L] B Malligan 5: ‘\’l’l \\\\\\\\\\1‘\/ ABSTRACT UNUSUAL BEHAVIOR AND THE ATTRIBUTION OF RESPONSIBILITY By Robert Dean Fischer A laboratory experiment was used to test the hypothesis that perceivers attribute a greater degree of responsibility for the outcome of his behavior to an actor who has behaved unusually than they would attribute to an actor who has behaved in the usual manner. Forty male and forty female subjects observed a video-tape recording of two confederates who played a laboratory game which re- quired each of them to make a decision which effected the outcome of the game. The subjects were instructed that the behavior of each confederate was either usual or unusual. The analysis of the data revealed that the hypothesis was supported. UNUSUAL BEHAVIOR AND THE ATTRIBUTION OF RESPONSIBILITY By Robert Dean Fischer A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Sociology 1972 M“; (£3. Copyright by Robert Dean Fischer 1972 With deepest appreciation to my grandparents Robert Lewis and Dorothy Mae Bown ii ACKNOWLEDGMENTS As the experimenter in this study, I have known all along who really was responsible and I would like to make an attribution or two, or three, myself. The following people get a "seven" on a seven-point bipolar rating scale from "Helped a Little" to "Could Never Have Done It Without Them." Much of the credit for this dissertation goes to my committee: Professors Jim McKee, Kevin Kelly, and Stan Kaplowitz. I want to say a special thanks to Bill Ewens, my committee chairman, for his encouragement, con- fidence in me, good counsel, and friendship. Those who lent their special talents to this project are: Greg Ingles and Mike Radeht, the two confed- erates; Chris Pruszynski the television cameraman; and Paul Emery and Ron Langley who did the computer programing. Others whose cheerful assistance made the work lighter, and whose pleasant qualities have endeared them to the graduate students in the Department are Marilyn, Jan, Colleen, Kay, and Mary Jane. A fond thanks goes to each for their many kindly deeds. Any remaining credit for this endeavor I will share with my wife Carol, just as we have shared the long hours of work, worry, and waiting that went into it. iii TABLE OF CONTENTS List Of Tables 0 O O O O O O O O O O O O O O O O UNUSUAL BEHAVIOR AND ATTRIBUTIONS OF RESPONSIBILITY . . . . . . . . . . . . . . . . Statement of the Problems . . . . . . . . . . Review of the Literature. Theoretical Development of the Hypothesis COLLECTION OF DATA . . . Experimental Design . Sample. . . Laboratory Setting. Cover Story . Stimulus Situation. Treatment . Instrument. Method of Analysis. . ANALYSIS AND INTERPRETATION. Unusual Behavior and Responsibility . Analyses of Variance. Summary of the Analyses of Variance Planned Comparisons Among Means . The Displacement of Responsibility. Mean Scores of Individual Treatment BIBLIOGRAPHY . APPENDICES . Groups. Degree of Association between Attributions. Summary and Conclusion. APPENDIX A: PAYOFF STRUCTURES . . . APPENDIX B: POST OBSERVATION QUESTIONNAIRE. APPENDIX C: TREATMENT INSTRUCTION SHEETS. . APPENDIX D: SCORE SHEET FOR DECISION MAKING EXPERIMENT . . . . . . . APPENDIX E: RECRUITING FORM . . . . . APPENDIX F: TIME SCHEDULE FOR SUBJECTS. . . APPENDIX G: REPORT REQUEST. . . . . . . . . APPENDIX H: RECEIPT FORM. . . . . . . . . iv 14 17 17 18 20 22 23 29 31 33 35 35 35 41 43 46 51 57 62 65 65 66 8A 88 89 91 92 93 TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE 10 11 12 13 LIST OF TABLES DISTRIBUTION OF SUBJECTS BY BEHAVIOR AND SEX. . . . . . . . . . THREE-WAY ANALYSIS OF DEGREE OF BLUE'S RESPONSIBILITY. . . . . . THREE-WAY ANALYSIS OF DEGREE OF RED'S RESPONSIBILITY . . . . . THREE-WAY ANALYSIS OF DEGREE OF GAME'S RESPONSIBILITY. . . . . . TWO-WAY ANALYSIS OF DEGREE OF RED'S RESPONSIBILITY--MALES ONLY. . TWO-WAY ANALYSIS OF DEGREE OF GAME'S RESPONSIBILITY--MALES ONLY . TWO-WAY ANALYSIS OF DEGREE OF RED'S RESPONSIBILITY--FEMALES ONLY. TWO-WAY ANALYSIS OF DEGREE OF GAME'S RESPONSIBILITY-—FEMALES ONLY MEAN LEVELS OF RESPONSIBILITY BY BEHAVIOR . . . . . . . . . . . . MEAN LEVEL OF RESPONSIBILITY BY BEHAVIOR AND PERCEIVER'S SEX. . . . MEAN LEVELS OF RESPONSIBILITY BY TREATMENT GROUP AND PERCEIVER'S SEX ASSOCIATION BETWEEN ATTRIBUTIONS BY TREATMENT GROUP AND PERCEIVER'S SEX ASSOCIATION BETWEEN ATTRIBUTIONS BY PERCEIVER'S SEX. . . . . . . . . 19 36 37 38 39 40 #0 Al AA 46 53 55 56 UNUSUAL BEHAVIOR AND ATTRIBUTIONS OF RESPONSIBILITY Statement of the Problems When the massacre of the women and children of Mai Lai was revealed, many Americans were stunned and outraged, but they were in the minority. Far more Americans defended Lt. Calley and his men on the grounds that the destruction of civilian pOpulations is commonplace in this war, and that there is no difference between annihilating entire villages from 40,000 feet with explosives and plastic pellets, and doing the same thing on the ground with small arms. After the 1968 elections, the news media exposed the fact that certain corporations had used phantom political organizations to funnel millions of dollars into political campaigns illegally. Although the corporations had violated the law, people shrugged it off because such chicanery is not at all uncommon in American politics. Recently, the National Poultry Farmers Association petitioned congress for permission to destroy one-fourth of all the laying hens in the country in order to reduce the supply of eggs and increase the selling price. Their solution to the problem of having surplus of food while, at the same time, having millions of hungry people in the nation and the world is to destroy the surplus food. Incredible! Criminal! And yet, not a murmur of protest is raised because creating 2 artificial scarcities in order to increase profits is standard operating procedure in this society. 0n the other side of the coin, all Americans are familiar with the intense hostility aroused by Black students who trespassed by sitting at lunchcounters where no Black had sat before. Equally familiar are the moral indignation and heated criticism with which radical changes in fashion are met: young men who were the first to wear long hair and beards, and young women who behaved "unusually" by wearing mini-skirts, hot pants, blue jeans, or pant suits, or simply by not wearing bras, before it became common to do these things, found themselves in serious trouble with school authorities and personnel managers. In each of these instances, the guilt or innocence of the people involved was judged not by the consequence of the act, nor by the intentions of the actor, but simply by the fact that their behavior was or was not conventional. In those examples where the actors' motives were selfish or cruel, and the results harmful to others, the actors were excused because they had behaved as most people do. In those examples where the actors' motives and effects were to enjoy themselves and harm no one the actors were censured because they had behaved as most people do not. The point the author wishes to argue is that there appears to be a pervasive tendency for perceivers to evaluate observed behavior on the basis of what they believe is the usual behavior in that particular situation. When the observed 3 behavior conforms to the conventional pattern, the actor is less likely to be perceived as responsible for that behavior than he would be were his behavior unlike that of most people in the same situation. The problem, then, is to demonstrate that perceivers who believe that observed behavior is unusual will tent to attribute greater responsibility to the actors for their actions, than will perceivers who believe the actors have behaved as most people do in that situation. Review of the Literature Research on the attribution of responsibility inci- cates that