, 1980) have provided empirical support for t h i s theoretical proposition using a t r a i t sorting technique with f a v o r a b i l i t y and unfavorabi1ity ratings toward a s p e c i f i c target object. They found extreme evaluations (both pro and con) related to more s i m p l i s t i c structures whereas less extreme evaluations related to more complex structures. Tetlock (1984) has also reported a c u r v i l i n e a r pattern in the complexity of discourse of parliamentarians as a function of their p o l i t i c a l ideologies. A similar c u r v i l i n e a r pattern was found in this study wherein more extreme (pro and con) attitudes toward c a p i t a l punishment (as determined by the CPQ) were associated with lower levels of complexity (as determined by the issue composition) whereas more moderate attitudes were associated with increasing levels of complexity. This substantiates the claim by Schroder et a l . and elaborates on the work of L i n v i l l e and others in at least two important ways: (a) Integrative complexity was employed in t h i s study (the assessment of conceptual structure as a function of d i f f e r e n t i a t i o n and integration), as opposed to the t r a i t - s o r t i n g technique employed by L i n v i l l e who assessed complexity as a function of discrimination and d i f f e r e n t i a t i o n , the l a t t e r being subsumed in the preliminary processes of the 73 former. (b) Even with t h i s alternate measure, the "complexity-extremity" e f f e c t , as L i n v i l l e c a l l s i t , was supported in the important and controversial a t t i t u d i n a l realm of c a p i t a l punishment. Moreover, this relationship c l e a r l y demonstrates the content-free nature of the measure of complexity in the absence of any uni d i r e c t i o n a l effect or bias. Clearly, these results are also consistent with those reported by Tetlock (1984). Recall that Tetlock found extreme s o c i a l i s t s and conservatives to display similar low levels of integrative complexity and s i g n i f i c a n t l y lower levels than moderate s o c i a l i s t s and conservatives on current policy issues and policy-making processes. Tetlock (1984) accounted for these findings with a value pluralism interpretation. It i s assumed that advocates of ideologies in which freedom and equality are both highly valued ( i . e . , moderate socialism) are under greater pressure to consider such values in more int e g r a t i v e l y complex terms than are advocates of ideologies that place greater weight on only one value ( i . e . , freedom for conservatives). The p r i n c i p l e s of such an interpretation might well apply in accounting for the findings of t h i s study. That i s , perhaps i t could be said that embedded in the issue of c a p i t a l punishment are the two values of l i f e and law. Following through with this assumption, i t would seem reasonable to suggest that extreme positions on this issue embrace one of these two values, whereas a more moderate position would encompass both. That i s , extreme opponents of c a p i t a l punishment might value law (and perhaps to the exclusion of l i f e considerations) whereas extreme proponents might value l i f e . A 74 more moderate position might embrace these two values equally and thus pr e v a i l upon such advocates to consider t h i s issue in more complex terms. This remains speculation and subject to empirical v a l i d a t i o n , but such a model parsimoniously provides a perspective in which these results may be interpreted. It i s interesting to note that support for the hypothesized relationship between complexity and attitudes toward c a p i t a l punishment was generated through the • issue composition (integrative complexity) and not through the PCT (conceptual complexity). This perhaps stems from the differences between these two measures as previously discussed. The PCT i s seen as a more global or general assessment of an individual's c h a r a c t e r i s t i c l e v e l of conceptual complexity. The issue composition, on the other hand, i s more of a s p e c i f i c measure of the way in which an individual processes information relevant to a p a r t i c u l a r domain, and, in t h i s case, one that is especially controversial. That i s , because of t h i s controversial nature of c a p i t a l punishment, i t has had great exposure and v i s i b i l i t y in the form of debates, speeches, and a r t i c l e s in the popular press. Moreover, part of what makes i t controversial i s that i t is an issue that c a l l s into play an individual's moral, philosophical, and r e l i g i o u s ideology which l i e s at the core of strongly held convictions and b e l i e f s and which perhaps predisposes one to p a r t i c u l a r orientations--an idea not unlike that proposed by Harvey et a l . (1961). These two factors in combination, the high exposure of an issue with i t s moral, philosophical and/or r e l i g i o u s component, may work to produce effects to which the more mundane PCT i s i n s e n s i t i v e . It i s 75 perhaps reasonable to assume that c a p i t a l punishment l i e s on a d i f f e r e n t plane of relevance than do stems such as "when I don't know what to do" or "confusion" that form the PCT. Tetlock's (1984) value pluralism