D \. 1‘ - ‘v‘: ‘ o 9 ' - . - a A r- ‘ 'o} -L ." u... 5:95am}; ‘. .15; ?R‘>sjt?£i-‘i'ifil,-Z'$s mat? ?» i 3.51%} £1 WW??? 0 3334‘ - 3353 mm ' o u I n a . ¢. . .. o. . o ‘ .0 c 0 C _. . c -- '0 - . . .1 .. . -Dhq"'v.- -‘f .I o. -I-’ I. . I It” -dOi.o . ... I. t v o of . . 0" .4 I o. -- . .a I - v I . ‘- - - _‘ v." - A *ILA“ A A n A w . -I I o'r-w--__ __-.__.—.——._ - ‘ \v‘ ‘_ V v .. v ." v ‘ ‘ > ' ‘ ‘ - . O - . . ' . s. - ' . g ‘ . _ . - Q - ' - - - ° . _ g._ V . . - b ___ . O V o , v . _ . . . - i . . - *- .- ‘- ' ‘ o. _ - . - -_ _ . . . - . . 9- ' - _ a - v - > . .- - . . . > - _ - - . . ~ . - . _ ' . ‘o‘ - . u . . - ~ A - - . . . . . . - - o ' I . a » ‘ t. ' . o - . . - u . ' I ' l - V - . - ' .. . ‘ . _ ' ~ 4 0 > . o ,. - . . . I . a _ ‘ ' _ . ' ' _ . ‘ . . a _ . ‘ h ‘ - o . c , . . - - -. . .\ - . ‘ t g o I ‘. ‘O' 1".-ool’r -‘v "o g>oo I ‘ . “, -s..- , .a.-)r"qu-‘ ‘... 'rfl'lo-Htlx" ‘~r ll . I :‘l. 9.1.4" on. W 9 "".J. O o . 0'" u~‘—_-or.-. ' a... a... . ”I'ol.co’--f_“ .‘, _-.a.-~oo 0—0”. Lo‘pa’P9'” "_':,' 'w,.-.: ' -‘ 3'0. .I‘I'rrl'oo filo} (‘iw'r""' 3; "at. o ’ -’ Q—l-fl 5".-l(""-‘Y.O " 0PM“"' ' rw'fl ...,."‘r “a o’u‘. -~- ‘ ’fi'" . Mm r'..-o-qv:~or'""r"""" H m -<~op~v"' ”"'" "' ir'flu'o" :-~m"v‘~ -—"" """ "'u'u’:¢ovlh trwnwfl“”"" - ..Or :0 vmtda'fl~o 'rmmo~o..lfh“_’:?" org..v' 4- ~-’: .‘flr. .1... ’r‘ L‘ ". -. ' ‘ ' "o row ~o"~"' "’ ‘ ' " - ' on” --Orqu¢'~~ "'""'""’ " 'o. ow. "' “ -. ”a. mormvvrrcon'WOmw..- 400:0.10- - e 0 .N'V’m""°‘m""cx M. ‘ "' 0'01. Gait-9.0.. “-1" ”pnorf.'. " ' ‘ o""fa ‘wwflvfin"; o . 00".“fo -‘ (no- ” .A.A A -wnfi? 1‘; L I B R A R “I lflllIHHllllll W11"! Illl Ill HI Hllll ”ll ”1 lllllill m, g n $21 3 1293 10196 1112 University DIFFERENTIAL CONCEPTIONS OF PHYSICIAN RESPONSIBILITY AMONG MEDICAL SUBGROUPS by Gregory T. Loftus Abstract of Master's Thesis Completed Summer Term, 1970 The term responsibility has been used to convey either a characteristic of an individual, or an inter- personal commitment. Limited psychological research has prOposed a number of personality, socioeconomic, or more concretely behavioral correlates of high or low ratings of responsibility. Research has also investigated the variables which interact in the process of making judgments about, or attributing responsibility. The present study examined the behavioral elements and conceptual components of responsibility as that concept is exemplified in the role of physician. An extensive literature review was followed by a statement of the problem. Despite frequent verbal reference to the obvious importance of responsibility and its high social desira— bility, the empirical referents of the concept have remained vague. Efforts to evaluate or predict an individual's level of responsibility have met with very limited success. A step toward improving present capacities lay in a closer examination of different com- ponents of responsibility, and their relative weight or importance. Six components were derived from the results of previous research. Any behavioral example or descrip- tion of responsibility from the previous research could be classed as one of these six components. This study obtained examples of responsible behavior and definitions of responsibility from five groups of respondents: Practicing physicians, medical faculty, medical students, paramedical personnel, and patients (n 300). Responses were analyzed to determine the relative importance or frequency of use of the different components of responsibility. Dependability was found to be clearly predominant, characterizing more than 50% of the responses in the survey. Initiative, altruism, and sociability were used frequently in typifying responsibility for the physician. Independence and conformity were used very infre- quently. This latter represented a departure from earlier research results. The different findings were attributed to the operation of socialization procedures in the different settings. Significant differences were observed between the distribution of items to categories depending upon whether the item was given as an example or definition of responsibility. It was hypothesized that this result may have derived from the difference in the response task. Despite overall similarities, significant differences were recorded among the different groups of respondents in the frequency with which they used the various components. These differences did not fully correspond to the differ- ences predicted from analysis of role interaction. An inventory of behaviors considered indicative of responsibility was compiled from the responses. These behaviors were examined to reveal the special concerns of the particular groups. flCommittee ChairmaW mng/fl/70 Thesis Committee: James Phillips Henry Smith Arthur Elstein DIFFERENTIAL CONCEPTIONS OF PHYSICIAN RESPONSIBILITY AMONG MEDICAL SUBGROUPS by Gregory T. Loftus A THESIS Submitted to the Department of Psychology Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS 1970 «\.J/$.> I1 913s) W5. A ” 'lllllllul‘u . n l in...» :11 1‘ p‘l) a Klf‘. ”nkrwéx 18 £41. :4 m... VIII! DEDICATION MR. AND MRS. JOSEPH P. LOFTUS Who demonstrated to me how a responsible life is lived. ACKNOWLEDGMENTS The author wishes to express his gratitude to all those persons whose sense of responsibility and professional com- petence aided in the completion of this project. Among them: The members of the Office of Medical Education Research and DevelOpment for their continual advice and support. Especially - Hilliard Jason, Director, who guided me through many periods of great difficulty. — Arthur Elstein who responded so well when asked for his help. - Susan Green who demonstrated each of the components of responsibility, repeatedly and in full measure while preparing this manuscript over the period of several years. James Phillips, Thesis chairman, who utilized his pro- fessional competence in the continual challenge which pro- duced this report. Henry Clay Smith whose genteel communications inspired confidence and courage. Bruce Forche who printed the final manuscript. TABLE OF CONTENTS Page INTRODUCTION AND REVIEW OF LITERATURE. . . . . . . . . l Behavioral Descriptions and Correlates of Responsibility. . . . . . . . . . . . . . . . . . 2 Attribution of Responsibility and Causality . . . . l6 PROBLEM. . . . . . . . . . . . . . . . . . . . . . . . 28 Hypothesis. . . . . . . . . . . . . . . . . . . . . 37 METHOD . . . . . . . . . . . . . . . . . . . . . . . . 38 Subjects. . . . . . . . . . . . . . . . . . . . . . 39 Instrument. . . . . . . . . . . . . . . . . . . . . D2 Procedure - Judging . . . . . . . . . . . . . . . . AB RESULTS. . . . . . . . . . . . . . . . . . . . . . . . h9 Response Characteristics. . . . . . . . . . . . . . 50 Analysis of the Distributions . . . . . . . . . . . 52 Results of Analysis . . . . . . . . . . . . . . . . 7h DISCUSSION . . . . . . . . . . . . . . . . . . . . . . 82 Response Characteristics. . . . . . . . . . . . . . 83 Distribution of Responses . . . . . . . . . . . . . 8h Group Differences . . . . . . . . . . . . . . . . . 91 Future Research Involving Responsibility. . . . . . 97 REFERENCES . . . . . . . . . . . . . . . . . . . . . . 99 APPENDICES . . . . . . . . . . . . . . . . . . . . . .th LIST OF TABLES Table Page I Components of Responsibility. . . . . . . . . . 32 II Index of Agreement Among Judges . . . . . . . . MB III Survey Response Rates . . . . . . . . . . . . . h8 IV Distribution of Responses Given as Examples of Responsibility . . . . . . . . . . . . . . . 53 V Distribution of Responses Given as Definitions of Responsibility . . . . . . . . . . . . . . . 55 VI Chi Square Analysis of Group Distributions Comparing Observed Frequencies to Hypothesis of Equal Frequencies. . . . . . . . . . . . . . 58 VII A Chi Square Comparison of Distributions of Examples vs. Definitions. . . . . . . . . . . . 58 VIII Distribution of Responses into Categories by Percent. . . . . . . . . . . . . . . . . . . 61 IX Two by Six Chi Square Analysis Pairwise Comparison of Respondent Groups: Examples of Responsibility . . . . . . . . . . . . . . . 63 X Two by Six Chi Square Analysis Pairwise Comparison of Respondent Groups: Definitions of Responsibility . . . . . . . . . . . . . . . 63 XI Response Distribution Sociability Versus Other Categories: Examples of Responsibility . 68 XII Chi Square Analysis Pairwise Comparison of Groups Sociability: Examples of Responsibility. . . . . . . . . . . . . . . . . 68 XIII Response Distribution Sociability Versus Other Categories: Definitions of Responsibility. . . . . . . . . . . . . . . . . 69 XIV Chi Square Analysis Pairwise Comparison of Groups Sociability: Definitions of Responsibility. . . . . . . . . . . . . . . . . 69 XV Response Distribution Altruism Versus Other Categories: Definitions of Responsibility. . . 72 Table Page XVI Response Distribution Altruism Versus Other Categories: Examples of Responsibility. . . . . 72 XVII Chi Square Analysis Pairwise Comparison of Groups Altruism Component: Examples of Responsibility . . . . . . . . . . . . . . . . . 73 XVIII Response Distribution: Examples . . . . . . . . 9h XIX Response Distribution: Definitions. . . . . . . 9h LIST OF ILLUSTRATIONS Figure I Graph of Response Percentages: Examples of Responsibility. . . . . . . . . . . . II Graph of Response Percentages: Definitions of Responsibility. . . . . . III Profile of Group Response Distribution: Examples . . . . . . . . . . . . IV Profile of Group Response Distribution: Definitions. . . . . . . . . . Page 514 56 65 66 LIST OF APPENDICES Appendix A. Items from the Check List Relevant to Responsibility: Havighurst and Taba, l9h9. Personal Scale of Social Responsibility: Gough, McCloskey and Meehl, 1952. Scale for Measuring Attitudes of Social Responsibility in Children: Harris, D. B., 1957. Instrument. . . . . . . . . . . . . . . Page 101: 110 11h 118 INTRODUCTION A significant dimension in interpersonal evaluation is the degree to which an individual is perceived to be responsible. Anderson (1968) found that the character— istic "responsible" was ranked 28th. in a list of 555 personality trait words ordered by their degree of favorableness. Over the last forty years, a limited amount of investigation has been directed toward the way in which the term responsibility was used, and the behavioral components it implied. Some of this research was concerned with responsibility as a function of the association between prescribed and observed behaviors. It sought to establish what behaviors were prescribed for certain settings, and to determine how those pre- scriptions related to judgments of responsibility. Other research focused upon the aspect of causality. These studies investigated the variables which effect, for example, the amount of responsibility attributed to an individual. The function of perceived freedom versus perceived interference, the degree of association between agent and outcome, and, the nature of the outcome have each been examined in relation to judgments of responsibility. 2 The First Type of Research to be Examined was Behavioral Descriptions and Correlates of Responsibility Since the beginnings of controlled investigation around 1930, there have been several descriptions of the "responsible" person pr0posed by different researchers. A composite of these characteristics will be presented, followed by a description of the differing methodologies and subject populations used in the research. Four studies have contributed the majority of this descrip- tive material, and their contributions will be designated according to the following key: Havighurst and Taba (l9h9)= I; Gough, McClosky, and Meehl (1952)=II; Harris et al. (195b, 1957, 195ha, l95hb, 1955, l958)=III; and, Brown and Landsberger (196O)=1Vo According to these researchers: The responsible person has "a strong nonauthoritarian personality" (III), with a "strong and unflagging sense of confidence in himself" (II). He considers his first duty is toward his own success (1). He requires little supervision (III), demonstrating personal independence (III) from both peers and adults (I), in carrying out his task (I) and in fulfilling his promises (I). He demonstrates initia- tive as "foresight" and "readiness to accept new tasks" (IV) or even seeking out opportunities for service (I). He is willing to face difficulties in carrying out what is expected of him (I) and exhibits initiative in solving such difficulties (I). He is resourceful (III). Others can 3 count on him (II) to carry through activities (III) and fulfill tasks expected of him (I). By always completing accepted jobs (I), he demonstrates dependability (II, III), reliability (IV), conscientiousness (IV), and trustworthi- ness (II, IV). He is punctual, and demonstrates this in his attendance (I). He demonstrates almost excessive emphasis on carrying his own share of burdens and duties (II). Further, he shows a ready willingness to accept the consequences of his behavior (II). He is orderly (III). He has a sense of rightness of the larger social world (II). He conforms to social norms (behavior) (III), and there is conformity between his attitudes and societal expectations (III). He has a sense of obligation to the group (II), and is socially oriented or group centered (III). He demonstrates sociability in terms of good relations with others (IV) and helpfulness to others (III, IV). He participates well (III). He is mature (III), and has constructive interests (III). He has a high, but somewhat rigid, set of self-demands (II), and highly developed standards (I, II) which he applies under a variety of circumstances (I). He is a straight shooter (II) with a strong sense of justice (II). He shows deep concern over the broader ethical and moral problems (II), and rejects the light, trivial, or dangerous (II). He also rejects privilege or favoritism (II), and disapproves of leniency toward athletes, or talented or forgetful people (I). Although work outside school is taken somewhat h more seriously than school work or activities, duties are taken seriously at school, home, and employment (I). The responsible person demonstrates the qualities listed above in all situations with both supervisors and peers (I). The methodology associated with the derivation of the above composite description resulted largely from the well-known Studies in the Nature of Character, conducted by Hartshorne and May (1930). In their investigation, Hartshorne and May and their associates devised a number of laboratory situations to examine behaviors indicative of character traits. The comparisons they made included: honesty versus cheating, c00peration versus selfishness, lying versus telling the truth, and service versus neglect. They developed multiple laboratory exercises for each of these. The correlations which they obtained between Idifferent tests for the same type of conduct were generally low. Ranging from 0.12 to O.h0, the average correlation for different laboratory behaviors within a trait was 0.20. Hartshorne and May reported that when observations of laboratory test behavior was used, almost no general conclusions could be drawn from the research. Hartshorne and May (1930) concluded that the low correlations which they obtained reflected the specificity of the testing situation and the consequent specificity of behaviors observed. Searching for an alternative index of character traits which suffered less from this difficulty, Hartshorne and May investigated the criterion of social 5 reputation. This measure reflected a series of impressions based on real life situations in a variety of different settings. Different measures of reputation were found to correlate between 0.36 and 0.77, with an average of 0.60. These facts led Hartshorne and May to suggest: When enough opinions can be gathered with reasonable care and from contrasted sources—- as from pupils, teachers, and parents--the resulting score becomes a fair substitute for an elaborate and expensive program of objective testing. (Hartshorne and May, 1930, p. 369.) This methodological recommendation was put into effect twenty years later in the next major investigation of character: Adolescent Character and Personality (Havighurst and Taba, 19h9). The purpose of this research was to in- vestigate the impact of social settings and individual characteristics on the development of the moral character of an individual. Defining character as a composite of moral traits, they selected honesty, responsibility, loyalty, moral courage, and friendliness as representa- tive of the traits which make up moral character. Following the suggestion of Hartshorne and May (1930), they set out to obtain reputation measures of these traits. Havighurst and Taba developed four specially designed instruments to obtain judgments of reputation. Two of these were used with the adults and two were used with the age-mate or peer group. Adults responded to the Check List, and a character sketch instrument. The check list was composed of 126 descriptions of behavioral situations in which varying degrees of the five traits 6 were manifested. The second instrument consisted of a series of short verbal sketches, each designed to repre— sent an individual of some degree of honesty, moral courage, loyalty, responsibility, or friendliness. There were between seven and ten sketches for each character trait. For age-mates, the Guess-Who and Portrait Guess-Who instruments were used. The former consisted of about ten word pictures for each trait in paired opposites--one high, one low. The Portait Guess-Who differed in that each word picture contained several manifestations of a trait as they might be found in some hypothetical indivi- dual. These instruments were used to evaluate all (112) of the sixteen year olds in the community studied. Havighurst and Taba reported the results of tabula— ting those evaluations in the form of a description of the "responsible" person. That description was included in the composite presented by this paper. The emphasis in their description lay primarily on the component of dependability. "Punctuality and completing accepted jobs characterize the concept of responsibility (19h9, p. 86)." They also reported other components; one reflecting initiative; one a complex involving indepen- dence, self-confidence and ascendency. An important part of Havighurst's and Taba's description was the implication that accepted tasks were carried through despite, or even especially when confronted by diffi- culties. (Appendix A contains the Check List items used by Havighurst and Taba to discriminate responsibility.) 