observers' perceptions are effected by a number of factors. Shaw (1967), and Sulzer and Burglass (1968) have presented evidence that the age of the perceiver is important in assigning responsibility, adults taking intentions more into account than children do. Nevertheless, Pepitone and Sherberg (1957), and Jones and Nisbett (1971) found that intentions are not nearly so important as be- havior in attributing responsibility, even among adults. Shaw and Sulzer (1964), and Shaw, Floyd and Gwin (1971), demonstrated that behavior which produces negative results will evoke attributions of greater responsibility to the actor than will behavior that results in positive outcomes. Walster (1966) found that the more severe a negative out- come the greater the degree of attributed responsibility, however, her results could not be replicated in two other studies of severity and responsibility, Shaver (1970), and L, Shaw and Skolnick (1971). A study by deCharms, Carpenter, and Kuperman (1965) revealed that greater responsibility is attributed to an actor for behavior which is intended to make a gain than for similar behavior intended to prevent a loss or to avoid punishment. Strickland (1958), and Kruglanski (1970) found that actors who performed tasks without supervision were accorded greater responsibility for their actions than were actors who worked under surveillance. The literature makes no mention, however, of the effect of unusual behavior on attributions of responsibility. To introduce this independent variable into the growing literature one must make a fresh start. A logical starting place seems to be a discussion of the nature of information about people which is available to perceivers. There are two major forms of information about persons: preinterpreted reports of behavior, and behavioral cues. Much of the information people receive about other people comes from some third party and is already evaluated. Gossip, report cards, letters of recommendation, and eulogies are common sources of ready-made attributions. The most common source of this kind of information is the mass media. People learn far more about other people, many of whom are remote from them in space, time, and social position, from the mass media than from any other source. The other form of information is behavioral cues from the actor, himself. Cues are only observable hints about the nature of the person. They must be cognitively assembled and organized 5 by the perceiver before he can make attributions about the actor. Both forms of information, the ready-made attri- bution from the mass media, and the do—it—yourself cues from the actor, present problems for the perceiver. A problem for perceivers (actually receivers), who must depend upon the mass media for information about current events, social problems, political movements, and so forth, is that very frequently the information about a particular individual or group is very sparce. The reporter's guideline is Who, What, When, and Where. "Why" is not included. Perhaps this is due to the difficulty of getting such information, or to the time pressures of dead— lines, or perhaps the media do not have the money and personnel necessary to pursue that question. But more likely, it is not necessary to explain why a particular person behaved in a manner so singular as to attract the notice of the media. Perhaps the editors assume that, given their audience's knowledge of what is usual behavior, people can well decide for themselves why it is that "man bites dog." The author would like to suggest that in the case of information from the mass media all a perceiver needs to know in order to make attributions about an actor is: (a) what did he do, and (b) what would most people do? Inter- estingly, in the case of making attributions on the basis of cues the problem is different, but the solution is similar. 6 The problem for the perceiver is not that cues from which to make attributions are scarce, but rather that they are too abundant. A perceiver is inundated by a flood of stimuli from both the focal person and the environment. Anything an actor does or does not do is a potential cue. His every choice from among the courses of action open to him, the statements he makes, the words he chooses, the tone of his voice, the look in his eyes, and even his posture, provide cues about his character and the inner forces which motivate him. The perceiver's knowledge of the situation provides another torrent of cues. The social situation: cultural values and beliefs, norms and social control mech- anisms, and societal goals are especially important cues about the causes of behavior. Even the physical setting in which the behavior takes place: the location, the time of day, down to the temperature and humidity may share the responsibility for a person's behavior. The problem for the perceiver, then, is to process the information efficiently so that he avoids overloading his information handling capacity. If he is unable to do this quickly, and with a minimum of effort, he will be swamped by the incoming stimuli and will be unable to make predictions about others' behavior and the effects of his behavior on them. Heider (1958) claims that the ability to make reliable predictions about the outcomes of interaction is critical to one's capacity to create a stable, controlled environment. If a person is to survive and benefit from his 7 day-to—day confrontations with other human beings, he must be able to anticipate correctly, at least most of the time, just how they will act and react toward him. That is, he must develop an adequate "theory" of the causes of people's behavior, a "theory" that may be applied parsimoniously to the cues provided to him by actors and situations. Heider suggests two such "theoriesP He points out that men are perceived as causal agents, a fact that makes them unique in the world as we understand it. Unlike other causes of natural phenomena which are themselves, seen as being caused, man is seen as the first cause at least of the moral quality of his acts. Thus, one theory about the cause of an individual's behavior is that there is something about the character of the individual, himself, that is respon- sible. The second theory, based on the unstated assumption that man, too, is a natural phenomena holds that man's behavior is frequently caused by impersonal situations, external to him, and located in the social or physical world. But which "theory" to apply to any given observation? What is needed is a principle by which to select the most reliable and parsimonious explanatiOn for each social inter- action. Although they do not consider the person versus situation causal attribution problem, Jones and Davis (1965), in their classic paper on person perception, point out that when a person's behavior conforms to the model of standard behavior, the perceiver learns very little about the unique 8 characteristics of that person. In other words, to learn that a man chooses to act in the conventional manner is to learn only that he is like most other men in that regard. This observation, taken by itself, is not especially dramatic, but it can be developed into the basis for the principle which we are seeking. Although knowing that a person behaves like others may be inadequate if the perceiver's purpose is to discover the other's character traits, this information may be a sufficient basis upon which to predict his future behavior. If the perceiver can assume that the other's behavior will conform with conventional behavior, then he does not need to know about the other's unique characteristic to predict his behavior, he needs only to know what behavior is con- ventional. Returning to Heider, we are told that the perceiver seeks to find a sufficient reason why a particular person acted in a