model is also relevant here. It could be proposed that the values inherent in the issue of c a p i t a l punishment (and the ways in which these values are brought into play) d i f f e r in some substantial way from the value(s) encompassed in the stems of the PCT. That i s , perhaps each of the PCT stems has only a single value attached to i t , in contrast to the two values of the issue composition. Such a difference would manifest i t s e l f in the r e l a t i v e scores of each of these measures, for i f the concurrent consideration of two values predisposes an individual to consider an issue in more complex terms, the consideration of only a single issue might necessarily p r e c i p i t a t e more s i m p l i s t i c thought. This i s perhaps the q u a l i t i a t i v e difference between the PCT and the issue composition and the reason for the d i f f e r e n t i a l results in r e l a t i o n to the c a p i t a l punishment questionnaire. Conceptual Complexity and Moral Reasoning Kohlberg (1981) has described his theory of moral reasoning as being a cumulative hierarchy of cognitve complexity in that each successive stage i s a more d i f f e r e n t i a t e d and better integrated structure. These terms also represent the integral processes upon which Schroder et a l . have elaborated their formulation of conceptual complexity. D i f f e r e n t i a t i o n characterizes the lower range of the simplicity-complexity continuum and represents the necessary condition for integration, which i s c h a r a c t e r i s t i c of the upper range. Both 76 theories focus on the cognitive structures that underlie thought, with the structures becoming increasingly complex as one attains subsequent stages or levels along the continuum. Conceptually, at least, these stages seem to address themselves to similar processes. This i s demonstrated by the moderate cor r e l a t i o n reported e a r l i e r (r_ = .43), where increasing stages of moral reasoning correspond to increasing levels of conceptual complexity. The subsequent analyses reported s i m i l a r l y displayed such a relationship. This i s consistent with the results reported by Sullivan et a l . (1970); what d i f f e r s i s the strength of the relationship or the magnitude of the co r r e l a t i o n . Sullivan et a l . presented an ove r a l l c o e f f i c i e n t of .62 between moral development and conceptual l e v e l . This difference may be attri b u t a b l e to several factors. F i r s t , Sullivan et a l . employed an older method of scoring for both the PCT and moral reasoning development. The PCT was scored according to the 1964 manual of Hunt and Halverson in which responses were assigned scores of 0 to 4 representing stages of the now abandoned developmental orientation or model. The moral reasoning protocols were scored according to Kolhberg's (1958) procedure which has subsequently undergone substantial revision resulting in a r e c l a s s i f i c a t i o n of many of the stage-typed responses. For example, much of what was formerly c l a s s i f i e d as Stages 4 or 5 is presently c l a s s i f i e d as Stage 3. This study employed the better validated, most recent systems of coding for both conceptual complexity and moral reasoning. Second, subjects in the Sullivan et a l . study were drawn from three age l e v e l s : 12, 14, and 17 years. 77 Subjects in t h i s study were drawn from three educational levels with ages ranging from 17 to 45 years resulting in a less r e s t r i c t i v e sample with a greater range in ages. Third, this present study assessed moral reasoning on the basis of a composition and statements on the single issue of c a p i t a l punishment. The rationale for such an approach comes from the work of Kohlberg and Elfenbein (1981) and Gibbs and Widaman (1982) and has been discussed e a r l i e r . Suffice i t to say that i t represents a departure from the method employed by Sullivan et a l . which contained nine hypothetical c o n f l i c t stories and corresponding sets of probing questions. Given that both moral reasoning and conceptual complexity assess the structural complexity of thought, i t is interesting to note that there is less than perfect correspondence between them. That i s , they are not synonomous measures. The differences that exist may, in part, be due to the inadequacies and v a r i a b i l i l t y in the methods of measurement. For example, diff e r e n t methods are used in the derivation of item scores increasing measurement error (paragraph completions versus answers relevant to p a r t i c u l a r dilemma issues and questions) just as d i f f e r e n t methods are employed in the c a l c u l a t i o n of ov e r a l l scores increasing the potential for information loss (mean scores versus weighted average scores). Nevertheless, the fact that a relationship does exist between conceptual complexity and moral reasoning can be seen as supporting Kohlberg's rather controversial claim of hierarchy. There appears to be a st r u c t u r a l hierarchy in that each succeeding stage i s both better d i f f e r e n t i a t e d and integrated. This 78 finding i s congruent with the findings of other more direct studies (e.g., Moran & Joniak, 1979; Rest, 1973; Walker et a l . , in press). The controversy, however, does not r e a