7 A methodology similar to Havighurst and Taba's was used by Gough, McClosky, and Meehl (1952). They gave a definition of responsibility to principles, teachers, and pupils in several high schools. Ratings were made of students on the basis of that definition. Criterion groups of students high and low on those ratings were selected. A battery of items was then administered to the two groups, and the 56 items which discriminated adequately were chosen to form a final scale (see Appendix B). Correlations between criterion ratings and scale scores were moderate, 0.50 to 0.59. Correla- tions between self—ratings and scale scores ranged from 0.1h to 0.11. Between self—ratings and criterion ratings, correlations were less than 0.10. The a priori definition which Gough et a1. presented to their subjects, and the description with which they summarized their results were both included in the composite given earlier. Briefly their definition stated: "The responsible person is one who shows a ready willing— ness to accept the consequences of his own behavior, dependability, trustworthiness, and a sense of obligation to the group (1952, p. 79)." The description based on their results emphasized a concern with broader ethical problems, and internalization of social norms and expectancies. It should be noted that the personality scale for social responsibility derived by Gough et al. was intended 8 as an instrument for the prediction of group trends and differences only. As was indicated by the size of the correlations reported above, that scale had limited value in predicting individual levels of responsibility. A series of investigations was subsequently carried out to establish individually valid and reliable assess- ments of responsibility. An introductory report by Harris, Clark, Rose, and Valisek (195ha) presented the Citizen- ship Scale of Social Responsibility. This scale was a modified h8—item version of the Personality Scale of Social Responsibility (Gough et a1., 1952). Harris et al. also employed the Teacher Check List (Havighurst and Taba, 19h9). Harris and his associates reported finding that these two measures were not significantly correlated, and showed no developmental trend, i.e., scores did not increase with age. Searching for other criteria of responsibility, Harris (195h) asked 215 female education students to describe the most responsible and least responsible students they had encountered during practice teaching. They were asked to emphasize specific behaviors rather than respond in terms of traits or more general characteristics. The description of the responsible student which Harris abstracted from his results was included in the previously presented composite of this paper. The stress in that description was upon conformity and social orientation, extending Gough et al's. (1952) emphasis upon awareness of social norms. 9 Following up on these findings, Harris, Clark, Rose, and Valisek (1965) investigated the relationship between responsibility and personality differences. Criterion groups were selected from the data of a previous study (Harris et a1., l95ha). A Guttman scale analysis was used to determine personality components of two scales, the Citizenship Scale, and the Teacher Check List. They derived from these scales seven components or dimensions of responsibility. From the Citizenship Scale came: 1) group centeredness; 2) conformity to social norms; 3) personal independence; h) possession of constructive interests; and 5) strong and nonauthoritarian personality. The Teacher Check List produced: 6) dependability; and 7) conformity between a child's attitudes and societal expectations. Differences on these dimensions were found to discriminate between children evaluated as responsible and those rated less responsible. 0n the basis of a theory of responsibility as a "composite of attitudinal elements reflecting behavior classifiable as reliable, accountable, loyal, or doing an effective job," Harris (1957) published a Scale for Measuring Attitudes of Social Responsibility in Children. This scale showed a positive trend in mean score with age, and substantial correlation with other measures of personal and social adjustment. The scale was composed of 21 items from the Gough et a1. scale (1952), and 29 new items. The new items were developed a priori and 10 selected for their discrimination of criterion groups formed by teacher and peer nominations. No correlations between ratings and scale scores were reported. A retest after four months yielded a reliability of 0.60 to 0.70. (See Appendix C for the scale.) A final study by Harris (1958) reported little consistency among parents' ratings of their own childrens' responsibility. He suggested that parents' judgments about responsibility are confounded by their successes and failures in relationships of discipline, guidance, and affection. Brown and Landsberger (1960) published a two part report entitled, The Sense of Responsibility Among Young Workers. They included an overview of the research up to that time, and presented the results of their own inventory of behaviors characteristic of responsibility. They proposed three subcategories of responsibility: I Initiative (Foresight and readiness to accept new tasks.) II Reliability (Conscientiousness and trust- worthiness.) III Sociability (Helpfulness and good relations with others. N.B., not affability, but ability to get along.) The reliability and exhaustiveness of these categories was verified by asking five judges to assign behavior items to the appropriate category. They reported that this test was very successful, but did not include any coefficient of agreement among the judges. Brown and Landsberger 11 obtained ratings of 99 subjects on 2h of the behaviors and derived correlations between the subcategories. Between initiative and reliability, the average correlation was 0.80; between reliability and sociability, 0.71; and between initiative and sociability, 0.56. These correla- tions led them to conclude that they were dealing with one basic trait with different subcategories. To test whether the concept of responsibility was continuous across different situations, i.e., school and work, they sought an independently established scale. Rejecting the Harris scale as pertinent mainly to school situations, they selected instead the Gordon Personal Profile which contains a subscale for responsibility. An average rank correlation of O.h8 (with wide fluctuation) was obtained between scores on the Gordon and supervisor ratings of responsibility. Brown and Landsberger found that the evaluation of responsibility by supervisors weighed most heavily on the component of reliability. Seventy-four out of 10h supervisors considered this aspect most important. The paradigm for establishing behaviors indicative of responsibility, as derived from the similarities among these four research efforts is as follows: Present a definition of responsibility to a group of subjects; on the basis of that definition, have subjects select some associate or associates whom they consider to have demonstrated this characteristic in their social l2 interactions; obtain from the subjects behaviors which they consider indicative of this characteristic responsibility; compile and consolidate these descriptions into a single description of the "responsible person." Some of the re- search proceeds--using the selection by associates as criteria, establish groups high and low on responsibility, and obtain correlations with other measures. Such measures have included demographic or developmental profiles, and questionnaire responses to attitude surveys. The descriptive definitions of a responsible person, given to subjects as stimuli or obtained as responses, have been previously summarized in this paper. The methodology of the research-—techniques and scales for establishing criterion ratings of responsibility-—was reviewed. There were further findings associated with that research, however, and other variables found to be associated with responsibility have emerged. 1) Persons from rural areas tend to be evaluated as higher on responsibility than those from urban backgrounds (Harris et a1., l95ha; Brown and Landsberger, 1960). 2) Males were rated less responsible than females (Harris et a1., 195ha) in school settings, but this effect reverses in work situations where males were rated more responsible (Brown and Landsberger, 1960). 3) There is a positive correlation between socio-economic class and ratings of responsi- bility (Havighurst and Taba, 19h9; Brown and Landsberger, 1960; and Harris et. a1., 1955). This variable appears to be mediated by school achievement. In so far as socio-economic class leads to greater achievement in school, it also leads to higher ratings of responsibility. ’4) 5) 6) 7) 13 When the positive correlation of socio-economic class with school achievement dissipates, it also loses its positive correlation with higher ratings of responsibility. There is a positive correlation between achieve- ment in school and ratings of responsibility (Havighurst and Taba, 19h9; Brown and Landsberger, 1960; Gough et a1., 1952; Harris et a1., 1955). Havighurst and Taba (19h9) implied, without actually testing it, that children who had household duties for which they were respon- sible were rated as more responsible in the high school and community setting of their study. This assumption was common in both the folklore and semiprofessional literature of develOpmental psychology. Despite this, research showed no significant correlation between a child's household chores and develop— ment of responsibility (Harris et a1., l95ha; 195hb; Harris, 1958). In a work setting, job aspiration was signifi- cantly correlated with high ratings of respon- sibility. Length of time on the job, in so far as it correlated with aspiration, was also positively correlated with ratings of responsibility. It should be noted that these variables were used to explain why women were generally rated lower in respon— sibility. Sex roles being what they were, women had lower job aspirations and higher turnover rates (Brown and Landsberger, 1960). In addition to the behavioral description of the responsible person, several other characteristics have been reported. a) Gough et a1. (1952) reported that responsible students are more intelli- gent, active, and tolerant, while least responsible students score significantly higher on the MMPI scales D (depression), Pd (psych0pathic deviate), Sc (schizo- phrenia), and Ma (mania). b) Harris (195k) suggested that responsibility may not be a single trait at all, but rather, a parsimonious label which ex- pressed the quality of adjustment by an individual to the demands and expectations made of him. Responsible persons are 1h healthy persons, while irresponsible children "present a clear picture of poor personality integration and adjust— ment." This was repeated in a later report (Harris et a1., 1955) where it was concluded that responsibility "is closely associated with general emotional and social adjustment." Areas which Harris et al. found to discriminate high and low responsible students included: general background data (IQ, socio-economic index, mother's and father's education); play activities; family adjustment scale; and test items reflecting moral judgment, interest-attitude (delinquency), and psychoneurotic items. c) Brown and Landsberger (1960) found no consistent relations between responsibility and personal adjustment as measured by six personality items; five from the Cornell "Faith In People" scale. They further found that while responsibility did seem to be highly related to the generally "good character," it could be reliably discriminated from other aspects of such a "good character." For example, they demonstrated that on the subscales of the Gordon Personal Profile, respon- sibility correlated only 0.10 to 0.1M with emotional stability. They generalized from this to the conclusion that respon- sibility was a discriminable trait, and not merely an aspect of general adjustment. d) Homant (1969), using a semantic differen- tial technique, found that dimensions associated with responsibility involved trustworthiness, competence, honesty, respectability, and maturity. Research subsequent to these major investigations has added further information: 8) Bronfenbrenner (1961) reported some effects of parental authority on the development of responsibility. He used teacher ratings as the measure of responsibility and compared them to the child's reports of parental behavior. There was found to be a curvilinear relationship between responsibility and discipline with optimal points different for 9) 10) 15 boys and girls. Paternal authority was found to facilitate responsibility in boys, but impede it in girls. High levels of responsibility in boys was associated with moderately strong discipline, especially from the father, and warmth and nuturant attitudes, especially from the mother. For girls, a low moderate level of discipline was associated with highest responsibility. Too much dominance by either parent led to lower levels of responsibility. Mischel (1961) reported a study in which he tested the hypothesis: "that subjects preferring smaller immediate rewards would show less social responsibility than subjects preferring larger delayed rewards (1961, p. 2)." His results indicated that delinquent subjects achieved lower scores on the Harris (1957) scale and preferred smaller immediate reinforcements. Nondelinquents achieved higher scores and tended to prefer the larger delayed reinforcement. Berkowitz and Daniels (1963) carried out a series of experiments to test the existence, validity, and strength of the "social respon- sibility norm." That norm was derived from a norm of reciprocity proposed previously by Gouldner (1960). The social responsibility norm stated: not only should peOple help and not injure those who hurt them, but people should aid those dependent on them. Using a condition in which a supervisor's evaluation was dependent upon the production of his supervisees, they found greater production in high dependency conditions (1963). In conditions where subjects had previously been helped by another peer, they found even greater effort on behalf of the now dependent peer. They found also that the effort measure was significantly correlated with a self report scale assessing socially responsible tendencies. The scale used by Berkowitz and Daniels (196k) was a modified version of the 1957 Harris scale. Based on a sample of 77 college students, Berkowitz and Daniels selected those 22 items which correlated .h5 or higher with the scale as a whole. Berkowitz and Conner (1966) replicated support for the social responsibility norm. In this study, using a similar methodology, they further found that prior frustrations decrease the probability of conforming to the norm, and success experiences increase the probability of conforming. 16 11) Stone (1965a, 1965b) reported a correlation of 0.65 between the Social Responsibility Scale and the Marlowe-Crown Social Desirability Scale. He also reported a correlation of 0.53 of the Social Responsibility Scale with the Edwards Social Desirability Scale. He argued that this indicated that the Social Responsibility Scale was being inflated by the motivation to answer in a socially approved manner and, hence, was not accurately reflecting an individual's level of responsibility. Berkowitz replied (1965) that what those correlations showed was the importance of awareness of societal norms as a concommitant of both the approval seeking behavior and the conformity aspects of responsibility. 12) It should be emphasized that the ratings of responsibility thus far reported were all external to the individual evaluated. The assessment scales developed with the previously stated paradigm of these research efforts were validated on external ratings by peers, teachers, and other supervisors. Reported correlations between ratings of responsibility by others and self-ratings of responsibility have been below 0.10. There is obviously a difference in perception between those who judge the figure of external activities, and those privy to the range of mitigating circumstances which provide the ground from which the figure emerged. This difference in perception has lead to another major approach to the study of responsibility. The Study of the Perception of Responsibility and Causality, In concession to the essentially private status of motivation, and to the even greater difficulty of apprehend- ing all the facets of any given event in order to rigor- ously and accurately study causality, some investigators have approached the research on responsibility in a different manner. An outstanding example of this approach was a series of studies conducted by Shaw and Sulzer. They chose to focus upon the process of making judgments concerning responsibility rather than on responsibility as a personality characteristic. 