particular way. Once he discovers a sufficient reason the search stops (Heider, 1958). If we incorporate this into the Jones and Davis argument, we may now assert that if the perceiver "discovers" that the actor behaved in a particular manner because society values that specific kind of behavior, that is, in that particular situation most people behave in a conventional manner, the search process stops and the perceiver attributes the cause of the behavior to the societal values. Should the perceiver decide that 9 the actor has not acted as most other men would in that situation, then the search must continue and he will con- sider the actor's unique characteristics as a causal locus. Here, then, is a principle by which perceivers may select between the personal and impersonal explanations of the cause of behavior. To identify this principle we will borrow Stain's (1971) label for the earlier concept from Jones and Davis and call it the "priority principle." The priority principle may be stated as follows: Perceivers consider societal values as a possible explanation of ob- served behavior before they consider personal factors, and if the societal values are perceived as a sufficient explan- ation, personal factors receive little or no consideration. However, when societal values fail to provide a sufficient explanation, the perceiver's attention is then directed to personal factors. The perceiver is almost certain to "discover" some personality trait to account for the behavior, his search being guided by the folk wisdom that "they wouldn't act that way if there wasn't something wrong with them." A few more words are required on this point. As Mills has pointed out, we ask about people's motives only in those situations which involve alternative or unexpected purposes or conduct (Mills, 1940). The author would like to extend this to point out that questions about an actor's motives, or questions in general about the possible cause of his behavior are probably never raised 10 unless his behavior is unusual. Ordinary behavior is taken for granted and questions are asked about the reason for it only rarely. More than likely, most people are not aware of the causes of ordinary behavior, and are probably unaware that it is caused. It is not until something happens, such as an experimenter who insists that they answer questions about the cause of the particular socially accepted pattern of behavior, do they become aware that they have never con- sidered the reason for that behavior, but have always taken it for granted. The Putneys call this the "yellow-eyed cat" phenomenon (1966, p. 7). When people are forced to consider the cause of common behavior, as in an experiment such as this one, it seems reasonable that they would consider situ- ational causes first, as the priority principle states, and it seems unlikely that they would attribute the behavior to the idiosyncracies of the actors. Out of the ordinary behavior, however, invites attributions about the actor's personal characteristics. Unusual behavior cannot be explained by referring to society's values, and, indeed, the first cue that the actor is different from other people is the fact that he acts unusually. Obviously, this is circular reasoning, but, nevertheless, probably a very common cognitive pattern. The important implication of such a phenomenon, if it does exist, is that because of the way in which perceivers process cues (and evaluate information) there exists a tendency to blame individual idiosyncracies or perhaps individuals' 11 psychological emotional problems for deviant behavior and social problems and to stop the search for the cause of anti-social behavior before considering that the real cause may lie with societal conditions in which these people find themselves. Frequently, people who behave unusually simply are labelled as lazy, violent, brain-washed, or just plain crazy, and no attempt is made to go beyond these labels to seek the actual sources of deviant behavior. 0n the other hand, and perhaps just as frequently, perceivers tend ggt to make personal causal attributions with regard to social problems if the behavior is regarded as being usual and this, also, is a form of negligence. This model of person perception assumes that most people operate on a kind of naive sociology. They are aware of a societal standard of behavior they uncritically accept as the correct way of behaving. They may be unaware of sub- group behavior patterns which conflict with the behavior patterns of the larger society, or perhaps, they simply refuse to recognize as legitimate other group's standards of behavior. Of course, there are many exceptions to this "Archie Bunker" approach to the world. It appears, however, that this is more frequently the case than not--unfortunate1y. One final point remains to be established. As stated above, when the observed behavior is unusual, it is likely that the perceiver will infer personal characteristics as a way of accounting for the other's actions. These personal characteristics may vary in the degree to which 12 they correspond with the behavior to be explained. WQQEEQ— spondence refers to the extent that the act and the under— lying characteristic or attribute are similarly described by the inference" (Jones and Davis, p. 223). The greater the correspondence, the greater the degree to which the inferred characteristics are perceived to account for the observed behavior. To put it the other way around, the correspondence of an inferred characteristic declines to the degree that the action observed appears to be common. For example, if a person behaves in a dominant fashion in a situation where dominant behavior is common, say the person is an Army platoon leader giving orders to his men, a perceiver would most likely not search for an explanation for his behavior. If the perceiver were asked to rate him on a scale from dominant to submissive, in terms of his personal characteristics, the perceiver would probably rate him as "somewhat dominant." He would be unlikely to rate him as "extremely domineering." What is more, it would be likely that the perceiver would not place great confi- dence in this inference. In such a case, the inference would have low correspondence with the observed behavior. On the other hand, if the actor had behaved in a domineering fashion in a situation where dominance is not common, say in a conversation with a peer, an observer would probably infer that this behavior is a reflection of a disposition to dominate others. In this case, the perceiver would probably rate the actor as "extremely dominant," and would be likely 13 to place high confidence in this attribution. In this latter case, the inference would have high correspondence with the observed behavior. An experiment by Jones, Davis, and Gergen (1961) demonstrates that behavior which conforms to clearly defined role requirements (usual behavior) is seen as uninformative about the individual's personal characteristics, whereas a considerable amount of information will be extracted from unusual, out-of-role behavior. The subjects listened to one of four tapes recorded "job interviews" in which the inter- viewee was instructed to appear to be interested in quali— fying as a submariner or as an astronaut. The subjects were told that the ideal submariner is "other-directed," and that the ideal astronaut is "inner-directed." In the two conditions in which the prospective submariner responded as if he were "other—directed," and the prospective astro- naut responded as "inner-directed" (in-role—behavior) the subjects inferences of what the interviewee "was really like as a person," revealed that they perceived him as moderately conforming and moderately affiliative. The confidence the subjects place in each rating was extremely low. On the other hand, out—of-role behavior (submariner- inner, and astronaut-other) produced very extreme ratings on the two characteristics, and both inferences were made with high confidence. In other words, the unusual behavior produced highly correspondent inferences about the stimulus person's personal characteristics, but usual behavior produced only low correspondence. 