l l y come from the hierarchy claim. It stems from a claim of content h i e r a r c h y — t h e notion that some advanced stage is more "moral" than that preceding i t . It should be mentioned, however, that the correlation procedure, although useful in examining the relatedness of constructs, f a i l s to provide s p e c i f i c information about how constructs are related. Perhaps i t could be said that reasoning about moral issues represents a subset of the t o t a l reasoning or decision-making environment. To the extent that conceptual complexity attempts to examine decision-making processes operative in the t o t a l environment, perhaps i t could also be said that moral reasoning i s a part of this greater whole. In such an interpretation, the relationship between moral reasoning and conceptual complexity may be analogous to a part-whole c o r r e l a t i o n , describing domain s p e c i f i c ways of assessing some more general, yet i n t e r n a l l y consistent, c h a r a c t e r i s t i c . Moral Reasoning (Pro versus Con) and Attitude towards Capital  Punishment Kohlberg has provided an extensive manual (Colby et a l . , in press) for the coding of reasoning about selected moral issues, of which c a p i t a l punishment is one. This manual i s primarily based on the responses from his longitudinal sample, from which he has been able to generate moral reasons at a l l stages for both sides of a l l issues. These stages are based on the structure, and not the content, of an individual's moral 79 reasoning. Content refers to one's attitudes or b e l i e f s or judgments and behavioral choices, whereas structure or form refers to the reasoning that underlies and j u s t i f i e s t h i s content. Content i s irrelevant in determining stage of development. In fact, opposing positions regarding moral issues can be defended at each stage of moral development. Thus, form and content are conceptually independent. However, the approach would be of l i t t l e theoretical or s o c i a l significance i f there were no empirical relationship between form and content. The development of higher levels of moral reasoning would hardly be of much consequence unless i t could be argued that i t produces judgments and behaviors that could be considered more moral. Thus, an aim of th i s study was to examine the relationship between attitudes toward c a p i t a l punishment and l e v e l of moral reasoning: between the form and content of moral reasoning. Kohlberg and Elfenbein (1981) examined th i s relationship in a study with 30 males and found that at Stage 3 and below, almost a l l subjects argued that c a p i t a l punishment i s right under cert a i n circumstances; at Stage 4, subjects were divided; and at Stage 5, subjects rejected the use of c a p i t a l punishment. The results of the present study, as reported e a r l i e r , demonstrate a similar pattern while extending upon the work of Kohlberg and Elfenbein (1981). S p e c i f i c a l l y , Kohlberg and Elfenbein (1981) used an exclusively male sample (with the questionable procedure of repeated observations on single subjects such that 105 data points were presented for 30 in d i v i d u a l s ) , a now outdated moral stage scoring system, and an assessment of attitudes regarding c a p i t a l punishment by only a 80 single question. This study included both men and women in the sample, used the most recent scoring manual (incorporating the substantial revisions to stage d e f i n i t i o n s ) , and assessed attitudes toward c a p i t a l punishment by both a dichotomous ( i . e . , single question) and continuous ( i . e . , CPQ) measure. The cor r e l a t i o n between the CPQ and moral reasoning i s consistent with that reported by DeWolfe and Jackson (1984) and substantiates the previous finding indicating increasing opposition to c a p i t a l punishment with higher levels of moral reason i ng. Thus, there is a rel a t i o n s h i p between the form and the content of moral reasoning, at least in the realm of c a p i t a l punishment. Such a finding i s congruent with those of other studies which indicate that higher stages of moral reasoning are t y p i c a l l y associated with l i b e r a l attitudes (Candee, 1976; Emler, 1983; Fishkin, Keniston, & MacKinnon, 1973; Haan, Smith, & Block, 1968). This rel a t i o n s h i p may be interpreted in several ways. For example, perhaps i t could be said that within student populations (the subjects in a l l of the c i t e d studies associating moral reasoning with l i b e r a l attitudes) there i s an overrepresentation of the l i b e r a l ideology. This has a certain i n t u i t i v e appeal u n t i l one p i t s t h i s finding against the c u r v i l i n e a r relationship reported e a r l i e r r e l a t i n g integrative complexity to c a p i t a l punishment. Clearly, t h i s l a t t e r relationship i s void of any l i b e r a l bias. Perhaps the r e l a t i o n s h i p between attitudes about c a p i t a l punishment and moral reasoning i s representative of the nature of the issue of c a p i t a l punishment i t s e l f . That i s , the two 81 sides of the c a p i t a l punishment issue may be imbalanced; not a l l alternative points of view are equally defensible, or even equally well known. DeWolfe and Jackson (1984) suggest