17 Shaw and Sulzer (196A) suggest that "one significant determinant of interpersonal behavior is the degree to which one individual holds another individual responsible for events in the interpersonal life space (l96h, p. 3h)." By "holds responsible" they mean that the judge perceives the acting agent to be the cause of a given outcome. Shaw and Sulzer propose a direct linear relationship between the degree to which a judge perceives the acting agent as the cause of a given outcome, and the amount of responsibility the judge will.attribute to that agent. The perceived causality of action can be influenced by many variables. Among these are: 1) degree of environ- mental determinacy, 2) developmental level of the perceiver in terms of perceptual differentiation, 3) the quality of the outcome-~good versus bad, h) the intensity of the out- come, and 5) characteristics of the specific situation and their interaction with the characteristics of the judge. The first of these variables, and the first examination of the attribution of responsibility, was pr0posed by Fritz Heider (1958). Heider suggested that five levels of attributed responsibility could be 1 may be labeled as follows: considered. These levels Level I. Global Association: The person is held responsible for any effect that he is connected with in any way. In Piaget's (1955) terms, responsibility at this most primi- tive state is determined by 1These labels assigned to levels are not Heider's, but were formed by Shaw and Sulzer and accepted by this author to facilitate communication. Level Level Level Level II. III. IV. 18 syncretistic, pseudocausal connections. Thus a person may be blamed for the harmful acts committed by friends. Extended Commission: The person is held responsible for any effect that he pro- duced by his actions, even though he could not have foreseen the consequences of his actions. As in Piaget's (1932) "objective responsibility," the person is judged according to what he does, but not according to his motives. Careless Commission: The person is held responsible for any foreseeable effect that he produced by his actions even though the effect was not part of his goals or intentions. He is held responsible for the lack of restraint that a wider cognitive field would have produced. Purposive Commission: The person is held responsible for any effect that he produced by his action, foreseeing the outcome and intending to produce the effect. This corresponds to Piaget's "subjective responsibility" in which motives are the central issue. Justified Commission: The person is held only partly responsible for any effect that he intentionally produced if the circumstances are such that most people would have felt and acted as he did. That is, responsibility for the act is at least shared by the coercive environment (Shaw and Sulzer, 196%. p. 39). Heider theorized that little responsibility would be attributed to an individual for acts performed by the individual's friends (Level I). Judges would perceive in- creased causality and, hence, attribute increased respon- sibility as the situation in question changed from Level II to Level III and Level IV. Level V should show a decrease in perceived causality, and also a decrease in l9 attributed responsibility. Heider also proposed that adults would evidence a greater differentiation of personal and environmental attributed responsibility than would younger children. Shaw and Sulzer (196k) tested Heider's hypotheses of the relationships among the levels, and on the develop- mental differentiation. In an experiment with groups of children and adults, they found that the five levels of environmental influence did produce significant effects. They also found a significant interaction between popula- tions and levels, thus supporting Heider's conception. In addition to the variables proposed by Heider, Shaw and Sulzer (196A), Sulzer (196A), Shaw and Garcia-Esteve (1968), and Shaw, Brisco, and Garcia-Esteve (1968) have reported that outcome quality exerts significant effects upon attributed responsibility. They found that negative outcomes resulted in greater attribution than did positive outcomes. "...Apparently, individuals are more willing to blame another than to give him credit for his actions (Shaw and Sulzer, 196A, p. hh)." This effect was modified for different levels in different experiments, but held stable as a trend across them all. The intensity of the outcome has also been shown to have a significant effect upon the degree of attributed responsibility. Shaw and Sulzer (196A), Sulzer (l96h), Sham Brisco, and Garcia— Esteve (1968) reported that very good or very bad out- comes appeared to result in greater attribution of responsibility than a milder outcome provided. 20 Shaw and Sulzer observed in early experiments that the characteristics of the story used to present the situation to be judged could have significant effects, especially in interaction with the population variable of personal relevance. In the 196h experiment, they discussed the impact of group norms, and shared exper- iences upon the production of the observed effects: Children attributed less responsibility when the actor was a child, and adults were "less willing to give credit for unintended good effects, or for good effects performed under 'coercion' when the actor was a child." This is consistent with Wright's (1960) findings that individuals are more willing to attribute responsibility to authorities than to peers. They also attribute a greater amount of responsibility to authorities than to peers. In addition to the variables discussed by Shaw, Sulzer, et a1., other variables which attenuate perceived responsibility have been postulated. Wallach, Kogan and Bem (1962, 196A) have reported "...individuals, when constituted as a group, experience a diffusion of responsibility as a product of the knowledge that one is deciding upon an action jointly with others rather than deciding by oneself (196A, p. 263)." They found that this diffusion resulted in decisions which had greater attached risk than decisions made by isolated individuals. Subsequent research suggested the operation of a norm 21 favoring risk taking. Further, evidence was gathered that manipulation of such variables as moral context of decision, riskiness of most vocal group member, or accurate knowledge of probabilities outweighed the diffusion effect (Bem, 1970, personal communication). In another area, the effect of diffusion was drama- tically demonstrated and replicated under a variety of conditions. Barley and Latane (1968) reported testing the hypothesis that: the more bystanders to an emergency, the less likely, or the more slowly, any one bystander will intervene to provide aid. Their results clearly supported this hypothesis. This aspect of the diffusion of responsibility was used in discussing social phenomenon often explained in terms of apathy, alienation, and anomie. Often researchers have approached the study of psychological aspects of causality and responsibility using different terminology. Brehm and Cohen (1962) discussed a concept labeled volition, by which they meant the power for "initiation and selection of behavior," and acceptance of accountability for its consequences. Rotter (1966) began an extensive research project into perception of causal efficacy using the principle of locus of control; internal versus external. The research to date has indicated a number of variables which influence the interpersonal evaluation of responsibility. One approach could be characterized as Rated Responsibility. That previously stated paradigm 22 proposed the evaluation of subjects as high or low on responsibility as a personality characteristic, or perhaps, a quality of adjustment. Usually the definition or criterion of responsibility was supplied by the researchers. Sometimes it was implicit in the assessment technique. Personality characteristics, test response patterns, and some antecedent experiences were demonstrated to correlate with high or low ratings on responsibility. A second approach may be characterized as Perception of Responsibility. In this approach, rather than study the persons about whom judgments of responsibility were made, the research concerned itself with the process of making judgments about responsibility. Independence of activity, nature of the outcome of activity, and relations between agent and judge have been shown to affect that process. While the research has provided information about responsibility, it has not resolved many of the diffi- culties involved in making accurate evaluations of responsibility. Several types of behaviors have been shown to be indicative of high evaluations of respon- sibility. These behaviors can be grouped into several components or subcategories of responsibility. There has been some disagreement about the number and labeling of such components. There has been little work done to establish the differential importance of the various components in actual evaluations of responsibility. 23 The results of the previous research can be used to establish some characteristics of a situation suited for further investigation of responsibility. The work of Harris et a1. suggested that responsibility became better deve10ped and more salient with the passage of adolescence. It was argued that the more pressing realities of adult life called forth previously undemon- strated standards of responsibility, as when the irrespon- sible adolescent suddenly manifests responsibility in the face of a new situation involving job, wife, children, etc. The work of Shaw and his associates supports those notions. They demonstrated that the perception of responsibility was more differentiated in adults. Their results also indicated that pigh realitonf outcomes affected judgments of responsibility. Havighurst and Taba, Gough et a1., and Harris et a1., all reported that responsibility was especially manifest in issues of broad moral concern. The work of Shaw et a1., and Darley and Latane demonstrated that the relationship of the individual judged to those making judgments affected the judgments made. The clear— est judgments came from situations involving independence and visibility of the behavior to be judged. The greater the intensipy of the outcome, the greater the perceived responsibility, according to Shaw et al. This was especially true when the acting agent was seen as a figure of authority. 2h One area of investigation, which appears to meet the above criteria, is presented in the field of medicine. The role of the individual physician involves lengthy training in preparation for performing a wide range of professional activities. These activities deal with the ultimate realities of life and death, as well as many more mundane considerations. Although sometimes con- sidered the head of a health delivery team, the activi- ties associated with the role of the physician are largely individual. The authority and responsibility are con— sidered to rest on his shoulders. The physician acts in the presence of a range of perceivers with different statuses; peers, subordinates, coworkers, and clients. All these characteristics suggest that the field of medicine is an appropriate setting for the continued study of responsibility. The salience of the concept of responsibility can be clearly demonstrated in medicine. Virtually every school of medicine professes that the preparation of responsible physicians is one of its central objectives. Responsibility is a central tenet in the practice of medicine, as can be seen in the Hypocratic Oath, and the concern evidenced on the t0pic during frequent ceremonial oratory. Despite these considerations, there has been little controlled investigation pertaining to medical respon- sibility. The professional journals use the term 25 responsibility in an amorphous way, usually referring to general policies for institutions}frequently associated with fiscal concerns. The accepted grounds for evaluating responsibility have not been clearly codified and widely disseminated. For example, Snyder (1967) states: Relatively few studies have investigated trainees' concept of the qualities which characterize established members of a particular profession.... Previous studies in medical student socialization for example have been limited to comparison of the conceptions, values, and attitudes among students with different amounts of medical school training. In reference to his own study, Snyder says: "Professional responsibility is defined as a sense of obligation to four recipients: the medical discipline, patients, colleagues, and society at large." Among the results of his own study, Snyder reports: 1) that medical school experiences do affect professional responsibility orientation; 2) that premedical school experiences retain lasting effects on students' orientation to professional responsibility; and 3) that informal contacts with faculty and students in that order are the most influential experiences. A) Another very important fact reported in his study was that about 3h$ of the students in the study had no clear-cut conceptions of the professional respon— sibilities of the physician. The findings of Snyder's study were consistent with the findings of earlier researchers on the operation of social norms and expectancies upon conceptions of respon- sibility (Havighurst and Taba, 19h9; Gough et a1, 1952; Harris et a1., l95ha). In a socialization process as 26 extensive and intensive as most medical schools, vagueness by students about group expectations can perhaps most readily be explained in terms of contradictions or vagueness about such expectations among the socializing agents. Such an interpretation is consistent with the experiences of the divergent views associated with most medical school faculties. In a technical report of the American Institute for Research, there was some discussion of the responsibilities of the physician as these might be derived from observation of their activities. That report classified responsibility according to: l) Accepting responsibility for welfare of the patient. a) Devoting necessary time, effort, and care. b) Asking for help on consultation. c) Following instructions or advice. 2) Accepting responsibility to institution. a) Coordinating with other hospital staff. b) Furthering the interests of the organization. c) Promoting good public relations. 3) Accepting responsibility to medical profession. a) Cooperating with public health programs. b) Preventing illness in patient's family and contacts. c) Preventing illness in hospital personnel and other patients. 27 h) Accepting responsibility to medical profession. a) Maintaining ethical standards. b) Developing one's own knowledge and skills. c) Contributing to medical knowledge. They further stated: Responsibility is concerned with actions reflecting the doctor's attitudes toward his patients, the hospital staff, the community, and the medical profession. The emphasis is on the doctor's awareness of his important responsibilities, and on his maturity in fulfilling them. This codification of ideals did not present the actual behaviors appropriate to an evaluation of responsibility in physicians. Rather, it attempted to specify the areas in which responsible behavior must be enacted by the physician. It also implied the existence of clear group norms. That implication was not supported by Snyder's (1967) research. Gough, McClosky, and Meehl (1952) reported using a sample of hO senior medical students in their develOpment of a Personality Scale for Social Responsibility. Comparing faculty ratings of responsibility to scores on their scale, Gough et a1. obtained a correlation of only 0.22 with a standard error of 0.16. These few studies relevant to medicine have not contributed much toward a resolution of the difficulties involved in making an accurate evaluation of responsibility. At best, they have further delineated the areas on which investigation should be focused. At worst, they imply a clarity of conception and perception which is inaccurate and easily misleading. PROBLEM Earlier research began by presenting a definition or description of responsibility to judges. Judges were then asked to nominate associates who showed very low or very high congruence with the description, or to rank them in terms of congruence. The various descriptions used lead to questions concerning the accuracy of that initial description. Such questions cannot be answered on the basis of the accumulated research reports. The variation does, however, suggest the need for further research. What kind of consensus exists among various groups about the meaning of the term responsibility? What kind of consensus exists about the value of given behaviors as indicators of responsibility? What is the importance of the different components of responsibility? Does this differ from group to group? Previous research derived criteria on the basis of social reputation. A person's reputation is influenced by a variety of factors which may distort the delinea- tion of responsibility. It has been shown that persons tend to attribute good qualities excessively to those they like, and underestimate good characteristics in persons whom they do not like (Newcomb, 1961). PeOple also tend to see good qualities as highly correlated and, thus, tend to judge a particular quality not by its particular indicators, but on the basis of other 29 30 good qualities (Smith, 1968). In order to minimize these influences of interpersonal perception, descriptions of responsibility should be derived by asking respondents to exemplify the concept, rather than rate some individual. A second influence of interpersonal perception and attribution was demonstrated by Shaw and Sulzer (196A) and Wright (1960). They demonstrated the influence of the relative status of the acting agent and the judge. This influence can be minimized and/or standardized: 1) by having respondents attribute to a role model rather than a particular individual, and 2) by controlling, as an independent variable, the relative status positions of the respondent and the role. Hence, respondents can be classified according to their role relations to the role of the physician. The nature of the role relationships implies certain factors previously shown to effect perception of respon- sibility. In so far as responsibility reflects helping dependent others (Berkowitz et a1., 1963), the pattern of interdependence among the different roles should be reflected by different emphasis by different groups. To the extent that responsibility is developmentally influenced (Harris et a1., l95h; Shaw and Sulzer, 196A), groups differing only on an age-experience dimension should show greater similarity than groups differing on a larger number of dimensions. This should be reflected both in the conceptualization of responsibility, and the behaviors selected as indicators of responsibility. 31 The previous research has presented descriptions of the behaviors found to be indicative of responsibility. The number of behaviors reported is quite large. Several categories have been proposed to group these behaviors into a lesser number of components. A limited number of these major components were represented as subsuming different types or classes of behavior. While there has been a good deal of variation among the specific behaviors cited by different researchers, the categories generally showed more similarity, and the overlap of the major com- ponents was large. 0n the basis of these facts, a group of components were selected which encompassed all the categories and behaviors reported in previous research. Any example or description of responsibility from the previous research could be classified within these six components. Table I presents a chart of the categories given by earlier researchers and shows the portions covered by the new component system. The definitions below outline the kinds of behavior subsumed by each component. 1) Independence: Items in this category exempli— fied a spirit of autonomy. The agent does things his own way, even if that is frowned on; he tries to rely only upon himself, both in terms of thought and performance. 