14 The point then is that the attribution of responsi- bility for his behavior to the actor, and the attribution to him of unique character traits are not two separate attri- butions but rather only one. The explanation of the actor's behavior is in terms of his perceived characteristics. Therefore, to attribute particular traits to an actor is to explain the reason for his behavior. The degree to which the perceiver attributes extreme characteristics to the actor, and the confidence he places in those attributions are a measure of his certainty that the actor is indeed the cause of the observed behavior. Theoretical Development of the Hypothesis The theoretical development of the hypothesis may be summarized as follows: Heider states that the ability to predict other peoples' behavior is necessary to the success of social interaction. To predict behavior, one must know the cause of the behavior and Heider suggests two types of causes: personal causes and situational causes. It is assumed that if cause is the situational type, then people in that situation will behave as the situation demands, that is, they will behave similarly. Jones and Davis point out that if a person behaves in a particular situation as others do, a perceiver does not learn much of the actor's character, but he does not need to know much about the actor's character in order to predict his behavior, he needs only to know how others act in that 15 situation. Heider subgests that people look for "sufficient explanations" of behavior and once one is found, they look no further. These two ideas were combined to form the "priority principle," which claims that people first con- sider the situational causes of behavior, i.e., they ask if the behavior observed is conventional behavior and if it is the search for the cause is terminated because a "sufficient cause" has been located. If the behavior is not ordinary, the personal causes must be considered and the perceiver asks what is it about the actor's unique character that causes him to act in this manner. Some character trait (or traits) will be identified as responsible for the act because usual behavior elicits attributions of corresponding unusual traits. The greater the correspondence the greater the de- gree to which the inferred unusual trait is perceived to account for the observed unusual behavior. And, therefore, since the person is seen as the source of his unusual be- havior, it is hypothesized that he will be held responsible for that behavior to a greater degree than he would be if his behavior had been conventional. The author suggests that a major factor in the attri- bution of responsibility is the perceivers concept of what constitutes conventional behavior. The purpose of this dissertation is to demonstrate that the manipulation of information given to perceivers about what constitutes "usual" behavior, will effect their attributions of respon- sibility to the stimulus persons. In particular, it is 16 hypothesized that: Perceivers will attribute a greater degree of responsibility for the outcome of his behavior to an actor who has behaved unusually than they will attribute to an actor who has behaved in the conventional manner. To anticipate what is to be presented in the section on experimental design, four treatment groups are created by instructions about the unusualness or usualness of the behaviors of two confederates. In the data collection pro— ceedure, the subjects are asked to indicate, on a seven- point scale, the degree to which each of the confederates and the rules of the laboratory game they play are respon- sible for the game's outcome, i.e., the amount of money won or lost by each of the confederates. It is reasonable to assume that these three attributions will not be independent, and that if the amount of responsibility attributed to one of the confederates is higher in the condition in which the subjects are instructed that his behavior is unusual, the degree of responsibility attributed to the other confederate, and to the game, would be less than in the "usual condition. Therefore, it is hypothesized that: If greater responsibility is attributed to an actor because his behavior was unusual, the amount of responsibility attributed to the other actor, and to the game, will be less than if the first actor had behaved in the conventional manner. COLLECTION OF DATA Experimental Design In order to test the hypotheses, it was necessary to create a situation in which two essential conditions could be met. First, it had to be possible to manipulate the per- ceivers' beliefs about the usualness or unusualness of a specific behavior in a particular situation. The requisite circumstance was one in which one group of perceivers be— lieved the action observed was usual, while a second group of observers believed the identical behavior was unusual. Holding constant the behavior, while varying the "usualness" of that behavior, would permit the observation of the effect on the perceivers' attributions of responsibility to the actors. The second necessary condition was one in which interfering variables were eliminated or controlled. To be assured that the effects observed were in fact produced by the independent variable, it was essential to avoid the possibility that extraneous factors, such as potentially applicable cultural norms, religious, ethical or political values, or influence from others, might contaminate the results. These two requirements are best met by the method of laboratory experimentation. The laboratory provides an environment in which it is possible to regulate the 17 18 information about a specific event which is available to the perceivers. Specifically, by designing a laboratory game, about which the perceivers could have no previous knowledge of how people behaved when playing the game, it was pOSsible to supply them with a variety of information about what was "usual" and "unusual" behavior. Furthermore, it was hoped that the fact that the stimulus situation was a novel one would reduce the likeli— hood that the behavioral information supplied by the experi— menter would be superseded, by those standards of behavior which apply to events in day-to-day life, in influencing the subjects' perceptions. Sample The subjects for this experiment were volunteers from two sections of an undergraduate social psychology class, in the Spring of 1972. They were recruited with the aid of the Subject Recruitment Form [see APPENDIX E] which also served as the instrument by which the background data were collected. There were forty male subjects and forty female subjects. Because of scheduling difficulties, the number of males and females in each condition was not equal. The distribution of male and female subjects in each condition is shown in Table 1. Each subject is reported twice, once for Blue's behavior, and once for Red's. 19 TABLE 1 DISTRIBUTION OF SUBJECTS BY BEHAVIOR AND SEX Perceiver's Blue's Behavior Red's Behavior Sex Usual Unusual Usual Unusual Males 19 21 24 16 Females 21 ‘12 16_ 23 Totals 40 40 40 40 About seventy-five percent of the subjects were in the nineteen through twenty-one years old age group. The rest were over twenty-one, but only three were twenty—five or older. About fifty percent of the subjects were social science majors, fifteen percent were in the humanities and the rest from a wide variety of schools and programs through- out the University. The majority (seventy-three percent) were sophomores and juniors. Over eighty percent reported their social class to be upper-middle or middle-middle, and all were white with the exceptions of one Black, one Mexican American, and one American Indian. Three people were eliminated from the sample because they recognized one of the stimulus persons. No other sub- jects were eliminated from the sample for any reason. 