that many of the familiar arguments thrown into the controversy in support of c a p i t a l punishment are at lower stages ( i . e . , "an eye for an eye") in contrast to the upper l e v e l , more s o c i e t a l and r e s p o n s i b i l i t y oriented arguments of opposition. Perhaps, too, this is representative of a l i b e r a l ideological bias in Kohlberg's theory with i t s focus on individual rights (Sullivan, 1977). This i s a more d i f f i c u l t position to defend, but one that cannot be discounted given the present findings. A related issue, within the realm of moral reasoning, was also addressed by this study: Do subjects use a d i f f e r e n t l e v e l of moral reasoning to substantiate their own, versus an opposing, position on c a p i t a l punishment? Within the present scoring system, the way in which weights are applied to the chosen and nonchosen positions implies that, in some way, reasoning on the chosen side i s more advanced, better formulated, and more representative of subjects' reasoning than is reasoning on the nonchosen side. (The weights are 3 and 2, respectively). This strays somewhat from the cognitive-developmental notion that a stage represents a structured whole with the implication that individuals w i l l be consistent and use the same l e v e l of moral reasoning, regardless of the content and context. The analysis reported e a r l i e r examined the l e v e l of moral reasoning used by subjects to both support and oppose c a p i t a l punishment as a function of their attitude on t h i s issue, and 82 yielded interesting r e s u l t s . F i r s t , related to the previous discussion, subjects who oppose c a p i t a l punishment use higher levels of moral reasoning o v e r a l l than do those who support i t . Equally interesting, however, is that regardless of subjects' attitude toward c a p i t a l punishment, they use higher moral reasoning in opposition to i t , than in support of i t ; a difference of about one-third stage. Thus, in contradiction to implications drawn from Kohlberg's system, subjects do not always use higher moral reasoning to substantiate their own position; they use higher moral reasoning when opposing c a p i t a l punishment, regardless of their chosen position. This raises questions regarding procedures for weighting scores in determining o v e r a l l l e v e l of moral reasoning. This comment, though, must be somewhat q u a l i f i e d to the extent that the measure of moral reasoning employed in t h i s analysis d i f f e r s from that t y p i c a l l y used by Kohlberg in ways already described. The important s i m i l a r i t y i s that both sides of the issue are addressed and measured. It would be a worthwhile venture to investigate the relationship between reasoning on both the chosen and the nonchosen positions on standard interview protocols and other issues (e.g., abortion) to see i f these differences are maintained and/or validate the weighting procedure. This also raises questions regarding the cognitive-developmental assumption of structured wholeness of stages inherent in Kohlberg's stage model. Although the difference of about one-third stage i s not in i t s e l f s u f f i c i e n t l y damaging, differences of up to two stages were noted and over 23% of the 83 subjects evidenced differences in reasoning of at least one stage. The same arguments presented before are v a l i d here. That i s , perhaps the moral domain of c a p i t a l punishment is imbalanced where opposing arguments are more accessible, more compelling and at higher stages. An extension of th i s l i n e of thought is that the higher staged opponents of c a p i t a l punishment are more v i s i b l e and verbal, supplying reasons that are reproducible but not necessarily understood by lower stage subjects whose reasoning might be best represented in their arguments of support. Ad d i t i o n a l l y , t h i s discrepancy might point to a l i b e r a l value bias in Kohlberg's theory and model. Conclusion This thesis has addressed several inte r r e l a t e d issues in psychology. Within the realm of conceptual complexity, i t has been demonstrated that the PCT and measures of integrative complexity are comparable--but not equivalent—assessments of the cognitive structures an individual engages in the processing of information. They both adequately expose the complementary processes of d i f f e r e n t i a t i o n and integration (as proposed by Schroder and others), but d i f f e r in the extent to which these processes are exposed as a function of d i s p o s i t i o n a l and si t u a t i o n a l factors. It i s suggested that these are the operative factors accounting for the d i f f e r e n t i a l results of the PCT and the issue composition on the c a p i t a l punishment questionnaire. The former yielded no interpretable relationship; the l a t t e r yielded a s i g n i f i c a n t , c u r v i l i n e a r pattern (consistent with previous research and theoretical positions) with extreme attitudes r e l a t i n g to conceptual 84 sim p l i c i t y and more moderate attitudes r e l a t i n g to increasing complexity. Just as Suedfeld and others have demonstrated that endogenous and extraneous stress can influence an individual's l e v e l of complexity, i t is proposed