32 TABLE I Components of Responsibility Identified by Various Research Teams Loftus Havighurst Gough, Harris, Homant, & Taba McCloskey et. a1. Berkowitz & & Meehl Stone, Brown & Landsberger Independence Self con- Strong fident Personal Non-au- security thori- tarian person- ality Assurance Mature Requires Requires little little super- super- vision vision Reject Reject Aprivilege privilege Initiative nitiative Poise Initiative Solves pro- Resource- blems ful Seeks service Faces diffifi culties Reliability Efficient or dependabil- Fulfills Does his Gets things Capable ity tasks part done i Carries thru Dependable Dependable Dependable Reliable Punctual Punctual onsistent 33 TABLE I (cont'd.) Homant, Loftus avighurst Gough, Harris, & Taba McCloskey et. a1. Berkowitz & & Meehl Stone, Brown & Landsberger Reliability Accepts Account- or conse— able dependabil- quences ity Trust- Can count Trust- worthy on worthy Serious Square Honest shooter Orderly Conformity Conven- tional Conforming Conforms to social norms Sociability Loyal to Sense of Loyal group obliga- tion to group Group High social centered desirabil- ity "Socially forth- coming" Thinks for Helps depen— good of dent others others Sociable Respected 3h TABLE I (cont'd.) of world Loftus Havighurst Gough, Harris, Homant, & Taba McCloskey et. a1. Berkowitz & & Meehl Stone, Brown & Landsberger Altruism Sense of justice Prepare Concern for con- for tribution broader to society social moral issues Construc- tive interest Rightness 2) 3) h) 5) 6) Initiative: Dependability: Conformity: Sociability: Altruism: 35 The agent does more than the contract requires. He may exercise foresight in antici- pating difficulties and/or their solutions. He may exhibit either a new technique or a thoroughness beyond the expected in carrying through his tasks without supervision or special encouragement. He does the job. He is con- sistent and trustworthy and can be counted on to fulfill the major (and most of the minor) expectations. He can and does make commitments, and carries through on them. He accepts the consequences of his actions. The agent fulfills the expecta- tions made of him by the position within which he is working. He performs tasks in the prescribed manner. He avoids activities that rock the boat. He can be counted on to follow the party line. The agent contains his own feelings and individuality sufficiently to maintain effective relations. He controls his relations to yield effective task orientation. Able to work even in strained situations, he exhibits thoughtfulness to co-workers, associates, patients, and others. The agent sacrifices himself for the benefit of another (especially when that sacrifice is not expected by the other). He per- forms well beyond the normal contract in attempting to do good for the other. He is con- cerned with the broader aspects of his position and with the ramifications as well as the immediate consequences of policies he may make or uphold. 36 The present study was designed to elicit the behaviors which were used in making judgments or attri- buting responsibility appropriate to the role of physician. Different subgroups within the field of medicine were asked to give behaviors which they believed demonstrated high responsibility or irresponsibility, and to give a definition of responsibility. The subgroups varied on several dimensions, including identification with the role of practicing physician, and the nature of the interdependence of activity with that role. The questions to be asked of the data were as follows: 1) 2) 3) h) 5) 6) Can subjects give specific behaviors which they consider responsible behavior or irresponsible behavior? Can subjects give a definition of responsibility? Can these responses be reliably coded into one of the six aspects of responsibility represented by the given component system? Does this coding result in'a distribution of items to categories which indicates greater frequency of use for some components? Are the.distributions of responses different for the different subgroups? If the distributions for.subgroups are different, can these differences be explained by the differential role relations of the subgroups to the physician? Are medical students more similar to faculty or physicians than to paramedical personnel or patients? Since paramedical personnel and patients interact with physicians in a social context of relative dependency, do they emphasize aspects of sociability? Faculty represent a kind of special dedication to the discipline of medicine per se, and are charged with training future physicians in the proper understanding and performance of the role. Do they emphasize or employ more frequently than 7) 37 other groups the component of conformity to norms? Patients represent the group with the most obvious dependence on the physician. Do patients more frequently cite behaviors which reflect the aspect of altruism which character- izes outstanding response by the physician to that dependency? Do other differences in the use of the components represent differences which can be attributed to clear differences in the nature of the role interdependencies? What is the conception of responsibility relevant to physicians, and what are the behaviors judged to reflect this? Can this be used to begin a more systematic program of enhancing responsibility in physicians? Some of these questions can be statistically tested as empirical hypotheses: 1) 2) 3) The first null hypothesis to be tested will be that: There will be no significant differences in the use of different categories, i.e., HO = there will be equal frequencies in each category. The second null hypothesis will be that: There are no differences among the subgroups, i.e., H0 = the distribution of responses to components will not be statistically significantly different across subgroups. The third null hypothesis to be tested will be that: Any differences that are found to be statistically significant will not reflect the nature of the role relations among the subgroups. Specifically: a) HO = medical students will be no more similar to practicing physicians and faculty than to patients and paramedical personnel. b) HO = paramedical personnel and patients will use the component sociability no more frequently than other groups. c) HO = faculty will use the component conformity no more frequently than other groups. d) HO = patients will use the component altruism no more frequently than other groups. METHOD Subjects Five groups of subjects were selected for partici— pation in the study. These groups reflect different role relationships to the role of the practicing physician. The groups were: 1) Practicing physicians: These subjects were selected from the staffs of patient care facilities in the Lansing, Michigan area, including: a) St. Lawrence Hospital, b) Edward W. Sparrow Hospital, c) Ingham Medical Hospital, d) and 01in Memorial Health Center, Michigan State University. 2) Medical school faculty: Members of the faculty of the College of Human Medicine at Michigan State University who were actively engaged in teaching and/or evaluation of medical school students during spring term, 1969. 3) Medical students: The members of the medical school classes, years IV, V, VI, at Michigan State University during spring term, 1969. h) Paramedical personnel: Members of the suppor- tive health professions, including nurses, registered and practical; x—ray technicians; and physical therapists from the institutions listed under Item 1. Paramedical personnel from Ingham Medical Hospital were not sampled. 5) Patients: The consumers of medical care at the various institutions—-in-hospital patients at the St. Lawrence and Edward W. Sparrow Hospitals, and both in-patient and out-patient groups at Olin Memorial Health Center were sampled. 39 AD The procedure for administration of the survey questionnaire was as follows: 1) 2) Practicing physicians a) b) C) d) St. Lawrence Hospital: More than 200 questionnaires were distributed to the in-hospital mail boxes of all staff physicians. The approval and endorse- ment of the staff physicians' executive committee was first obtained, and the cooperation of the hospital administra- tion was given throughout the project. Edward W. Sparrow Hospital: Following the approval of the administration and staff executive committee, the investi- gator gave a short introduction to the survey, and then distributed 60 copies of the form to attending physicians at a monthly staff luncheon. This was followed by a short period of time for completing the questionnaires. The form was collected following the luncheon. Ingham Medical Hospital: The investi- gator was introduced at a staff meeting of hospital physicians by the Director of the hospital. After a short intro- duction, the investigator distributed 50 copies of the questionnaire, and collected them at the close of the meeting. Olin Memorial Health Center: Following the distribution of a Director's memo urging cooperation, nine questionnaires were distributed via internal mail service to the student health service staff physicians at Olin. Faculty of the medical school a) b) Forty-five questionnaires were distri- buted by campus mail or courier to members of the faculty who were actively engaged in teaching or evaluating medical students. Twenty—five questionnaires were administered at a meeting of the Office of Medical Education Research and Development. 3) 5) Al Medical students A 20-minute portion of class time was reserved for the introduction and administration of the questionnaire to year IV, V, and VI, respectively. Paramedical personnel a) b) C) St. Lawrence Hospital: An explanation, and procedure for administration were given to the Head of Nursing, who then distributed h5 questionnaires to members of the staff. Edward W. Sparrow Hospital: The investi- gator addressed a nursing staff meeting and distributed 30 questionnaires. Four nursing students were surveyed at the time of the staff meeting mentioned under Item l-b. Olin Memorial Health Center: Twenty questionnaires were distributed to nurses, x-ray technicians, and physical therapists to be filled out and returned at their convenience via house mail. Patients 8) C) St. Lawrence Hospital: Twenty-five questionnaires were given to the Head Nurse for distribution to in-hospital patients. Edward W. Sparrow Hospital: Twenty questionnaires were distributed to the nursing heads of the various services for sampling of in-hospital patients. Olin Memorial Health Center: Twenty questionnaires were delivered by the investigator to in—hospital patients, left overnight, and collected the following day. A supply of questionnaires was available at the out-patient desk. Each patient who came for medical care was offered the opportunity to respond. A2 Instrument One basic questionnaire was used in this study. It was modeled after those used by Flanagan (l95h) in his development of the Critical Incident Technique. That tech- nique was evolved to enhance the reliability of judgments made in field settings, and improve their reliability in terms of subsequent coding. Due to the extremely limited free time available to most of the subjects in the study, the form was kept as brief as possible. The questionnaire was pretested, and showed that a single response to each item required approximately five minutes for the average subject. The form also allowed for expansion of any response at the subject's option. The questionnaire asked each subject to: "Give an example--a specific act——by a physician or student physician you have known or heard of who demonstrated what you consider to be respon— sible behavior." "Give an example--a specific act-—by a physician or student physician you have known or heard of who demonstrated what you consider to be irrespon— sible behavior." "Give a definition of responsibility." Each questionnaire was attached to a cover letter appropriate to the sample group being surveyed. This letter explained the purpose of the research project. It also gave an example of the type of response requested. The example was a behavior appropriate to the concept of honesty. The cover letter also listed the endorsements of the appropriate peOple. (See Appendix D for a copy of the questionnaire and cover letter.) A3 Procedure--Judging The responses elicited by the survey were to be rated, by three judges, as to which of the six components of responsibility they reflected. In order to prepare the judges for that task, a series of practice sessions were held using behaviors drawn from the previous research, or from pilot studies associated with pretesting the instru- ment for this study. The judges were given the following instructions: "Here is a list of responses that were intended to describe what the respondent meant by respon— sible behavior. Look over the list. As you can see, some responses are identical to others; some are similar in vocabulary; some are similar in meaning. Here is a table which shows the results of previous research. It includes various types of behaviors. 0n the left are some words that have been used to label different components of responsibility. Your task is to take each of the response items on the list in front of you and assign it to one of the six components of responsibility." Practice sessions continued until the judges began to feel comfortable with the task and were developing consen- sus on the meaning of each component. The judges were then asked to write a definition for each of the terms used to label the components. These definitions were used as the established meaning for each component. Practice coding continued until reliability reached acceptable levels, i.e. index of agreement = 0.80. The definitions of the components were reviewed at this time to verify that there was a clear consensus among the judges on the meaning of each term. The definitions presented on page 28 were drawn up and used to code the responses of the survey. M: This coding procedure produced a list of behaviors given as examples of responsibility and a distribution of response frequency per component for each of the sample subgroups in the study. It also produced a list of examples of irresponsibility and the distribution of response frequency per category for each subgroup, although this list had special difficulties to be discussed below. A list of definitions and the distri- bution of response frequencies per component was also produced. The practice sessions began by using the behaviors previous research presented as being descriptive of responsible persons. In progressing from research examples to data gathered in pilot tests of the instru- ment for this study, difficulties developed in rating the obtained examples of irresponsibility. The scheme of components for coding the behaviors was developed from the previous research. The scheme resulted in confusion and unreliability in dealing with examples of irrespon- sibility. Several attempts were made to develop a component scheme which would yield reliable coding of examples of irresponsibility. None of these efforts were successful. The failure to develop a reliable coding scheme for examples of irresponsible behavior can be attributed to several factors. The first involved a decision as to whether irresponsibility represented the absence of a 145 positive attribute, e.g. initiative, or the opposite of the positive. Should the irresponsible equivalent of initiative be lack of initiative? Or should it be some opposite? Should the irresponsible pole of initiative be represented by inertia, or sloth; indifference, or conformity? These questions could not be addressed on the basis of previous research since no previous research had provided any evidence regarding the nature of irresponsibility. A second group of factors, which contributed to the lack of a coding scheme for irresponsible behaviors, derived from the characteristics of the responses them- selves. There was more technical language, and more exact details were given. Greater emphasis was placed by the respondents upon the uniqueness of this situation. These facts made the process of abstracting and classify- ing these items more difficult. Shaw and Sulzer (196A) pointed out that persons are more willing to give blame than to give credit, especially to authorities. This effect may have contributed to the stronger attention to detail exhibited in examples of irresponsible behavior. Responses given as examples of irresponsibility were put aside for analysis at a later time. In contrast to the failure to achieve reliability in coding responses given as examples of irresponsibility, the component system worked very well for coding examples of responsible behavior, and definitions of responsibility. 1+6 The responses obtained during the survey were compiled and arranged into lists. The lists were organized according to the item of the questionnaire to which the response was addressed. For each group, there were three lists: one which included all definitions of respon- sibility, a second which contained all examples of responsible behavior, and a third which included all examples of irresponsibility. For reasons explained above, the list of items for irresponsibility was put aside for subsequent study. The lists of items for the other two response types were of such a length that each group's responses could be scored as a unit. Thus, all the examples of responsible behavior given by faculty members could be scored in a single session. The lists were presented to the judges in accord with the instructions outlined previously. The judges then rated each item according to the aspect of respon— sibility it most clearly represented. Occasionally, an item would represent more than one of the components. In such cases, the judges were instructed to choose the category with which the item was primarily or predominately concerned. If the item could not be scored in a single category, the judge was told to list the multiple categories to which it might belong. Approximately ten percent of the items were handled this way. The scoring of that item was then based upon the component which was most frequently used by all three judges. 1:7 To estimate the reliability of the judges in the process of coding the items, an index of agreement was computed on each list. That is, one score was derived for faculty definitions, one score for faculty examples of responsible behavior, and so on. The index of agree- ment was computed by summing the number of items on which any two of the judgments disagreed, and dividing that sum by the total number of judgments made. When sub— tracted from 100%, this number indicates the proportion of judgments on which there was agreement. Thus: # of actual disagreements of possible disagreements Index of agreement = l - The profile of the scoring process in terms of the judges' agreement is given by the following table (Table II). Table II indicates that the judges did in fact show a high degree of agreement about the way specific behaviors, or definitions fit a given aspect or component of responsi- bility. The overall average across groups and items was well above the minimum satisfactory level. It does indicate a high reliability in the scoring process for this study. On the basis of this common and reliable discrimination of the different aspects of responsibility, statistical tests were made to examine the hypotheses previously stated for the study. AB TABLE II Index of Agreement Among Judges Definitions Examples Average Physicians 9A% 80% 87% Faculty 93% 83% 88% Medical Students 92% 95% 93% Paramedical Personnel 92% 87% 89% Patients 87% 86% 86% Total average 91% 88% 90% TABLE III Survey Response Rates # Forms # Forms % Response distributed Returned Physicians 320 A3 15% Faculty 70 AA 63% Medical students 77 66 85% Paramedical personnel 99 85 85% Patients - in 65 h1* 63% out ? A2 ? Total ? 