20 Laboratory Setting The subjects, in groups of up to five persons each, were met by the experimenter in the waiting room of the Sociology Department's experimental laboratory. Here they were greeted by the experimenter, who identified himself and directed them into the laboratory. The laboratory is a medium-sized room of about 15 X 30 feet. The walls and ceiling are covered by white acoustical tiles, and the room is well lighted. A one—way mirror runs the length of one wall, but since this was not used it was covered by drawr drapes. At one end of the laboratory, flush against the wall and midway from either side of the room, was a small table upon which stood the game board. The game board was a A X A sheet of plywood, it was painted white, and held erect by a wooden stand. Along the left side of the board were mounted two blue light bulbs, one above the other. The left side of the board was labelled "BLUE." Across the top of the board were two red light bulbs, mounted side by side. The top of the board was labelled "RED." The four lights were controlled by switches placed on two smaller tables located about five feet in front of the larger one. One set of switches, for the "Blue" player, were labelled "Up'and "Down;" the second set, for "Red" were labelled "Left" and ‘ "Right." The game board had two hooks located under the two light bulbs at the top of the board. These hooks were for attaching the pay off structures to the game board. 21 The two pay off structures were four-cell matrices with numbers in each cell to indicate the amount each player would win if that cell were chosen. One matrix was labelled "Low-Risk Payoff Structure," the other was labelled "High-Risk Payoff Structure" [see APPENDIX A for Payoff Structures]. The payoff structures were printed on 3 X 3 feet pieces of white posterboard. The four light bulbs were located so that the blue lights indicated which one of the two rows (up or down) the "Blue" player selected and the red lights indicated "Redb" choice of the left or right column. Chairs were placed at each of the two smaller tables so that the confederates sat facing the game board. At the opposite end of the room, facing the game board, was the television camera used to make the video- tape of the two confederates who were the stimulus persons. Beside the camera was a special-effects generator which is a piece of television studio equipment that has two tele- vision screens in its console. These two pieces of equip- ment were not used during the experiment, but were left in the laboratory as props, to demonstrate that the video-tape was actually made in the room. Along the length of one side of the room were five desk-top chairs, each isolated from the one beside it by a rather large room-divider. It was in these chairs that the subjects sat as they watched the video-tape as it was played on a television set located on the opposite side of the room. The Ampax video-tape recorder was on the table 22 beside the television set so that the subjects could see the experimenter turn the tape on and off, and so that they could see the actual tape on the equipment as it played. The only other items in the room were a table, in the corner of the room just to the right of the game board, on which were stacked the questionnaires and other materials,and a chair for the experimenter to use while the subjects filled out the questionnaires. Cover Story After the subjects were taken into the laboratory, they were told the following cover story: "We are interested in the way in which people non- verbally project cues about their character traits to other people. We want to know more about how people reveal what kinds of persons they are with- out actually telling others verbally. We want to discover how much information about a person's character is revealed by a person's behavior. In other words, how much can we know about a person without talking to him. We are going to have you watch two subjects play a laboratory game and then have you report back to us what kinds of impressions of the subjects you picked up just by watching them. Last term, we administered a questionnaire to a large sociology class. The questionnaire consisted of a battery of questions designed to reveal how the sub- jects see themselves as persons. Using the results from this, we selected thirty subjects and enlisted their cooperation in this research. We asked each for a list of people who knew him very well, and who would be willing to tell us their impression of him. We then contacted these people by mail and asked them to fill out a questionnaire similar to the one each subject had filled out about himself. Of course, all of the replies are kept confidential. So now, at this point, we had information about how each subject, and several of his close friends, see him as a person. What we want to do now is see how 23 each subject is seen by complete strangers who have observed his behavior, but who have never talked with him. What we are going to have you do is watch a video- tape recording of two of these subjects as they perform a task in a laboratory setting. We will then ask you to fill out the same questionnaire that our other respondents have filled out. Are there any questions?" If there were, they were answered in a manner in keeping with the cover story. Stimulus Situation After they were told the cover story the subjects' attention was directed to the game board on the table at one end of the room. They were then given the following explanation of the game: "What you will see on the video-tape is two of the thirty subjects playing this game. This is a very simple laboratory game which is often used in exper- iments like this one. There are two players: "Blue" and "Red." Each player has two switches which operate his set of lights. Blue's switches are labelled "Up" which turns on that light right there, [the experi- menter demonstrated each operation as the explanation proceeded] and "Down," which lights that one. Red's switches are for "Left," . . . and "Right." Any questions so far? [If there were questions at any time, they were answered; if there were none the experimenter proceeded with the explanation.) Now, this part of the game board is called the "pay- off structure." There are four cells in the payoff structure, and these numbers inside each cell indicate how much money each player wins on each trial. There are ten trials altogether, and at the end of the ten trials, each players winnings are added up and he actually gets paid that amount of money. The amount that the "Blue" player wins is indicated by the number in the lower left-hand corner of each cell; the number in the upper right-hand corner indicates how much the "Red" player wins. Now then, which cell is selected on each trial depends upon which lights Blue and Red turn on with their 24 switches. They take turns going first; on the first trial Blue turns on his light first and then Red lights his; on the second trial, Red goes first and then Blue. Let's say that it's Blue's turn to go first and he chooses to light the "Up" bulb--1ike this. And, then Red decides to go to the 1eft--like this. The combin- ation of the two lights indicates which cell has been selected. In this case, "Up" and "Left" are lighted, so the upper left-hand cell is the cell that has been selected. This means that Blue would receive forty cents for this trial and so would Red. If they used these same two lights on each of the ten trials, at the end of the game, each player would be paid four dollars. [At this point, the experimenter used the various combinations of switches to indicate each of the remaining cells. With each cell, he asked one of the subjects to report which cell has been selected and how much money each player would receive. When he was satisfied that each subject understood the mechanics of the game, he asked if there are any questions, and proceeded to explain the low-risk, and high-risk payoff structures.] 