that issue salience and personal conviction can function in a similar capacity. The relationship between conceptual complexity and moral reasoning was also examined, y i e l d i n g , as expected, a moderately high c o r r e l a t i o n . It i s suggested that this result i s perhaps best interpreted by considering conceptual complexity as a superordinate c l a s s i f i c a t i o n system of information processing within which considerations in the moral domain are embedded. One par t i c u l a r aspect of this moral domain examined in this study was c a p i t a l punishment. Consistent with previous research, i t was found that opposition increases with stage of moral reasoning. Moreover, subjects, when j u s t i f y i n g opposition to c a p i t a l punishment, evidenced higher moral reasoning than when they supported i t , regardless of chosen position on the issue. This l a t t e r finding c a l l s into question Kohlberg's procedure for weighting scores in determining o v e r a l l level of moral reasoning, as well as the cognitive-developmental assumption of structured whole. Furthermore, the findings taken together provide the basis for some interesting questions regarding the nature of morality in general and the nature of morality as defined by Kohlberg. This thesis is by no means the d e f i n i t i v e work about the issues i n , and the relationships between, conceptual/integrative complexity, moral reasoning, and attitudes toward c a p i t a l punishment. It does, however, address some basic dimensions of 85 these separate, yet related theories, and generate some provocative questions. 86 References Adorno, R. , Frenkel-Brunswick, E., Levinson, D. , & Sanford, R. (1950). The authoritarian personality. New York: Harper. A l l p o r t , G. W. (1958). The nature of prejudice. Garden City, NJ: Doubleday Anchor. B i e r i , J. (1968). Cognitive complexity and judgment of inconsistent information. In R. P. Abelson, E. Aronson, W. J. McGuire, T. M. Newcomb, M. J. Rosenberg, & P. H. Tannenbaum (Eds.), Theories of cognitive consistency (pp. 633-640). Chicago: Rand McNally. B i e r i , J. (1969). Category width as a measure of discrimination. Journal of Personality, 37, 513-521. B i e r i , J., Atkins, A. L., Briar, S., Leaman, R. L., M i l l e r , H., & Tripodi, T. (1966). C l i n i c a l and s o c i a l judgment: The discrimination of behavioral information. New York: Wiley. B l a s i , A. (1980). Bridging moral cognition and moral action: A c r i t i c a l review of the l i t e r a t u r e . Psychological B u l l e t i n , 88, 1-45. Candee, D. (1976). Structure and choice in moral reasoning. Journal of Personality and Social Psychology, 34, 1293-1301. C a r r o l l , J., & Rest, J. (1982). Moral development. In B. Wolman & G. Strieker (Eds.), Handbook of developmental  psychology (pp. 434-451). Englewood C l i f f s , NJ: Prentice-H a l l . Colby, A., Kohlberg, L., Gibbs, J . , Candee, D., Speicher-Dubin, B., Kauffman, K., Hewer, A., & Power, C. (in press). The  measurement of moral judgment: A manual and i t s re s u l t s. New York: Cambridge University Press. 87 Colby, A., Kohlberg, L., Gibbs, J., & Lieberman, M. (1983). A longitudinal study of moral judgment. Monographs of the  Society for Research in Child Development, 48(1-2, S e r i a l No. 200). DeWolfe, T. E., & Jackson, L. A. (1984). Birds of a brighter feather: Level of moral reasoning and s i m i l a r i t y of attitude as determinants of interpersonal a t t r a c t i o n . Psychological  Reports, 54, 303-308. Emler, N. (1983). Morality and p o l i t i c s : The i d e o l o g i c a l dimension in the theory of moral development. In H. Weinreich-Haste & D. Locke (Eds.), Morality in the making:  Thought, action, and the s o c i a l context (pp. 47-71). Chichester, England: Wiley. Falconer, R. D. (1973). Moral judgments: A methodological  study, u t i l i z i n g Kohlberg's moral dilemmas and B i e r i ' s  repertory g r i d . Unpublished master's thesis, University of New Brunswick, Fredericton, NB. Fishkin, J., Keniston, K., & MacKinnon, C. (1973). Moral reasoning and p o l i t i c a l ideology. Journal of Personality and  Social Psychology, 27, 109-119. F l a v e l l , J. H. (1968). The development of role-taking and  communication s k i l l s in children. New York: Wiley. Gibbs, J. C , & Widaman, K. F. (1982). Social i n t e l l i g e n c e :  Measuring the development of sociomoral r e f l e c t i o n . Englewood C l i f f s , NJ: Prentice-Hall. Guttman, L. (1950). The basis for scalogram analysis. In S. A. Stouffer, L. Guttman, E. A. Suchman, P. F. Lazarsfeld, S. A. Star, & J. A. Gardner (Eds.), Measurement and 88 prediction (pp. 60-90). Princeton, NJ: Princeton University Press. Haan, N., Smith, M. B., & Block, J. (1968). Moral reasoning of young adults: P o l i t i c a l - s o c i a l behavior, family background, and personality correlates. Journal of Personality and  Social Psychology, 10, 183-201. Harvey, 0. J . , Hunt, D. E., & Schroder, H. N. (1961). Conceptual systems and personality development. New York: Wiley. Kelly, G. A. (1955). The psychology of personal constructs: A  theory of personality (Vol. 1). New York: Norton. Kogan, N. (1979). B e l i e f s , attitudes and steroetypes about old people. Research on Aging, J_, 11-36. Kogan, N. (1979). A study of age categorization. Journal of  Gerontology, 34, 358-367. Kogan, N. (1982). Research on b e l i e f s and attitudes about old people. In G. Gutman (Ed.), Canada's changing age structure (pp. 297-325). Burnaby, BC: SFU Publications. Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach to s o c i a l i z a t i o n . In D. A. Goslin (Ed.), Handbook of s o c i a l i z a t i o n theory and research (pp. 347-480). Chicago: Rand McNally. Kohlberg, L. (1973). Continuities in childhood and adult moral development r e v i s i t e d . In P. B. Baltes & K. W. Schaie (Eds.), Life-span developmental psychology: Personality and  s o c i a l i z a t i o n (pp. '179-204). New York: Academic Press. Kohlberg, L. (1976). Moral stages and moralization: The cognitive-developmental approach. In T. Lickona (Ed.), Moral 89 development and behavior: Theory, research, and s o c i a l issues (pp. 31-53). New York: Holt, Rinehart and Winston. Kohlberg, L. (1978). Moral education reappraised. The Humanist, 38(6), 13-15. Kohlberg, L. (1981). Essays on moral development: The philosophy of moral development (Vol. 1). San Francisco: Harper & Row. Kohlberg, L., & Elfenbein, D. (1981). Capital punishment, moral development, and the constitution. In L. Kohlberg, Essays on moral development: The philosophy of moral  development (Vol. 1, pp. 243-293). San Francisco: Harper & Row. L i k e r t , R. (1932). A technique for the measurement of attitudes. Archives of Psychology, 140, 1-55. L i n v i l l e , P. (1982). The complexity-extremity effect and age-based stereotyping. Journal of Personality and Social  Psychology, 42, 193-211. L i n v i l l e , P. (in press). Affective consequences of complexity regarding the s e l f and others. In M. S. Clark & S. T. Fiske (Eds.), Affect and cognition: 17th Annual Carnegie Symposium  on Cognition. H i l l s d a l e , NJ: Erlbaum. L i n v i l l e , P., & Jones, E. (1980). Polarized appraisals of out-group members. Journal of Personality and Social Psychology, 38, 689-703. Loevinger, J. (1966). The meaning and measurement of ego development. American Psychologist, 21, 195-206. Moran, J. J., & Joniak, A. J. (1979). Effect of language on preference for responses to a moral dilemma. Developmental 90 Psychology, 15, 337-338. Nisan, M., & Kohlberg, L. (1982). Universality and v a r i a t i o n in moral judgment: A longitudinal and cross-sectional study in Turkey. Child Development, 53, 865-876. Osgood, C. E., Suci, G. J., & Tannenbaum, P. H. (1957). The measurement of meaning. Urbana, IL: University of I l l i n o i s Press. Palys, T. (1981). Capital punishment questionnaire. Unpublished manuscript, Simon Fraser University, Burnaby, BC. Piaget, J . (i960). The general problems of the psychobiological development of the c h i l d . In J . M. Tanner & B. Inhelder (Eds.), Discussion on c h i l d development (Vol. 4, pp. 3-27). London: Tavistock. P h i l l i p s , D. (1980). The deterrent effect of c a p i t a l punishment: New evidence on an old controversy. Amer ican Journal of Sociology, 80, 139-148. Porter, C. A., & Suedfeld, P. (1981). Integrative complexity in the correspondence of l i t e r a r y figures: E f f e c t s of personal and s o c i e t a l stress. Journal of Personality and Social Psychology, 40, 321-330. Princeton manual guidelines. (undated). Princeton, NJ: Princeton University. Rawls, J. (1971). A theory of j u s t i c e . Cambridge, MA: Harvard University Press. Rest, J. R. (1973). The h i e r a r c h i c a l nature of moral judgment: A study of patterns of comprehension and preference of moral stages. Journal of Personality, 41, 86-109. Rest, J. R. (1979). Development in judging moral issues. 91 Minneapolis: University of Minnesota Press. Rokeach, M. (1973). The nature of human values. New York: Free Press. Rosenberg, M. J., & Hovland, C. I. (1960). Cognitive, a f f e c t i v e , and behavioral components of attitudes. In M. J. Rosenberg, C. I. Hovland, W. J. McGuire, R. P. Abelson, & J. W. Brehm (Eds.), Attitude organization and change: An  analysis of consistency among attitude components. (pp. 1 -14). New Haven, CT: Yale University Press. S e l l i n , T. (1959). The death penalty. Philadelphia: American Law I n s t i t u t e . Schroder, H. M., Driver, M. J., & Streufert, S. (1967). Human  information processing. New York: Holt, Rinehart and Winston. Scott, W. A. (1962). Cognitive complexity and cognitive f l e x i b i l i t y . Soc iometry, 25, 405-414. Scott, W. A. (1969). Structure of natural cognitions. Journal  of Personality and Social Psychology, 12, 261-278. Selman, R. L. (1976). Social-cognitive understanding: A guide to educational and c l i n i c a l p r a c t i c e . In T. Lickona (Ed.), Moral development and behavior: Theory, research, and s o c i a l  issues (pp. 299-316). New York: Holt, Rinehart and Winston. Streufert, S., & Streufert, S. C. (1978). Behavior in the  complex environment. Washington, DC: Winston. Suedfeld, P. (1978). Charac t e r i s t i c s of decision making as a function of the environment. In B. King, S. Streufert, & F. F i e l d l e r (Eds.), Managerial control and organizational  democracy (pp. 203-213). Toronto: Wiley. 