321 ? 7 more came in too late to be scored indeterminate w a u u RESULTS Response Characteristics A differential return rate was observed among the questionnaire.distributed according to the description provided in the methods section. The following section reports on the response rates for each group in the study. 1) Practicing physicians a) Thus, St. Lawrence Hospital: Of the 200 questionnaires distributed through the hospital mail service, 1A were returned. Edward W. Sparrow Hospital: Of the 56 questionnaires distributed to physicians (four went to nursing students), 12 were returned at that time. One was subsequently returned by mail, for a total response of 13. Ingham Medical Hospital: Of the 50 forms handed out, 15 responses were obtained. Olin Memorial Health Center: From nine questionnaires, one reply was received. of approximately 300 questionnaires which were distributed to respondents, some A3 responses were obtained. This corresponds to a response rate of approximately 15%. 2) Faculty of the medical school a) b) 0f the A5 copies of the survey distributed to teaching faculty, some 23 were returned promptly. The return of 21 responses for 25 question- naires for the Office of Medical Education Research and DevelOpment reflects four persons who disqualified themselves from the classification as faculty. 50 51 The receipt of AA responses from the 70 questionnaires distributed established a return rate of about 63% for faculty. 3) Medical students The three classes of medical students have 25 to 27 students each. The return of 66 completed questionnaires is largely a re— flection of class absences at the time of administration. One student chose not to return the form at all; one other returned a blank form. Forms were sent to those who were not in c1ass—-only one returned. Of the 77 medical students polled, 66 replied. Thus, the response rate was nearly 85%. A) Paramedical personnel a) St. Lawrence Hospital: Of the A5 question- naires given to the Head of Nursing, A0 were returned. b) Edward W. Sparrow Hospital: Twenty—eight responses were obtained from the 30 nurses gathered at the meeting. Two declined to answer. Four responses were added by student nurses (see 1—b) for a total of 32. c) Olin Memorial Health Center: After much delay, and reminders, 13 of the 20 question— naires were returned. For the group, paramedical personnel, of the 99 questionnaires handed out, 85 responses were obtained for a return rate of approximately 85%. 5) Patients a) St. Lawrence Hospital: Of the 25 forms given to the Head Nurse, 2A were returned. b) Edward W. Sparrow Hospital: Of the 20 questionnaires given to the nursing staff, seven were returned in time to be scored by all judges. (Seven more were subsequently recovered, having been lost in the U.S. mails.) 52 c) Olin Memorial Health Center: Of the 20 copies distributed within the house, 10 were returned. The remainder of the patient responses came from the out-patient services associ- ated with Olin. There was no record kept of a response rate in this group, because none was feasible. Thus for the patient group, the data on response rate were indeterminate. All the data on response rates can be summarized as found in Table III (see page A8). Analysis of the Distributions As noted at the conclusion of the Method Section of this paper, the results of the judging process were lists of items given as responses. These lists were organized according to: l) The medical subgroup from which the items were obtained. 2) Their classification as examples or definitions of responsibility. 3) The component of responsibility into which they were coded and associated with each component, a frequency count of the number of items it contained. That data is summarized in the following tables. In addition to the tables, the data are presented in graphic form to facilitate comparison and to provide a succinct description of the nature of the distributions. Analysis of the data took several forms. The first was the description of the within-group distributions. The statement of the first null hypothesis was that the components of responsibility will have equal frequencies 53 TABLE IV Distribution of Responses Given as Examples of Responsibility Physi- Facul- Medical Para- Patients Total cians ty students medical Person- nel Dependability 1A 22 17 A0 32 125 Initiative ll 13 2A 17 1A 79 Altruism 7 7 19 5 1A 52 Sociability 5 2 3 20 16 A6 Independence 1 O l 0 2 A Conformity 0 0 l 0 0 l # of responses 38 AA 65 82 78 307 # of subjects in sample A3 AA 66 85 83 321 # not respond- ing 5 O l 3 5 1A spamssemnm Hmcnomnmm Hwowvmamndmu: mpqmcspm Hwoflvmzum hpddowmum mcwHOthnmuH wwrm m N_H_ _ 4 m N m : m_m a. m a m N H m z m N H m A N H II .HI H IL IL o m oa ma om mm om mm o: m: on mm om mm or m» ow mm Hwflanomnou monocnom6coH Asaaapsaeom smflsanas o>apmeHcH npsafinaeseaea owe Ipnmo Ihom hpwawpamsommmm Ho madmadxm mommpcwohmm mmnommom mo madam H mmDlo 55 TABLE V Distribution of Responses Given as Definitions of Responsibility Physi- Facul— Medical Para- Patients Total cians ty students medical person- nel Dependability 22 32 A8 69 52 223 Initiative A 0 2 6 5 l7 Altruism 2 3 7 l 6 19 Sociability 10 6 5 h 9 3A Independence 2 2 2 1 7 1A Conformity 1 0 2 2 3 8 # of responses A1 A3 66 83 82 315 # of subjects in sample A3 AA 66 85 83 321 # not respond- ing 2 l o 2 l 6 mpqmwpdmnm HonQOmnmm Hecadoamhmmu: mpnmosvm Hsofidmznm hpadocmum mssfiofimhnmua LN w,: m m A. A m a. m.: m m a n A m m A m A m m H_, n a m m H .7; IA IL oa ma em I. mN om mm o: m: L on A R om ri mm .IL o» L 2. I. on no hpwauomqoo monmocwmoonH hpwawnmwoom amwshpa< o>wpdflquH Hpfidwnwvnomon own Ipqoo Inom spaaanaasoasom so escapaaflaon mommpnmohmm omnommom mo nacho HH mmDOHm 57 of use. A Chi-square analysis was performed on the distribution of each group's items, and the frequency with which they fell into the various categories. This corresponds to a l x 6 Chi-square of each of the columns in Tables IV and V (see pages 53 and 55), comparing the hypothesis of equal frequencies with the observed fre- quencies. The results of that analysis are given by Table VI (see page 58). A Chi-square value of 20.52, with 5 degrees of freedom, indicates that there is less than one chance in a thousand that the obtained frequencies would have been observed if there were no differences in the frequencies with which the different components were used. On the basis of these findings, we reject the null hypothesis. This leads us to believe that there are differences in the frequency with which different components are used. Some components are used more frequently in making judg- ments of responsibility. In addition to the differences in use among the components, there are several other characteristics of the distribution which should be noted. There is a clear predominance of responses in the component of dependa- bility. In fact, more than 50% of the responses in the total survey were scored in this category. With the exception of one distribution in ten obtained, respon- sibility was most frequently judged by the component of dependability. The one exception was that, in the 58 TABLE VI Chi Square Analysis of Group Distributions Comparing Observed Frequencies to Hypothesis of Equal Frequencies Definitions Examples Physicians X2 = A8.15* X2 = 21.65* Faculty x2 = 106.07* ~x2 = 52.30* Medical students x2 = 151.25* x2 = A9.09* Paramedical Personnel X2 = 265.A8* X2 = 87.89* Patients x2 = 130.A3* x2 = 50.92* *p (.001 TABLE VII A Chi Square Comparison of Distributions of Examples vs. Definitions Physicians Faculty Medical students Paramedical personnel Patients **p (.01 i 10.75 20.A1** A0.31*** 29.29§** 19.91** ***p (.001 59 distribution of examples of responsibility, medical students gave more examples of initiative. Another characteristic of the distribution, which should be noted here, was the almost complete absence of examples of responsibility in the category of conformity. Only one item out of the more than 300 was classified in this category. This point will be further discussed in the next section. A second step in the analysis of the data resulted from the contrast of the marginal totals of the distri— bution of examples and definitions, as found in Tables IV and V. This comparison indicated that there may be differences between these two distributions in the use of the various categories of responsibility. This was tested statistically by a 2 x 6 Chi-square analysis of those two marginal columns. The result was a Chi-square value of 95.63. This is significantly well above the .001 level of confidence (with 5 df, p(.OOl, x2 = 20.52). This statistic indicates that there were indeed different frequencies associated with the different components depending on whether the response was given as an example or definition of responsibility. To determine which of the groups exhibited this difference between examples and definitions, a set of 2 x 6 Chi-squares was run to contrast the distributions of examples and definitions. This amounted to a column by column comparison of Tables IV and V. The results of that analysis are contained in Table VII (see page 58). 60 Four out of the five cases were statistically significant. In the fifth case, the statistic just missed significance. (For 5 df, p( .05, x2 = 11.07). Examining the differences between definitions and examples in another way may extend our understanding of these differences. By comparing the relative frequencies in terms of percentages, it should become clearer that one major difference between the two distributions is in the variance among categories other than dependability. In the distribution of definitions, where dependability accounts for such a large proportion of the responses (71.5%), the other categories appeared to be rather equally weighted. The data are presented in Table VIII (see page 61). The analysis of the data as presented in Tables VI, VII, and VIII lead to a rejection of the first null hypothesis. That hypothesis stated: There will be no significant differences in the use of the different categories, i.e. there will be equal frequencies in each category. Significantly different frequencies were observed in the different categories. Further, there were different frequencies in the use of the categories, depending upon whether the items were given as examples or definitions of responsibility. A third step in the analysis of the data was an investigation of the statistical differences among the frequency distributions obtained from the various groups 61 TABLE VIII Distribution of Responses into Categories by Percent Definitions Examples Average Dependability 71.5% h0.h% 55.9% Initiative 5.2% 26.5% 15.8% Altruism 6.2% 17.A% 11.8% Sociability 10.3% l3.A% 11.8% Independence h.h% 1.3% 2.85% Conformity 2.3% .03% 1.16% 62 of respondents. The null hypothesis was that there would be no significant differences among the different subgroups. To test this hypothesis, Chi-square analyses were made to compare the distributions. The first analysis was a 5 x 6 Chi-square of Tables IV and V respectively. The distri- bution of responses given as examples of responsibility produced a Chi-square value of AA.88. The analysis of items given as definitions of responsibility yielded a Chi-square value of 32.80. With 20 degrees of freedom, a Chi-square value of 31.01 corresponds to a confidence level of .05. Therefore, the obtained values of X2 examples = AA.88, X2 definitions = 32.80 are both significant at the .05 level. These significant differences are evidence for the rejection of the null hypothesis of no differences among the distributions of different groups. These overall tests indicated significant differences. A series of smaller comparisons were then run to determine which groups contributed to the differences. Tables IX and X give the results of a series of 2 x 6 Chi—squares which compared the distributions of each pair of groups. For a 2 x 6 analysis, Chi-square has five degrees of freedom, and a Chi-square value of 11.07 corresponds to the .05 level of confidence. Five significant differences were observed in Tables IX and X (see page 63). These differences, when added to the differences reported on the overall tests, indicate the rejection of the second null hypothesis. 63 TABLE IX Two by Six Chi Square Analysis Pairwise Comparison of Respondent Groups Examples of Responsibility Physi- Faculty Medical Para- Patients cians students medical personnel Physicians I 3.8 5.A 9.3 1.6 Faculty I 8.0 10.3 6.A Medical students I 23.7*** 17.3** Paramedical personnel I 7.8 Patients I *‘H' = P (.01 “HI“. = p (.001 TABLE X Two by Six Chi Square Analysis Pairwise Comparison of Respondent Groups Definitions of Responsibility Physi- Faculty Medical Para- Patients cians students medical personnel Physicians I 8.0 9.8 15.8“ A.8 Faculty I 1.6 15.8“ A.3 Medical students I 7.9 3.9 Paramedical personnel I 12.7* Patients I .= p<.05 6A The nature of any obtained differences between sub- groups was the subject of the third null hypothesis, which predicted that any differences which were found to be statistically significant would not reflect the nature of the role relations among the subgroups. Figures 111 and IV (see pages 65 and 66) represent the distribution of responses to components by the different subgroups. The third null hypothesis made four specific predictions. The first was that medical students and faculty would be no more similar to practicing physicians than patients and paramedical personnel. This implied that the direct association with the M.D. degree and the identification with the role of the physician could not be used to discriminate the way in which the components of responsibility would be used. Tables IX and X show that no significant differences were found among the groups of faculty, student and physician. Table IX indicated that medical students were significantly different from both patients and paramedical personnel in the use of components for examples of responsibility. Table X indicated that paramedical personnel are signifi— cantly different from physicians and faculty, but not medical students in the distribution of responses given as definitions. Table X also shows a significant differ- ence between paramedical personnel and patients. The obtained results did not provide clear grounds for rejecting the null hypothesis. While differences among FerJh.. 111 Distritutiun of Responses Examples of Re‘ptusibiiity Per- Dcpenda- Initia- Socia— Inde— Con- centgfic bility tive Altruism bility pendencc for-it! 80 75 70 as 60 55 50 us A0 35 3O 25 2O 15 10 Mn ii s“ '11 —- P‘fnedIC‘l can—- students personnel Physicians 'lCUlty—— Patients ICQRE 1V Distribution of RespUBScs Definitions of Res;uhsibility T {er- Deprnia- Initia- Cocia- inde- Cen- CthLflfif bility_ tive Altruism LilityAiehienco fnrmity 80 \ s \ \ To 65 60 55 50 £5 3 A0 35 30 Meli0a1--- Paramedical students personnel Physicians Faculty -— Patients “m 67 groups were obtained, these differences did not clearly show a discrimination between "M.D.'s" and others. The second prediction of the null hypothesis stated that paramedical personnel and patients will use the component of sociability no more frequently than other groups. This prediction was based on the setting in which patients and paramedical personnel interact with physicians. Based on expectations associated with role relations, since the patients and paramedical personnel were frequently dependent upon the physician, they would be expected to stress the component of sociability. Tables XI and XII (page 68) and, XIII and XIV (page 69) present the data. A comparison of the distribution of examples to sociability versus other categories (Table XI) yielded a Chi-square value of 16.92 which is significant at the .005 level of confidence. To determine the groups which contributed this significant difference, a series of smaller analyses were performed. Table XII gives the results of that series. A comparison of the distribution of definitions to sociability versus other categories (Table XIII) yielded a Chi-square value of 12.71 which is significant at the .05 level of confidence. To determine the groups which contributed this significant difference, a series of smaller analyses were performed. Table XIV gives the results of that series. 68 TABLE XI Response Distribution Sociability Versus Other Categories Examples of Responsibility Physi— Faculty Medical Para- Patients Marginal cians stu- medical sum dents person- nel Socia- bility 5 2 3 20 16 A6 Other 33 A2 62 62 62 261 Marginal sum 38 AA 65 82 78 307 TABLE XII Chi Square Analysis Pairwise Comparison of Groups Sociability Examples of Responsibility Physi- Faculty Medical Para- Patients cians students medical personnel Physicians I 2.09 2.AA 1.99 .93 Faculty I .00 7.82** 5.70** Medical students I 10.7A** 7.78** Paramedical personnel I .3A Patients I ** = p (.01 69 TABLE XIII Response Distribution Sociability versus Other Categories Definitions of Responsibility Physi- Faculty Medical Para- Patients Marginal cians stu- medical sum dents person- nel Socia— 10 6 5 A 9 3A bility Other 31 37 61 79 73 281 Marginal A1 A3 66 83 82 315 sum TABLE XIV Chi Square Analysis Pairwise Comparison of Groups Sociability Definitions of Responsibility Physi- Faculty Medical Para- Patients cians students medical Apersonnel Physicians I 1.A8 5.9A* 10.A9** 3.77 Faculty I 1.16 3.32 .238 Medical students I .A9 .A9 Paramedical personnel I 2.51 Patients I *=P<'O5 **=P<.Ol 70 Table XIIl indicated that there were significant differ- ences between the groups on the distribution of examples to the category of sociability. Paramedical personnel and patients were significantly different from faculty and medical students. Figures I (page 5A) and III (page 65) clearly support the interpretation that paramedical personnel and patients use the component sociability significantly more frequently than other groups in giving examples of responsibility. In contrast to the distribution of examples, the distribution of definitions presents a very different picture. Table XIV indicated significant differences be- tween physicians and paramedical personnel. Examination of Figures II (page 56) and IV (page 66) shows that physicians gave more sociability definitions than did paramedical personnel. Paramedical personnel gave gew sociability definitions, in contrast to their frequent use of examples coded as sociability. The null hypothesis, which predicted no greater use of sociability by paramedical personnel and patients, is clearly rejected for the distribution of examples of respon- sibility. Paramedical personnel and patients did give more examples of sociability than did other groups—-significantly more than faculty or students. The distribution of defini- tions did not support rejection of the null hypothesis. 