0.K. so far, so good. Now then, there are actually two payoff structures. This one is the lowerisk payoff structure; the other one is underneath [the experimenter raised the lowerisk matrix slightly to expose the other matrix beneath it, then dropped it down again and went on to explain the lowerisk matrix.] It's called a lowerisk payoff structure because of the way the payoffs are arranged. You can see that the "Blue" player receives forty cents whenever he chooses "Up," regardless of whether Red chooses "Left" or "Right." If Blue were to choose "Down" he would get nothing, so we would expect him to choose "Up" every time. The same kind of thing is true for Red. When he lights his left bulb he gets forty cents, no matter which light Blue chooses. If Red lights his right light he gets nothing, so we would expect him to choose "Left" on every trial. Now, since we can be pretty sure that Blue will choose "Up," and that Red will choose "Left" on each of the ten trials, its a safe bet that each player will have won four dollars at the end of the game, and that is why this payoff structure is called "LoweRisk." Are there any ques- tions about the lowhrisk payoff structure? [The experimenter removed the low-risk matrix from the hooks, exposing the high-risk matrix. He set the lowerisk matrix aside and said:] 25 This is the high-risk payoff structure. It is some— what like the other one with two important exceptions. First, if the combination of lights chosen by the players indicates that the upper left-hand cell has been selected, each player receives sixty cents in— stead of only forty cents. This means that if this cell were selected on each of the ten trials, both players would be paid six dollars at the end of the game. So there is the possibility of winning more money with the high—risk payoff structure. But now look at this: on the high-risk payoff structure the "Red" player can go to the right and win eighty cents on each trial. If he chooses the right light on the lowerisk payoff structure he would get nothing, but if he chooses it on the high-risk payoff structure he gets eighty cents and the "Blue" player gets nothing regardless of whether he chooses "Up" or "Down." So there is this risk to the "Blue" player that Red might go to the right and win eight dollars on the game, if he goes to the right every trial, and then Blue would get zero. And, that is why this payoff structure is called a high-risk pay- off structure: because of the risk to Blue. Are there any questions about the high-risk payoff structure? Now, since Blue is the player who takes the risk on the high-risk payoff structure, we allow him to decide whether the game will be played with the low- risk or high-risk payoff structure. Are there any questions about that? Alright, now then, there is one more important thing for you to know about this game. [At this point in the explanation of the game, the experimenter supplied the treatment information about the "usual" behaviors for both Blue and Red. The treatment conditions are discussed in the next section of this chapter. Note that in the explanation of the game, no mention was made about whether the goal of the game was to compete or to cooperate. Does any one have any questions now? None? Very well, if you'll have a seat over here, we'll watch the tape. You'll see the whole explanation, again, on the tape, as the game is explained to the two subjects [this was done to in ure that the subjects understood the game's mechanics.i You'll notice that the game begins with a toss of a coin to determine which subject will be "Blue" and which one will be "Red." ‘ 26 As you watch the subjects play the game, we will have you keep score on this score sheet. [:See APPENDIX D for the Score Sheet. The score sheet was employed to be sure that the subjects knew which player did what and how much he had won. It also acted as a double- check to be sure that the subjects correctly under- stood the treatment instructions.] It is easy to keep score, because the subjects were told how much they won after each trial, so all you have to do is write that down in the proper column on the score sheet. This is just a record for you so that when you are filling out the questionnaire you'll knOW'WhO did what on each trial. There are two questions on the bottom of the score sheet that you won't be able to answer until after you've seen the tape. Since the dividers block your view of the game board, I'll give each of you a copy of the payoff structures. [This handout also included the instructions about "usual" behavior, according to the subjects' treatment condition. See APPENDIX C for Treatment Instruction Sheets.] All of the subjects regardless of their treatment condition, watched the same video-tape. Therefore, any differences in perceptions, between the four groups, may be assumed to be due to the treatment instructions received and not to the behavior observed. Also, because video-tape was employed, the observed behavior is perfectly standardized, an important advantage over laboratory experimental designs that use "live" confederates each time the experiment is rlm. The two confederates were white, male undergraduates. They were clean-shaven, had medium length hair, and were dressed in sports shirts and wash trousers. Because of the camera angle, one's face was never visible on the video- tape, and the other's was seen for only a second or two. Their dialogues consisted of an occasional one-word response 27 to the experimenter's questions. Aside from their pre- arranged choices during the game, they evidenced no distinguishing characteristics, except that they were "typical" college students to the point of being practically nondescript. The confederates' roles called for them to make two important decisions which were apparently critical to the outcome of the game. The first of these was Blue's choice of payoff matrix. Blue "chose" the high—risk payoff structure, and this "decision" presented Red with the option of cooperating with his partner by electing to light the left bulb, and, thereby winning six dollars each, or of exploiting his partner by choosing to go to the right on each trial, winning eight dollars for himself while leaving Blue with nothing. Red "chose" to go to the right on each of the ten trials. It was important to the design of the experiment that both players have a hand in the determination of the final outcome of the stimulus game. The burden of respon- sibility could fall on Blue for having chosen the high-risk payoff structure, when he might have played it safe and opted for the low-risk, or it might fall on Red for having chosen to go to the right on each trial instead of to the left. Then, too, it is likely that both may be seen as being equally responsible. Since the stimulus situation is a novel one for the perceivers, it is reasonable to assume that the treatment information about whether each player's 28 behavior was usual or unusual would have an impact on the subjects attributions of responsibility. After the subjects had watched the video-tape, the experimenter asked them to answer the two questions at the bottom of the score sheet. The questions asked whether each player's behavior had been usual or unusual. The exper- imenter checked each subject's