92 Suedfeld, P. (in press). Intimations of mortality: Integrative s i m p l i f i c a t i o n as a precursor of death. Journal of  Personality and Social Psychology. Suedfeld, P., & Rank, A. D. (1976). Revolutionary leaders: Long-term success as a function of changes in conceptual complexity. Journal of Personality and Social Psychology, 34, 169-178. Suedfeld, P., & Tetlock, P. (1977). Integrative complexity of communications in international c r i s e s . Journal of C o n f l i c t  Resolution, 21, 169-184. Suedfeld, P., Tetlock, P., & Ramirez, C. (1977). War, peace and integrative complexity: United Nations speeches on the Middle East problem. Journal of C o n f l i c t Resolution, 21, 427-442. Sulli v a n , E. V. (1977). A study of Kohlberg's structural theory of moral development: A c r i t i q u e of l i b e r a l s o c i a l science ideology. Human Development, 20, 352-376. Sulli v a n , E. V., McCullough, G., & Stager, M. (1970). A developmental study of the relationship between conceptual, ego, and moral development. Child Development, 41, 399-411. Tetlock, P. E. (1981). Pre- to postelection s h i f t s in pr e s i d e n t i a l rhetoric: Impression management or cognitive adjustment? Journal of Personality and Social Psychology, 41 , 207-212. Tetlock, P. E. (1983a). Accountability and complexity of thought. Journal of Personality and Social Psychology, 45, 74-83. Tetlock, P. E. (1983b). Cognitive style and p o l i t i c a l 93 ideology. Journal of Personality and Social Psychology, 45, 118-126. Tetlock, P. E. (1984). Cognitive style and p o l i t i c a l b e l i e f systems in the B r i t i s h House of Commons. Journal of  Personality and Social Psychology, 46, 365-375. Thurstone, L. L., & Chane, E. J. (1929). The measurement of  attitu d e . Chicago: University of I l l i n o i s Press. UBC guidelines for the scoring of conceptual and integrative complexity, "(undated).. Vancouver, BC: University of B r i t i s h Columbia. Vidmar, N., & Ellsworth, P. (1974). Public opinion and the death penalty. Stanford Law Review, 26, 1245-1270. Walker, L. J. (1980). Cognitive and perspective-taking prerequisites for moral development. Child Development, 51, 131-139. Walker, L. J. (1982). The sequentiality of Kohlberg's stages of moral development. Child Development, 53, 1330-1336. Walker, L. J . (in press). The v a l i d i t y of the moral stage model and the r e l i a b i l i t y of the moral judgment interview. In L. Kohlberg & D. Candee (Eds.), Research in moral development. Cambridge, MA: Harvard University Press. Walker, L. J . , de Vries, B., & Bichard, S. (in press). The h e i r a r c h i c a l nature of stages of moral development. Developmental Psychology. Zajonc, R. B. (1960). The process of cognitive tuning in communication. Journal of Abnormal and Social Psychology, 61_, 159-167. Zimbardo, P., Ebbesen, E., & Maslach, C. (1977). Influencing 94 attitudes and changing behavior (2nd ed.). Don M i l l s , ON: Addi son-Wesley. 95 Footnote 1More s p e c i f i c a l l y , since a moderate relationship i s what was hypothesized, i t i s appropriate to determine that t h i s i s , in fact, what was achieved. "Moderate" is defined here as accounting•for less than 50% of the variance, but s t i l l s i g n i f i c a n t . Testing for this entailed the comparison of the obtained c o r r e l a t i o n with one accounting for 50% or more of the variance ( i . e . , a "strong" r e l a t i o n s h i p ) . A c o e f f i c i e n t of .71 was used in this comparison. Employing the Fisher Z-transformation, the resultant normally-distributed Z value was -4.14, 2 < .001. Thus, the co r r e l a t i o n between conceptual and integrative complexity can be said to be moderate as i t i s s i g n i f i c a n t l y greater than a n u l l relationship, but s i g n i f i c a n t l y less than one considered to be strong. Appendix A Test Materials 97 This study assesses the way in which people describe their perceptions of, or feelings toward, s p e c i f i c s o c i a l issues. You w i l l be asked to perform various tasks throughout th i s booklet: write a couple sentences in response to a word or phrase, write several paragraphs, complete a questionnaire. The aim of thi s study is to examine the relationship among these tasks. Please respond to a l l of the sections as completely as you can; however, you are free to refuse to respond to any of the sections or to answer any of the questions. If you complete thi s booklet, i t w i l l be assumed that your consent to participate in t h i s study has been given. Refusal to partic i p a t e or withdrawal at any time w i l l not influence your class standing in any way. This booklet should take about one hour to complete. Please complete the sections in the order in which they appear, without looking forward through the booklet. A l l information that you provide w i l l be kept completely c o n f i d e n t i a l . Thank you for your p a r t i c i p a t i o n . Please complete the following: - Sex: M F - Age (in years): - Number of years of post-secondary study: 98 Section I On each of the following pages, please complete the sentence which is begun for you and then write at lease one more sentence on the topic. After that, you may--if you wish—write two or three more sentences i f time permits. PLEASE, DO NOT TURN THE PAGE UNTIL ASKED TO DO SO. 99 When I am c r i t i c i z e d ... PLEASE DO NOT TURN THE PAGE UNTIL ASKED TO DO SO. 100 Rules . . . PLEASE DO NOT TURN THE PAGE UNTIL