1While some of the frequencies indicated in Table XI are small, the Chi-square analysis was accepted because none of the significant differences obtained resulted from a comparison in which the expected frequencies were smaller than 5. 71 That faculty will use the component of conformity no more frequently than other groups was the third null hypothesis speficically made about group differences. Tables IV and V presented the original frequency tabulation of the data. Examination of those tables showed that faculty gave pg examples or definitions of responsibility which were coded as conformity. There were no grounds for rejecting the null hypothesis. The fourth group comparison suggested by the hypothesis was that patients will use the component altruism no more frequently than other groups. This hypothesis derived from the analysis of the role of the physician which is frequently idealized as "healer". It predicted that patients, who are least exposed to continual interaction with the role of physician, would maintain the most idealized conception of that role. Tables XV, XVI (page 72), and XVII (page 73) present comparisons of the use of altruism. For the Chi-square analysis which was performed on Table XV, a X2 = 6.251 was obtained. This value did not indicate any significant differences among the distri- butions compared. For the comparison given in Table XVI, a Chi-square value of 13.997 was obtained. With five degrees of freedom, that value was significant at the .05 level of confidence. The obtained difference was not identified by the groups which contributed to it. In order to determine which 72 TABLE XV Response Distribution Altruism versus Other Categories Definitions of Responsibility Physi- Faculty Medical Para- Patients Marginal cians students medical sum person— nel Altruism 2 3 7 l 6 19 Other 39 AO 59 82 76 296 Marginal sum Al A3 66 83 82 315 TABLE XVI Response Distribution Altruism versus Other Categories Examples of Responsibility Physi- Faculty Medical Para- Patients Marginal cians students medical sum person- nel Altruism 7 7 l9 5 1A 52 Other 31 37 A6 77 6A 255 Marginal sum 38 AA 65 82 78 307 73 TABLE XVII Chi Square Analysis Pairwise Comparison of Groups Altruism Component Examples of Responsibility Physicians Faculty Medical Para- Patients students medical personnel Physician I .09 l.A8 A.11* .OOA Faculty I 2.52 3.23 .0A5 Medical students I 1A.A7*** 2.A52 Paramedical personnel I 5.365* Patients I a = p (.05 **n = p (.001 7A groups were significantly different, a series of smaller comparisons were made, as reported in Table XVI. No differences were found for the distribution of definitions, as was shown by Table XV. Table XVII showed that there were significant differences on examples between paramedical personnel and all groups except faculty. Examination of Figures I and III indicated that the obtained differences reflect the very low use of examples of altruism by paramedical personnel. The patient group did not exhibit significant differences from the other groups, and thereforq the null hypothesis could not be rejected. The results of the statistical analysis of the data provide for the following conclusions: 1) The first null hypothesis to be tested was that: There will be no significant differences in the use of the different categories, i.e. HO = there will be equal frequencies in each category. This hypothesis was rejected. There were significant differences in the frequency with which different categories of responsibility were used. These differences vary depending upon whether the items were given as examples or definitions of responsibility. Dependability accounted for more than 50% of all responses. 2) The second null hypothesis tested was that: There are no differences among the subgroups, i.e. HO = the distribution of responses to components will not be statistically signifi- cantly different across subgroups. This hypothesis was rejected. Significant differences were observed among the subgroups in the frequency with which they used the different components. 75 3) The third null hypothesis tested was that: Any differences that were found to be statisti- cally significant would not reflect the nature of the role relations among the subgroups. Specifically: a) HO = Medical students will be no more similar to practicing physicians and faculty than to paramedical personnel and patients. While there was some similarity among physicians, faculty, and medical students, they were not consistently different than paramedical personnel and patients. That is, the more direct association with the M.D. degree did not yield a consistent similarity which could discriminate between the groups. b) H = Paramedical personnel and patients will use the component sociability no more frequently than other groups. This hypothesis was rejected for the distri— bution of examples, but not for definitions. Paramedical personnel gave significantly more examples of sociability than did faculty or students, and patients gave significantly more than students. c) Ho = Faculty will use the component conformity no more frequently than other groups. This hypothesis was not rejected. In fact, faculty did not use the component conformity at all. d) HO = Patients will use the component altruism no more frequently than other groups. This hypothesis was not rejected. Patients did not use the component altruism signifi- cantly more often than other groups. Medical students gave significantly more examples of altruism and initiative than did other groups. In addition to the statistical analysis of the differences among groups of respondents, and the quantitative 76 measures of the importance of different components of responsibility, the data were subjected to further exam- ination. One of the purposes of this research project was to observe and record the particular behaviors that subjects used to respond to questions about the concept of responsibility. What are the behaviors used to identify responsible and/or irresponsible persons? What areas do the groups consider to be of special concern? The differences observed between items given as examples and those which were definitions present difficulties in answering these questions. This will be discussed in the next section of this paper. For examples of responsibility, the following observations were made: Physicians gave examples of responsibility which emphasized the importance of a comprehensive follow through on diagnosis and treatment, focusing frequently on calls and visits intended to verify the progress of the patient. They gave examples which indicated the importance of assisting other physicians at times of heavy loads, such as emergencies or when normal assistance was unavailable. Realization of limita- tions on time and knowledge, and making appropriate referrals constituted another class of examples. Faculty examples focused upon limiting one's commitments in order to insure adequate care, and making referrals on that basis as well as one's lack of skill in a particular area. Faculty also gave examples dealing with more abstract 77 ethical implications, such as the withdrawal of support from terminal patients. Medical student examples focused on the humanity of the interpersonal contact, stressing care for details in both physical management and psychological support. Students also stressed efforts toward self-improvement through continuing education, and through active concern with the policy making appropriate to the setting in which one practices. Paramedical personnel focused upon areas such as treatment of the "whole person" including many examples indicating the importance of sharing with the patient information as to his condition and progress. They frequently mentioned the follow through on both the patient and his family as indicative of responsibility. Paramedical personnel included several examples which stressed the interaction of the health team, and the requirements of cooperation for efficacious outcomes. This included taking over for other physicians who were unavailable and utilizing the competence of others, as well as complying with policies intended to allow others to exercise their competencies. Planning ahead was frequently cited as especially indicative of responsibility. Patients' examples covered such considerations of the interpersonal contact as explaining the illness and treatment to the patient, and taking time to make the patient feel treated as a person rather than as a mal— functioning machine. This concern was usually exempli— fied as attention to what appear to be small details. 78 They also cited the use of consultations and referrals, as well as a thorough follow through, as indicative of the responsible physician. The physician's availability to patients was the one behavior most frequently cited by patients as an example of responsibility. The responses which were given as definitions of responsibility showed a tendency to be more abstract and general. A synthesis of the definitions given by physicians would include such factors as: sensitivity to the needs and feelings of others, competent decision making and, treatment of others as one would like to be treated--to the best of one's ability. Faculty responses showed a clear and predominant concern with accountability: responding to perceived needs and obligations, especially as defined by the traditional role model. They emphasized that the physician was answerable for his performance-~to himself, and to the community. Medical students exhibited concern with making a response which was appropriate to the expectations of others. This was true especially in terms of the norms of interpersonal interaction, although it seemed also to hold for standards of practice. Awareness of one's capabilities, and the limitations thereof, and being "right" were mentioned by students as parts of responsibility. The theme of "do unto others as you would have them do unto you," was perhaps the single most pervasive response given as a definition of responsibility. All groups included at least some items which made 79 ' but medical specific reference to this "golden rule,‘ students used it most frequently. The notion of reciprocity included in this theme will be discussed further in the next section. Paramedical personnel gave definitions of responsibility which were clearly weighted toward a sense of obligation, and response to that obligation—-accountability. Paramedical personnel gave stress to an element of resiliency or adaptability to changed expectations, as when standard procedures are modified. Action to improve one's surroundings was also included in definitions of responsibility by paramedical personnel. Patients indicated that the physician's duty should include a considerate effort to improve the understanding of those with whom he is working. The physician should be Sensitive to the expectations of others, and demonstrate a sense of obligation and concern in terms of attention to detail. In general, the definitions stated that the physician should show awareness and consideration of others, adapt his behavior as appropriate to their expectations, and practice the kind of reciprocal considerateness implied by the "golden rule." Some of the directions in which physicians were urged to modify their actions were demonstrated in those behaviors cited as exampges of irresponsibility. Irresponsible behaviors exemplified by physicians indicated several areas of concern. Motivation was exemplified 80 especially by greed, or overwhelming concern with finances. Other motivations considered irresponsible were less tangible. Generally, they represented something other than the altruism prescribed by the Hypocratic Oath, and the standards professed by the profession. Inappropriate dress or language was cited as irresponsible behavior, indicating that one of the physician's responsibilities is to maintain a certain image. Vacationing without a replacement, overcommitment, and other behaviors which limited the physician's availability were cited as irresponsible actions. Failure to follow through, especially on follow up treatment of patients contributed several examples of irresponsible behaviors given by physicians. Faculty gave examples of irresponsible behaviOr which included: inadequate or improper delegation of authority, failure to control emotion-- especially failure to control expression of emotion, failure to be available and/or follow up on the conse- quences of prescribed treatments, and failure (of students, particularly) to demonstrate appreciation for the efforts of others. Most frequently, students gave examples of irresponsibility as incompetence or improper patient treatment. They also gave examples in which the physician failed to respond, e.g. preferred to sleep, or was in poor condition to respond-~drinking or drunk. Paramedical personnel most frequently cited emotional problems as irresponsible behaviors; Overreaction by the physician 81 to frustrations, verbal or even physical harrassment of support personnel, and lack of sensitivity to the emotional discomfort of the patient constituted most of their examples of irresponsible behaviors. Other areas of irresponsibility mentioned by paramedical personnel were: failure to visit hospitalized patients (sometimes for weeks), failure to respond to requests for such visits, disregard of established operating procedures, and improper delegation of authority. The focus of attention by patients in attributing irresponsibility to physicians was apparent incompleteness or inadequacy of diagnosis and/or being "rushed through" with insufficient attention. These factors frequently result in a third frequently mentioned behavior--pill pushing—-which may alleviate symptoms, but oft times leads to subsequent difficulties for the patient. Patients also mentioned gruffness and inconsiderate behavior in general as indicative of irrespon— sibility on the part of a physician. The latter was the most frequent complaint. Fully 10% of the patients gave no example of irresponsible behavior on the part of a physician or expressly stated that they knew of none. DISCUSSION The purpose of this study was to examine the particular behaviors which groups of respondents used to judge the concept of responsibility, particularly in reference to the performance of physicians. Previous research had presented some behaviors indicative of the characteristic respon— sibility, but had not presented an adequate outline for accurate determination of the importance or frequency of citation of different types of behavior. Components derived from previous research were used to combine sets of particular behaviors, and assess their relative impor- tance or weight in the process of making judgments about responsibility. Group differences in use of different components were examined. Response Characteristics Practicing physicians. Cooperation by administrative personnel at all levels (hospital director, professional committees, and hospital administrators) provided direct access to practicing physicians for both direct presenta- tion and mail distribution of survey forms. Despite the concern thus demonstrated, a response rate of only 15% was obtained from practicing physicians. The questionnaire was deliberately created to require a minimal amount of time, and, in fact, time was provided during the two cases of direct presentation. The failure of approximately 70% of the physicians attending those sessions to complete 83 8A the form implied a significant resistance to examination of the topic of physician responsibility. The attribution of resistance to the topic of responsibility, rather than to participation in survey research, was based on two factors. First was the evidence that other surveys of the same group of physicians obtained significantly higher response rates (Molineux, 1969-70, personal interviews and questionnaire--95%; Michaelman, 1968—69—-50%). The second factor was reflected by comments on some of the forms which were returned. The most explicit such response was, "We prefer not to put these on paper." Other comments directed at the questionnaire form, or the term responsibility, indicated that the topic was indeed important, but was not amenable to detailed specification. The combined evidence indicated that physicians felt either that responsibility was too nebulous to be defined empirically, or that it was inappropriate for empirical investigation and best left to the cOnscience of the individual practitioner. The former alternatiVe was rendered less tenable by the response of the A3 physicians who did cooperate, and by the responses of the other groups. Other gropps of respondents. The response rates for other groups was within the average range. No particularly outstanding characteristics were observed. The Distribution of Resppnses The first null hypothesis—-that the components of responsibility would be used with equal frequencies, was 85 rejected on the basis of the significant X2 reported in Table VI. The alternative hypothesis was that some components were more frequently used as the basis for making judgments of responsibility. This was true for items given as examples and items given as definitions of responsibility. The most logical and parsimonious interpretation of these data was that differences in frequency of citation reflected differences in the frequency of use of the components in making judgments of responsibility, and concomitantly, differences in the contribution of those components to the conceptualization of responsibility. On the basis of that interpretation, the results given in Table VIII can be used to delimit the importance of the various components in determining judgments of responsibility about physicians. Table VIII showed that dependability was clearly the most important aspect of responsibility. This finding was consistent with the reports of earlier researchers. Havighurst and Taba (19A9) stated that, "completing accepted jobs characterize(s) the concept of responsibility." Brown and Landsberger (1960) reported that reliability was selected as most important by 7A out of 10A supervisors. Their use of reliability was very similar to dependability in this study. Gough, McClosky and Meehl (1952) also emphasized the importance of dependability. 