answers right away to be sure that the treatment information had been understood. If any subject had questions about whether the behavior was usual or unusual, the experimenter reminded the subject of the instructions given about this, and referred him to his copy of the payoff structures which had the treatment instructions written out. The experimenter would then ask the subject to tell him whether the observed behaviors Were usual or unusual. When experimenter was satisfied that each subject understood and accepted the treatment instructions, the post-observation questionnaire was administered. As each subject completed his questionnaire, he was asked to fill out the report request form [see APPENDIX G] so that each subject would receive, by mail, a report on the experiment's purpose and findings. After data had been collected from eighty subjects, the experimenter discussed the experiment in class. It was felt that the discussion and individual reports would be an improvement over the more commonly used talk-down immediately following the experiment since the only deception was in the fact that the study was actually concerned with the effect of norms on 29 perceptions rather than with non-verbal_communication, and that the players observed were actually confederates. The experimenter felt that this in no way would be harmful to the subjects and that the explanation of the experiment would be a more meaningful learning experience if the re- sults of the experiment could be discussed along with its purpose. Each subject was paid two dollars for participating in the experiment, and the experimenter‘made a point of expressing his sincere appreciation, for his or her parti- cipation and cooperation, to each subject, individually, at the completion of each session. Treatment During the experimenter's explanation of the mechanics of the game, following the discussion of the lowa and high-risk payoff structures, he explained that the "Blue" subject was allowed to choose which payoff structure would be used, and then asked the subjects if they have any questions. If there were none, or after questions had been answered, he said: "Alright, now then, there is one more important thing for you to know about this game." At this point the experimenter supplied the treat— ment information; that is, he told the subjects what is the "usual" behavior for both players. The information for each of the four conditions is as follows: BO Condition One: Blue usual/Red usual. We know from previous experience that almost all of the subjects will choose the High-Risk Payoff Structure. In fact, more than 95% of the "Blue" subjects choose the High-Risk Payoff Structure. We also know that when the High-Risk Payoff Structure is chosen, more than 95% of the "Red" subjects will choose "Right" every trial so that the result of the game is that Red wins $8, and Blue gets $0. Condition Two: Blue usual/Red unusual. We know from previous experience that almost all of the subjects will choose the High-Risk Payoff Structure. In fact more than 95% of the "Blue" sub- jects choose the High-Risk Payoff Structure. We also know that when the High-Risk Payoff Structure is chosen, more than 95% of the "Red" subjects will co- operate with their partners and choose "Left" every trial so that the result of the game is that both players win $6. Condition Three: Blue unusual/Red usual. We know from previous experience that almost all of the subjects will choose the Low-Risk Payoff Structure. In fact, less than 5% of the "Blue" subjects choose the High-Risk Payoff Structure. We also know that if the High-Risk Payoff Structure is chosen, more than 95% of the "Red" subjects will choose "Right" every trial so that the result of the game is that Red wins $8, and Blue gets $0. Condition Four: Blue unusual/Red unusual. We know from previous experience that almost all of the subjects will choose the LowaRisk Payoff Structure. In fact, less than 5% of the "Blue" subjects choose the High-Risk Payoff Structure. But, we also know that when the High-Risk Payoff Structure is chosen, more than 95% of the "Red" subjects will cooperate with their partners and choose "Left" every trial so that the result of the game is that both players win $6. 31 Instrument The data was collected with an eighteen page questionnaire [see APPENDIX B for Post—Observation Questionnaire.] There were three parts to the question- naire: The first part of the questionnaire was concerned with the perceived character traits of Blue and Red. To examine the degree of extremity in attributions of character traits, two sets of fifteen seven-point bi- polar adjective rating scales were used. One set was for the "Blue" stimulus person and one set was for the "Red." The subject was asked to indicate what he thought each stimulus person was "really like as a person." The fifteen scale items comprise three clusters of five items each. The clusters were selected on the basis of their relevance to the observed behaviors. The first cluster deals with attributions of friendliness, i.e., it asks about the degree to which either stimulus person was perceived as displaying a disposition to make friendly overtures to others. The scale items in the friendliness cluster were: friendly-- unfriendly, trusting--suspicious, warm--cold, pleasant—- unpleasant, likeable--irritating. The second cluster deals with attributions of aggressiveness. This cluster is con- cerned with the degree to which either stimulus person was perceived as displaying a disposition to dominate others in a determined, calculating, and energetic pursuit of his own ends. The items in the aggressiveness cluster were: 32 competitive-—cooperative, smart--stupid, forceful--weak, dominant--submissive, stubborn--f1exible. The third, the altruism cluster dealt with the degree to which either stimulus person was perceived to display a regard for the interests of others. The items in this cluster were: con- siderate of others--selfcentered, honest-—dishonest, reliable-—untrustworthy, generous--not generous, humane-- ruthless. The measure of the subject's confidence in each of the above attributions was taken by a seven-point bipolar rating scale ranging from "Not Certain At All" to "Extremely Certain." The subjects were also asked to report their general impression of each stimulus person on a seven-point scale margin from "Very Bad Impression" to "Very Good Impression." They were also asked to write a brief statement describing what they felt each subject "was really like as a person." Questions 5 through 6a made up the section on responsibility. The first question in this section (0-5) asked "Who do you feel is responsible for the final score in this experiment?" This was a closed-ended question and the subject was required to select from "Blue," "Red," "both," or "neither." Questions 4, 5, and 6 asked "To what degree was the "Blue" subject (the "Red" subject, the rules of the game) responsible for the final score?" These three 33 questions required the subjects to indicate the degree of perceived responsibility on seven-point scales ranging from "Not At All Responsible" to "Totally Responsible." Each of these questions was followed by an open-ended question which asked "Why do you say that?" The third section of the questionnaire dealt with attributions of the degree to which each stimulus person was unrestrained in his behavior. Three seven-point scales ranging from "Not Free At All" to "Completely Free" were provided for responses as to Blue's freedom in choosing the payoff structure, his freedom in choosing "Up" or "Down," and Red's freedom in choosing "Right" or "Left." Each of the three scales were followed by an open-ended question which asked: "Why do you say that?" Method of Analysis The procedure employed to analyze the data from the scales was to compute means on each item for each condition (usual or unusual) for each of the two stimulus persons (Blue and Red). These means were compared by the use of a one-tailed t ratio. The one-tailed test is appropriate since the hypothesis predicted that responsibility scores would be greater in the "unusual" conditions. The data from non-scale items were analyzed by means of contingency tables. The obtained frequency distributions were compared with the expected frequency distributions by using Pearson's chi-square statistic. 