ASKED TO DO SO. 101 When I don't know what to do ... PLEASE DO NOT TURN THE PAGE UNTIL ASKED TO DO SO. 102 Confusion .. . PLEASE DO NOT TURN THE PAGE UNTIL ASKED TO DO SO. 103 Policemen .. . PLEASE DO NOT TURN THE PAGE UNTIL ASKED TO DO SO. 1 04 When a friend acts d i f f e r e n t l y towards me ... PLEASE DO NOT TURN THE PAGE UNTIL ASKED TO DO SO. 105 Section II The execution of criminals for committing serious crimes or offences, has been, and remains, a controversial topic and i t s appropriateness has been intensely debated for many years in many countries. The aim of thi s study is to examine and document the variety of opinions that exist in re l a t i o n to this topic of c a p i t a l punishment. Towards such an end, you are requested to write a brief composition ( i . e . five to ten paragraphs) discussing the issue of c a p i t a l punishment as you perceive i t . Please present your opinion and attitude towards c a p i t a l punishment and include discussion of any factors that you consider relevant. Because th i s topic i s a matter of opinion, there are no right or wrong answers. Please begin writing now. Please use the remainder of thi s page and as much of the following two pages [two blank pages followed] as you require to write t h i s composition. You should take about 20 minutes to complete th i s task. 106 [The order of Sections III and IV was counterbalanced.] Section III The previous section was concerned with a general discussion of your opinion and attitude towards c a p i t a l punishment. This section i s more s p e c i f i c a l l y concerned with only one side of this issue; that i s , you are requested to provide as many reasons as you can think of to support the use of c a p i t a l punishment. Please explain your reasons f u l l y (especially why you think' these reasons are important). Please take about 10 minutes to complete th i s section using the remainder of t h i s page and the following page [one blank page followed]. 107 Section IV Now adopt the other side of t h i s issue; that i s , please provide as many reasons as you can think of to oppose the use of c a p i t a l punishment. Once again, please explain your reasons  f u l l y (especially why you think these reasons are important). Please take about 10 minutes to complete this section using the remainder of t h i s page and the following page [one blank page followed]. 108 Section V Listed below are a number of statements of opinion regarding c a p i t a l punishment. Please read each statement and then indicate the extent to which you agree or disagree with that statement by c i r c l i n g one of the numbers from -3 to +3 to the l e f t of the statement, according to the following scale: -3 = strongly disagree -2 = moderately disagree -1 = s l i g h t l y disagree 00 = neutral, or no opinion +1 = s l i g h t l y agree +2 = moderately agree +3 = strongly agree Because the statements are matters of opinion, there are no right or wrong answers. You w i l l probably find that you agree with some statements and disagree with others. -3 -2 -1 00 +1 +2 +3 I feel that c a p i t a l punishment has no use in society. -3 -2 -1 00 +1 +2 +3 Each serious offence must be dealt with in d i v i d u a l l y to decide whether c a p i t a l punishment should be used. -3 -2 -1 00 +1 +2 +3 People who agree with c a p i t a l punishment are s a d i s t i c . -3 -2 -1 00 +1 +2 +3 Capital punishment puts the control of who s h a l l l i v e and die in the hands of imperfect humans. -3 -2 -1 00 +1 +2 +3 Capital punishment protects society. -3-2-1 00+1 +2 +3 The "common good" requires that dangerous and violent persons be destroyed. -3 -2 -1 00 +1 +2 +3 Capital punishment i s the same as murdering a person. -3 -2 -1 00 +1 +2 +3 Capital punishment i s useful in some societies but not in others. -3 -2 -1 00 +1 +2 +3 Whether c a p i t a l punishment i s enforced should depend on the circumstances of the individual case. 109 -3 = strongly disagree -2 = moderately disagree -1 = s l i g h t l y disagree 00 = neutral, or no opinion +1 = s l i g h t l y agree +2 = moderately agree +3 = strongly agree Capital punishment is an unjust method of dealing with convicted felons. -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 -3 -2 -1 00 +1 +2 +3 Capital punishment c i v i l i z e d society. Capital punishment barbaric. Capital punishment has no place in a for any reason i s is a necessary e v i l . An eye for an eye; people who k i l l others should be k i l l e d . C apital punishment is necessary to clear out the scum of society. Capital punishment isn't pleasant, but neither are the people who would be receiving i t . Capital punishment is a f a i r and just way of dealing with society's worst offenders. If we allow c a p i t a l punishment, some innocent people w i l l inevitably be k i l l e d . C apital punishment eliminates the p o s s i b i l i t y of r e h a b i l i t a t i o n and treatment, which should be the aims of sentenc ing. The k i l l i n g of another person i s b a s i c a l l y wrong and should never be done, even by the state. Murderers and rapists deserve to be executed. Capital punishment should not be used under any circumstances. Murderers should be murdered. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096288/manifest

Comment

Related Items