86 In this study, the particular behaviors which were coded as dependability for physicians called special attention to the areas which were used to evaluate whether the physician was adequately fulfilling his contract. Special emphasis was placed upon the importance of a thorough examination, with attention to detail, and a consistent and considerate follow up of the progress of the patient. This included reasonable limitations upon the number of patients accepted in order to guarantee adequate care, making appropriate referrals, and offering explanations to the patient and/or his family. Availability was frequently used to characterize dependability on the part of a physician. A second finding revealed in Table VIII is the similar frequency with which sociability, initiative, and altruism are used. Previous research has not provided any clear understanding of the relative importance of these other aspects of responsibility. While dependability is clearly predominant, it equally clearly cannot be used alone to make accurate evaluations of responsibility. There were significant differences in the way the different components were used in the distribution of examples versus the distribution of definitions. This point will be subsequently discussed. For the purpose of the moment, we will treat the average rankings of the components. That responsibility is more than mere dependability has been suggested by all previous research. The components 87 used in this study were composed of, or derived from, the results of previous research. This is the first study, however, which reported the frequency of use for these other components. This is the first time that some data has been offered as to the relative frequency with which the different components were used to characterize the concept of responsibility. Table VIII showed that sociability, initiative, and altruism account for 11.8%, 15.8%, and 11.8% respectively of the items in the study. Together they account for 39.5% of the items. Taking time to explain one's intentions and plans to patients, nurses, or other coworkers was the behavior most frequently cited as sociability within the physician's role. Initiative frequently reflected some appropriate action in unusual circumstances, and special efforts to offer high level medical care in spite of difficulties. Altruism was most frequently characterized by dependable responses in unusual circumstances. Being available during unusual times, rendering services without charge, and making special efforts to comfort patient's family were frequently cited. The relative importance attributed to the components, independence and conformity, in previous research was not supported in this study, which had fewer than 5% of the responses in these categories combined. The importance of these categories in previous research appears to derive from the school setting in which the research was conducted. 88 Conformity and independence were given more emphasis by Harris and his associates than by any other researchers. Gough, et. a1., were also closely associated with primary schools and reported emphasis upon these components. Havighurst and Taba, who used several sources in addition to the school setting, mentioned the qualities conformity and independence with less emphasis. Brown and Landsberger carried out their research in a work setting and put very little emphasis upon the aspects of conformity and independence. In this study, two groups gave no examples of indepen- dence, and four groups gave no examples of conformity. In contrast to this, there was only one group, faculty, who gave no definitions coded as conformity. All groups gave at least one definition, which was coded into the independence category. This study was carried out on an adult population in a work setting. It would seem that the primary school setting evidenced special concern with conformity and independence, partly because it has a limited capacity for imposing standards or norms and a vast number of such norms to instill. Such concerns are consistent with the function of the primary school as an important socializing agent. In contrast to this, established work or professional settings have a wide and interlocking set of procedures for enforcing conformity to group norms, and it is, there- fore, a concern which is less pressing and less obviously 89 difficult. In the primary school setting, with its socializing function, the group itself has done little toward the actual evolution and construction of the norms. Further, the situation is one in which authorities are attempting to enforce a conformity which is not yet fully supported by the peer group itself. Homans (1959) has pointed out some effects of group involvement in the evolution of group norms. One of these is that it becomes a less conscious concern of those leading the group, and enforcement becomes an automatic function of group membership. Snyder's (1967) findings, that orientation toward professional responsibility was largely developed during informal group interaction, suggests that such a mechanism may well be in operation here. It is suggested that the absence of responses which were coded as independence or conformity in this study reflects a transference of these qualities from conscious concern and verbal expression to an operational expression which is not verbalized because it is taken for granted. The work of Berkowitz and his associates demonstrated the existence and Operation of such a norm, viz., the social responsibility norm. Another factor which suggests the operation and importance of such norms was the assumed reciprocity appealed to in those responses which cited the "golden rule" in characterizing responsibility. Such norms are seldom referred to with direct verbalizations, but rather, are assumed and function as postulates for most social interaction. 90 In addition to establishing frequencies with which the different components were used in making judgments about responsibility, this study revealed a significant difference between the use of the components, depending on whether the responses were given as examples or definitions of responsibility. This unexpected finding does not reflect anything previously reported in the literature, either on responsibility, or on recall tasks. The obtained differences may be due to the overwhelming influence of dependability in the distribution of definitions. With 70% of the items in that category, there was less variance in the distribution of items to other categories. A reversal of the rankings of three of the components was shown by Table VIII. For definitions, the ranking was initiative, altruism, sociability. However, the average percentage frequency with which these components were used was very similar, and the difference among the importance of these categories does not appear to be too great. A second interpretation of the differences in frequency of use among these three categories may be derived from an analysis of the task given to survey respondents. Asking for a specific example of a concept, as opposed to asking for a wider conceptualization, may create different outcomes in terms of the respondent's set. No research on the difference between these tasks could be found in the literature, however, some hypotheses can be offered. When asked to respond with a specific 91 example, the subject attempts to recall some particular Observation he has made in the past. In recall of this nature, the principles of primacy, recency, and vividness are liable to produce a response which need not be typical of the performances normally observed. In fact, it can be argued that the more unusual it was, the more likely it is to be remembered. Thus, examples of respon- sibility could be expected to be more dependent upon recent experiences, and particular exposure. In contrast, the task of generating a definition calls for a response which clearly indicates the central aspects of the concept, even at the cost of losing some aspects which may be more peripheral. A definition must include the essentials, and fuller explication may be sacrificed to that purpose. While an example may consist of a particular physician's action on a particular occasion which caught the respondent's attention, a definition is expected to represent a more central tendency. The much greater frequency with which responses given as definitions were coded into the component dependability may indeed reflect just such an effect. Grogp Differences Significant differences were found in the frequency with which different groups used the components of respon- sibility. Examination of Figures I and II should put the investigation of those differences in the proper context. Those figures show that the general profile of responses 92 was quite similar across groups. There was a good deal of similarity in the way in which different groups ordered the components. There were, however, some statistically significant differences. The expectation expressed in hypothesis 3-b (page 37) was that paramedical personnel and patients would use the component of sociability to express the interaction out- comes which they would prefer to obtain from physicians. The rejection of the null hypothesis supported that expectation for the distribution of examples, but not for definitions. This finding could be best interpreted in the light of the previous discussion of task differences between the two types of responses. Such an interpretation presents some difficulty in explaining the performance of physicians, who gave significantly more definitions involving sociability than either paramedical personnel or medical students. No other interpretation, however, was developed from the obtained data. Other predicted group differences in the use of components for making judgments about responsibility were not supported. The clearest reason for this finding was the degree of similarity among all groups about the relative importance of the different categories. Some significant differences were obtained, which reflected the specific emphases expressed by different groups on different components, but did not reflect the predicted role relations among the groups. Following is an outline 93 by group of the response profile tabularly presented in Tables XVIII and XIX (see page 9A). Physicians gave fewer items which were coded as dependability than other groups. This was true for both examples and definitions. In fact, physicians were the only group which did not show significant differences between the two distributions. On the whole, physicians' responses were spread across the categories in a manner very similar to the average weighting of the categories based on all groups. The one exception to this was definitions of sociability, where physicians gave more responses. Physicians gave significantly more than paramedicals. An interpretation similar to that made of the dearth of independence and conformity items in this study can be made for physicians' use of dependability. It is a characteristic which is so demanded in their role, that they could come to take it for granted more than other groups. It would thus become less salient in recalling an outstanding example of responsibility. Sociability may be more frequently cited for a similar reason--that it is an outstanding performance which is difficult to maintain during the hectic day to day business schedule. Faculty responses did not exhibit the emphasis upon conformity which was predicted by the hypothesis 3-c of this study. In fact, they did not use this category at all. No clear interpretation of this finding can be 91+ TABLE XVIII Response Distribution Examples Depend- Initi- Altru- Socia- Inde- Con— ability ative ism bility pen- form- dence ity Physicians 37.8% 28.1% 17.9% 12.8% 2.6% 0.0% Faculty h9.9% 29.5% 28.7% h.5% 0.0% 0.0% Medical students 25.7% 36.2% 28.7% h.5% 1.5% 1.5% Paramedical personnel A8.8% 20.7% 6.1% 2A.A% 0.0% 0.0% Patients Al.6% 18.2% 18.2% 20.8% 2.6% 0.0% TABLE XIX Response Distribution Definitions Depend- Initi- Altru- Socia- Inde— Con- ability ative ism bility pen- form- dence ity Physicians 53.7% 9.8% A.8% 2A.A% A.9% 2.A% Faculty 7u.2% 0.0% 7.0% 13.9% h.6% 0.0% Medical students 72.5% 3.0% 10.5% 7.6% 3.0% 3.0% Paramedical personnel 82.8% 7.2% 1.2% A.8% 1.2% 2.A% Patients 63.h% 6.1% 7.3% 10.9% 8.5% 3.7% 95 drawn from the data. The reasoning about the nonverbal operation of the enforcement of group norms presented previously about conformity may apply a fortiore to the special agents of professional socialization. A further anomalie in the faculty responses applies to the category initiative. Almost 30% of the examples responses were coded as initiative, but no responses given as definitions were so coded. The interpretation suggested for this finding is that faculty evaluations respond to the pressures implicit in an academic setting, but do not consider initiative as truly necessary for the responsible practice of medicine. Paramedical personnel exhibited the most distinct response pattern. They showed frequent statistically significant differences from other groups on various components in both distributions. This was expected, although the specific hypotheses about the nature of the differences were not stated in a manner to be clearly supported by the data. As previously mentioned, they were significantly higher on examples of sociability. The paramedical group was also significantly lower on examples of altruism. These two facts, combined with an examination of the content of the responses from the paramedical group clearly indicate the conflicts which arise in the relationship between an authority and the subordinate who must implement decisions. Paramedical personnel, more than any other group, appear to suffer 96 from and react against vagueness about the physicians' responsibility. Patients exhibited a response pattern which was significantly different from some groups on particular components, but such differences were more likely to be due to the extremity of the other group. The one notable exception to that was the examples of sociability where patients gave a high frequency of responses. In general, patients' responses exhibited the dependability upon the role of the physician which is implied by the role relations. Their responses tended to emphasize the overt manifestations of good medical care--a thorough examination, time spent, and politeness. Medical students, as potential physicians, were expected to exhibit the effects of the socialization in which they are actively engaged. Their responses covered the whole range of components in each distribution. Medical students were significantly higher on examples of initiative and altruism, and low on examples of dependability. These results are fully consistent with the tradition of idealism attributed to aspiring physicians. As mentioned at the beginning of the discussion of group differences, the overall agreement among the groups about the relative importance of the categories outweighs the variation among the groups. Only part of one of the four predictions about the effect of role relations upon the differential use of components was supported. 97 Future Research Involving Responsibility One of the functions of the research reported in this paper was to establish the relative weighting of the different components of responsibility in order that those weights might be used in the evolution of a scale for measuring responsibility among physicians. Given the lengthy list of specific behaviors to be observed, and a preliminary knowledge of how important the various behaviors are, progress can be made in this direction. A combination of the data obtained in this research, and the methodology employed by Shaw, Sulzer, et. a1., could provide for a research program leading to an instrument for the prediction of responsibility. Armed with the list of specific events which were evaluated by the respondents as responsible or irresponsible, a series of vignettes could be developed which require the respondent to exercise his judgment in making decisions. The functional operation of such an instrument could truly fill the full meaning of evaluation, which includes learning as well. Rather than a trial and error method» ology, practice sessions could be established in which the subject is given opportunity to exhibit his sense of responsibility and obtain feedback about its congruence with professional standards. The concept of responsibility appears to be operation— ally defined by different behaviors according to different roles. The continued investigation of responsibility must 98 include establishing the specific behaviors used in making judgments of responsibility appropriate to the role under consideration, and establishing the relative frequency with which the different components of respon— sibility are used. This research has demonstrated at least one methodology for accomplishing these aims. REFERENCES 100 REFERENCES American Institute for Research. Classification of critical incidents--intern and resident performance. Technical publication--A. I. R., Pittsburgh: 1965. Anderson, N. H. The goodness of personality trait words. Journal of Personality and Social Psychology, 1968, 2, 272-279. Bem, D. J. A personal communication, 1970. Berkowitz, L. Response to stone. Journal of Personality and Social Responsibility, 1965, g (5), 757-758. Berkowitz, L., and Connor, W. H. Success, failure, and social responsibility. Journal of Personality and Social Responsibility, December, 1966, A (6), 66A-669. ~ Berkowitz, L., and Daniels, L. Affecting the salience of the social responsibility norm: effects of past help on the response to dependency relationships. Journal of Abnormal and Social Psychology, March, 196A, §§_(3), 275-281. Berkowitz, L., and Daniels, L. Responsibility and dependency. Journal of Abnormal and Social Psychology, 1963, g5 (5), h29-A36. Brehm, J. W., and Cohen, A. R. Explorationspgn cognitive dissonance. New York: Wiley, 1962. Bronfenbrenner, U. Some familial anticedents of respon- sibility and leadership in adolescents. I. L. Petrullo and B. M. Bass (Eds.), Leadership and Interpersonal Behavior. New York: Holt, Rinehart and Winston, Inc., 1961. Brown, A. W., and Landsberger, H. A. The sense of responsibility among young workers: part I, definition and measurement; part II, correlates, Occupational Psychology, January, 1960, 3A (1), l—lA and 73-85. Crowne, D. P., and Marlowe, D. The approval motive. New York: Wiley, 196A. 101 Darley, J. M., and Latane, B. Bystander intervention in emergencies, diffusion of responsibility. Journal of Personality and Social Psychology, 1968, 8 (A), 377-383. Flanagan, J. C. The critical incident technique. Psychological Bulletin, 195A, 2; (A), 327-358. Garcia-Esteve, J., and Shaw, M. Rural and urban patterns of reSponsibility attribution in Puerto Rico. Journal of Social Psygpology, 1968, 1A, lA3-1A9. Gough, H. G., McClosky, H., and Meehl, P. E. A personality scale for social responsibility. Journal of Abnormal and Social Psyphology, 1952, El. 73—80. Gouldner, A. W. The norm of reciprocity: a preliminary statement. American Sociological Review, 1960, 25, 161-179. Harris, D. B. How student teachers identify responsibility in children. Journal of Educational Psychology, Harris, D. B. A scale for measuring attitudes of social responsibility in children. Journal of Abnormal and Social Psychology, November, 1957, 52‘737, 322-326. Harris, D. B. Parents'judgment_of responsibility in children, and children's adjustment. Journal of Genetic Psychology, 1958, as, 161-166. Harris, D. B., Clark, D. E., Rose, A. M., and Vallesek, F. Personality differences between responsible and less responsible children. Journal of Genetic Psychology, 1955, £1. 103-109. Harris, D. B., Clark, K. E., Rose, A. M., and Vallesek, F. The measurement of responsibility in children. Child Development, March, 195Aa, 22, 21-28. Harris, D. B., Clark, K. E., Rose, A. M., and Vallesek, F. The relationship of children's home duties to an attitude of responsibility. Child Development, March, 195Ab, 22, 29-33. Hartshorne, H., and May, M. Studies in the nature of character, 3 vols. New York: Columbia University Press, 1928-1930. Havighurst, R. J., and Taba, H. Adolescent character and personality. New York: Wiley and Sons, 19A9. 102 Heider, Fritz. The psychology of interpersonal relations. New York: Wiley and Sons, 1958. Homant, R. J. The meaning and ranking of values. Unpublished Master's Thesis, Michigan State University, Department of Psychology, 1967. MisChel, W. Preference for delayed reinforcement and social responsibility. Journal of Abnormal and Social Psychology, 1961, 62 (1), 1-7. Newcomb, T. M. The acquaintance Process. New York: Holt, Rinehart, and Winston, 1961. Piaget, J. The moral judgment of the child. New York: Harcourt, Brace, 1932. Piaget, J. The language and thought of the child. New York: Meridian, 1955. Shaw, M. E., and Sulzer, J. L. A empirical test of Heider's levels of attribution of responsibility. Journal of Abnormal and Social Psychology, July, 196A, 62(1), 39-196 0 Shaw, M. E., and Schneider, F. W. Intellectual competence as a variable in attribution of responsibility. Research Report No._]. NSF Grant GS-6A7, University of Florida, 1967. Shaw, M. E., Briscoe, M. E., and Garcia-Esteve, J. Cross-cultural study of attribution of responsibility. International Journal of Psychology, 1968, 3 (1), 51-60. Smith, H. C. Social Perception: the development of inter- personal impressions; an enduringAproblem in psychology. Edited by Hans Toch and Henry Clay Smith. Princeton: Van Hostrand, 1968. Snyder, D. S. The relationship of_students' experiences before and during medical school to their conception of professional responsibility. Journal of Medical Education, 1967, As, 213-218. Stone, L. A. Rejoinder to Berkowitz, social desirability or social responsibility. Journal of Personality and Social Psychology, 1965a, g (57, 758. Stone, L. A. Social desirability correlates of social responsibility. Journal of Personality and Social Psycholosx. 1965b. 2 I57. 756-757. 103 Sulzer, J. L. Attribution of responsibility as a function of the structure, quality, and intensity of the event. Dissertation Abstracts, 1965, 25 (8), A8A6. Wallach, M. A., Kogan, N., and Bem, D. J. Group influence on individual risk taking. Journal of Abnormal and Social Psychology, 1962, 65, 75-86. Wallach, M. A., Kogan, N., and Bem, D. J. Diffusion of responsibility and level of risk taking in groups. Journal of Abnormal and Social Psychology, 196A, 66, Wright, J. M. Attribution of social responsibility and self—concept. Doctoral Dissertation, University of Colorado, Boulder, Colorado, 1960. Appendix A 105 Havinghurst and Taba, l9A9 Items from the Check List Relevant to Responsibility 69. Keeps appointments. 27. Dawdles at his work. 86. Is very conscientious. 2A. Is usually late for appointments. 71. Takes good care of school property. A7. Occasionally forgets an appointment. ll- Always forgets to do assigned homework. 5- Can never be depended on to complete a job. 20- Must be continually prompted to finish a task. 9A- Takes the initiative in assuming responsibility. 7o. Gets down to work without being prodded by others. 23- Is careless about employer's and school's property. 67. Is quite responsible for a boy (or_girl) of his age. 82. Finishes assigned work whether checked up on or not. 3A. When assigned homework, does only part of the assignment. 7A. Does helpful things at home such as cleaning up before a party. 92. Sees jobs to be done and does them without waiting to be asked. 93. Feels a strong obligation to finish well whatever he undertakes. 16. Never feels any need to care for his room or possessions at home. 18 Lets others do the work he has agreed to do for his class or club. 56. Carries through an undertaking about as well as others of his age. 83- Takes his share of the burdens in planning as well as in executing plans. 31. 1.1, 76. 7A. 13. A7. 1A. XX- 106 Quits work as soon as the "whistle blows,‘ even in the middle of a job. Finishes most assignments promptly but in a careless, slap-dash fashion. Works steadily and does not bother other people while teacher is out of room. When left in charge of a younger child, does not neglect it for something more interesting. People soon learn it is useless to assign him important tasks even if he is willing to accept them. Will carry out a task entrusted to him if it does not interfere too much with something he would rather do. Likely to drop or neglect a difficult responsibility without bothering to notify anyone or find a substitute. I do not have sufficient basis for judgment. 11. 12. 1A. 19. 22. 2A. 36. A8- A9. 50. 107 Havinghurst and Taba, l9A9 Items from the Student Beliefs Questionnaire Relevant to Responsibility Social Responsibility Scale A = Agree D = Disagree When assigned a somewhat difficult task at school or by your employer, you should be expected to work it out yourself, without the aid of adults. It is all right to be late to a meeting if you know that other people are going to be late also. Some persons are naturally carefree and forgetful, and so they must be excused when they fail to complete assigned duties. When you accept a job you should complete it, regardless of what happens to make it difficult to do so. People who have put off completing their school work should not be given extra time by their teachers to get their work in. When an adult tells you to do something which is more difficult than you thought it would be it is best to let him take over the task. When it is "quitting time" you should feel free to leave your job, even if the task on which you are working is not quite finished. When you have not the time to keep up your lessons and your friendships, it is better to neglect your lessons than your friends. Students who know the required subject matter should be given high grades even if they haven't completed all the required work, such as reports, notebooks, etc. A student need not feel that he should straighten up a disorderly room at school if he was not responsible for the disorder. When you see another student misusing school materials, you should not interfere in what is his own affair by trying to stop him. 53. 5A. 55- 72. 78- 79- 81, 83. 8A. 102. 106. 107. 108 You need not feel under any obligation to do things at school which have not been especially assigned to you by a teacher or a student officer. Students who are not willing to do the minor and somewhat boring tasks to help the school and teachers are not really good citizens of the school. Stars in athletics and other school activities should not be expected to be punctual about other school work. Members of a committee should expect club members not assigned to that committee to help them, if additional help is needed. If you see things which should be done at home or around your place of work, it would be foolish to do them unless you were asked to do so. Students who give much of their time to the school's activities should still be expected to work hard at their studies and not expect special allowances from their teachers. Talented people should not be expected to do the simple and uninteresting jobs of the school or of an organization. One should work hard for the school when needed, even if this means giving up an after-school job which is not a necessity but which would help ease the family financial affairs. It would not be fair to speak against a candidate who very much wants the office, even though he would not make a good officer. When it interferes with your job you should not talk in a friendly way with the-customers on whom you are waiting. You should be willing to drop out of school to help earn money for your family if your support is needed. If someone has "fallen down" on a responsibility, he should not again be entrusted with a responsible job. 109 D 108. Boys or girls should not be expected to help around the home if they are busy and if a maid or another adult is there to help. A 109- When the father is finding it difficult to get along on his wages, a high-school student should voluntarily help provide for the family, even if he is prominent in school and must give up school activities to do this. | l III||§ . Appendix B lo. ll. l2. 13. 1A. 15. 16. 111 Gough, McCloskey and Meehl, 1952 Personal Scale of Social Responsibility Every family owes it to the city to keep their sidewalks cleared in the winter and their lawn mowed in the summer. I wouldn't sneak into a movie even if I could do it without being caught. When I work on a committee, I like to take charge of things. Maybe some minority groups do get rough treatment, but it's no business of mine. Every citizen should take the time to find out about national affairs, even if it means giving up some personal pleasures. We ought to worry about our own country and let the rest of the world take care of itself. It's a good thing to know people in the right places so you can get traffic tags, and such things, taken care of. It is hard for me to act natural in a group of people. It's no use worrying my head about public affairs; I can't do anything about them anyhow. School teachers complain a lot about their pay, but it seems to me they get as much as they deserve. There's no use in doing things for people; you only find that you get it in the neck in the long run. When a person "pads" his income tax report so as to get out of some of his taxes, it is just as bad as stealing money from the government. We ought to pay our elected officials better than we do. A person who doesn't vote is not a good citizen. It makes me angry when I hear of someone who has been wrongly prevented from voting. If I get too much change in a store, I always give it back. 17. 18. 19. 20. 21. 22. 23. 2A. 25. 26. 27. 28. 29. 30. 31. 32. 33. 3A. 112 AS long as a person votes every four years, he has done his duty as a citizen. Police cars should be specially marked so that you can always see them coming. We ought to let Europe get out of its own mess; it made its bed, let it lie in it. I must admit I try to see what others think before I take a stand. When prices are high, you can't blame a person for getting all he can while the getting is good. I can honestly say that I do not really mind paying my taxes because I feel that's one of the things I can do for what I get from the community. People have a real duty to take care of their aged parents, even if it means making some pretty big sacrifices. I would be ashamed not to use my privilege of voting. I like to read newspaper articles on crime. When someone does me a wrong, I feel I should pay him back if I can, just for the principle of the thing. I have had very peculiar and strange experiences. AS a youngster, I was suspended from school one or more times for cutting up. Everything is turning out just like the prophets in the Bible said it would. I have never done anything dangerous for the thrill of it. I enjoy a race or game better when I bet on it. In school I was sometimes sent to the principal for cutting up. I feel that I have often been punished without cause. I liked school. 35. 36. 37- 38. 39. A0. A1. A2. A3. AA. A5. A6. A7. A8. A9. 50. 51. 52. 53. 5A. 55. 56. 113 I seldom or never have dizzy spells. I like science. I very much like hunting. My parents have often objected to the kind of people I went around with. I was a slow learner in school. I have never been in trouble with the law. In school I found it very hard to talk before the class. I do not dread seeing a doctor about a sickness or injury. I played hooky from school quite often as a youngster. I would like to be an auto racer. It is all right to get around the law if you don't actually break it. I have often found people jealous of my good ideas just because they had not thought of them first. In school my marks in deportment were quite regularly bad. I am fascinated by fire. I usually work things out for myself rather than get someone to show me how. When I get bored, I like to stir up some excitement. I like to read about science. A large number of people are guilty of bad sexual conduct. I am often sorry because I am so cross and grouchy. I would like to wear expensive clothes. I am afraid of being alone in a wide-open place. At times I feel like swearing. I I. | .1! I II Ill“ Appendix C 10. ll. 12. 13. 1A. 15. 16. 17. 18. 115 Harris, D. B., 1957 Scale for Measuring Attitudes of Social Responsibility in Children It is always very important to finish anything one has started. At school, it is easy to find things to do when the teacher doesn't give us enough work. Police cars should be especially marked so that you can always see them coming. It is no use worrying about current events or public affairs; I cannot do anything about them anyway. We ought to worry about our own country and let the rest of the world take care of itself. In school, my behavior gets me into trouble. I am hardly ever on time for meals. I have been in trouble with the law or police. When a person does not tell all his income in order to get out of paying some of his taxes, it is just as bad as stealing money from the government. A person who does not vote when he can is not a good citizen. I hardly ever get my school work done on time. I have played hookey from school. Every citizen should take the time to find out about current events if it means giving up free time. In school I am sent to the principal for being bad. Maybe some minority groups (Negroes, Indians, Mexicans, Jews, etc.) do get bad treatment, but it is no concern of mine. We ought to let Europe get out of its own mess. People criticize me for wasting time. When I work on a committee, I let other people do most of the planning. 19. 20. 21. 22. 23. 2A. 25. 26. 27. 28. 29. 30. 31. 32. 33. 3A. 35. 36. 37. 116 I am often late for school. If it is worth starting at all, it is worth finishing. I am the kind of person that people can count on. In school, I am one of those who can go on working even though the teacher is out of the room. People can count on me to get things done, without checking on me. I am frequently chosen as room helper or to run errands. I do my chores the very best I know how. I have been elected leader or president of my class. Nothing is more important than to be honest with other people. My teacher often complains because I don't finish my work. When you can't do a job, it is no use to try to find someone else to do it. It is more important to get the job done than to worry about hurting other people's feelings. Why bother to vote when you can do so little with just your one vote. "Never give a sucker an even break." Letting your friends down is not so bad because you can't do good all the time for everybody. Our country would be a lot better off if we didn't have elections, and people didn't have to vote. It's a good thing the Atlantic Ocean separates us from Europe because then we don't have to worry about them. It's more important to work for the good of the team than to work for your own good. I would never let a friend down when he expects something from me. 38. 39- A0. A1. A2. A3. AA. A5. A6. A7. A8. A9. 50. 117 People would be a lot better off if they could live far away from other people and never have anything to do with them. Every person should give some of his time for the good of his town or city. If everyone pitches in to do a job, it can always get done. It is a good rule to do something for your neighbor if he does something for you. Doing things which are important should come before doing things you enjoy doing. When a person doesn't like something he is supposed to do, he will get someone else to do it if he is smart. Cheating on examinations is not so bad as long as nobody ever knows. People have a real duty to take care of their parents when they are old, even if it costs a lot. I usually work things out for myself rather than get someone to show me how. I usually volunteer for special projects at school. Children often get punished when they don't deserve it. When given a task, I stick to it even if things I like to do better come along. It really doesn't matter whether parents attend Parent-Teacher meetings regularly. Appendix D 119 The attached questionnaire is part of a study* attempting to obtain a definition of what behaviors are meant by the term "responsibility". As you know, one of the goals of a medical school is to produce "responsible" physicians. It seems that you as a group could best help establish what that means. We will very much appreciate your response to the questions on the next page. Brief specific answers are best. As much as is possible, they should be descriptions of concrete behaviors. An example of an appropriate response in terms of "honesty" would be: Honest: 1. He returned borrowed money even though the lender had forgotten. 2. He returned extra change given him by a store's cashier. Dishonest: 1. He changed the price tag on a book he bought. 2. He told a lie to a professor about why he missed an exam. Pilot studies indicate that most people can complete the form in about ten minutes. The range is wide, however, and you should feel free to take as long as you desire. Your cooperation, and your time, are very much appreciated. When the form is completed, please fold and return to the address on the back of the questionnaire. Thank you, Greg Loftus *This study is part of a thesis for the M.A. degree with the Department of Psychology at MSU, and is sponsored by the Office of Medical Education Research and Development. 120 Please indicate your present position Please give at least one example-—a specific act--by a physician or student physician you have known or heard of who demonstrated what you consider to be responsible behavior. Please give at least one example-—a specific act--by a physician or student physician you have known or heard of who demonstrated what you consider to be highly irresponsible behavior. Please give a definition of responsibility. Do you have further comments that would be relevant to an investigation of medical responsibility? 2 l 137i. "17 ||||||||||||| 777717717 llll‘llllllllllllllml