54 Each analysis was done twice: once using the entire sample as a whole, and then, again, controlling on the sex of the subject to reveal possible differences in the responses of males and females. ANALYSIS AND INTERPRETATION Unusual Behavior and Responsibility The principle hypothesis tested was that unusual behavior will elicit more extreme attributions of respon- sibility than will usual behavior. The independent variables were the instructions given to the subjects about whether Blue's and Red's behaviors were conventional or unconventional, and the dependent variables were the attri- butions of responsibility. The data for the dependent variables were collected by three questions: "To what degree was the Blue subject responsible for the final score?; To what degree was the Red subject responsible for the final score?; and, To what degree were the rules of the game responsible for the final score?" Since the hypotheses clearly indicated the partic- ular effects expected, a "planned-comparison" research design is the appropriate analysis procedure to employ. However, it is frequently useful to begin analysis using a "post-hoc" design, the data for general patterns of relationships between variables. This method is presented below using three independent variables. Analyses of Variance The first two variables (A and B) are the instruc- tions about the confederates' behaviors. These were 35 36 explained in the section on treatment. The third variable (C) is the sex of the perceiver. The legitimacy of, and the necessity for, including the sex of the perceiver as one of the three variables becomes obvious in the analysis which follows. Table 2 shows the summary of the three-way analysis of variance of the data from the question about the degree of Blue's responsibility for the final outcome of the game. TABLE 2 THREE-WAY ANALYSIS OF DEGREE OF BLUE'S RESPONSIBILITY Source SS df MS Fa Blue usual/unusual (A) 5.55 1 5.55 1.99 Red usual/unusual (B) 5.51 1 5.51 1.27 Perceiver's Sex (C) .57 1 .57 .15 A x B 5.24 1 5.24 1.89 A X C 4.09 1 4.09 1.48 B X C .00 1 .00 .00 A x B x C .19 1 .19 .07 Error Term 199.65 72 2.77 aF ratio required for p <.05 = 5.98 Since all of the three-way analyses of variance were done with unequal cell frequencies, they were somewhat more conservative tests than they would be otherwise. As may be seen in Table 2, none of the F ratios, for the first analysis approached the level of significance, and it appears that neither of the instructions about the confed- erates' behaviors nor the sex of the respondent made a significant difference in the degree of responsibility 37 attributed to the Blue confederate. Furthermore interaction effects are not in evidence. Although they were not re— vealed in this analysis, there were effects on the degree of responsibility attributed to Blue which are revealed by another analysis techinque which provides a more appropriate test of the hypotheses. These effects are demonstrated in the section on planned comparisons. TABLE 5 THREE-WAY ANALYSIS OF DEGREE OF RED'S RESPONSIBILITY Source SS df MS F Blue usual/unusual (A) .00 1 .00 .00 Red usual/unusual (B) 7.62 1 7.62 5.91b Perceiver's Sex (C) 4.02 1 4.02 2.07 A x C 11.76 1 11.76 6.04a B x C 5.50 1 5.50 1.69 A x B x C 1.26 1 1.26. .65 Error Term 140.18 72 1.95 a p < .05 b p < .10 In Table 5, it may be seen that the instructions given about Red's behavior (B) had some influence on the degree of responsibility attributed to him. However, the effect is significant only at the .10 level of probability. Of greater interest is the significant effect caused by the interaction between the instructions given the subjects about Blue's behavior and the instructions about Red's (A x B). Apparently, both instructions taken together have 58 a major effect on the amount of responsibility attributed to Red. But still more interesting is the finding that the largest effect on the degree to which Red was perceived to be responsible was caused by the interaction between the instructions given about Blue's behavior and the sex of the perceiver (A x C). These two findings are explained in detail in the section on planned comparisons. TABLE 4 THREE-WAY ANALYSIS OF DEGREE OF GAME'S RESPONSIBILITY Source SS df MS F Blue usual/unusual (A) 7.88 1 7.88 2.60 Red usual/unusual (B) .49 1 .49 .16 Perceiver's Sex (C) 6.82 1 6.82 2.25 A x B .04 1 .04 .01 A x C 2.94 1 12.94 .97 B x C 16.26 1 16.26 5.36a A X B X C 12.05 1 12.05 3.983 Error Term 218.51 72 5.05 5p <.05 In Table 4, two more effects are revealed. The first of these was due to the interaction between the instructions given about Red's behavior, and the sex of the perceiver (B x C). These two variables operating together have a rather considerable effect on the degree to which the rules of the game were perceived to be responsible for the players' winnings. The second significant effect is due to the interaction between all three variables (A x B x C). 39 Tables 5 and 4 suggest that the next step in the analysis is to present a two-way analysis of variance for the degree of Red's responsibility, and for the degree of the game's responsibility, this time controlling on sex rather than using it as one of the variables. Since Table 2 indicated no significant effects on the degree of Blue's responsibility that could be attributed to the two treat- ment variables, or to sex differences, the degree of responsibility attributed to Blue will be omitted from con- sideration until the section on planned comparisons. Tables 5 and 6 are for male subjects only, and Tables 7 and 8 are for female subjects only. TABLE 5 TWO-WAY ANALYSIS OF DEGREE OF RED'S RESPONSIBILITY--MALES ONLY Source SS df MS Fa Blue usual/unusual (A) 5.89 1 5.89 2.18 Red usual/unusual (B) .45 1 .45 .17 A x B 8.93 1 8.95 5.31 Error Term 97.14 56 2.70 aF ratio required for p<<.05 = 4.11, for p <.1O = 2.86. Table 5 indicates that for male subjects, neither the instructions that Blue's behavior was usual or unusual, the instructions that Red's behavior was usual or unusual, nor interaction between the two sets of instructions had a significant effect on the degree of responsibility attri- buted to Red. 40 TABLE 6 TWO-WAY ANALYSIS OF DEGREE OF GAME'S RESPONSIBILITY--MALES ONLY Source SS df MS Fa Blue usual/unusual (A) 10.26 1 10.26 2.80 Red usual/unusual (B) 5.57 1 5.57 1.52 A x B 6.75 1 6.75 1.84 Error Term 151.98 56 5.67 aF ratio required for p <.05 = 4.11, for p14 TOTAL I M P O R T A N T After recording the scores and adding the total, answer the two questions below. "BLUE'S" choice of Payoff Structure was (circle appropriate response) 1. USUAL--like most other subjects. 2. UNUSUAL--not like most other subjects. "RED'S" choice of "Right" or "Left" was (circle appropriate response) 1. USUAL--like most other subjects. UNUSUAL--not like most other subjects. 88 ‘TL m’ ‘. APPENDIX E RECRUITING FORM Sociology 551 Spring, 1972 SOCIAL PSYCHOLOGY LABORATORY EXPERIMENT If you wish to participate in this experiment, please check the appropriate boxes below to indicate the times when you will be available. Please check a number of different times so that we will be able to fit you in. The experiment last 50 minutes. PLEASE CHECK THE TIMES WHEN YOU ARE AVAILABLE: Mon. Tues. Weds. Thurs. Fri. Sat. 10:20—11:10 11:50-12:20 5:20-4:10 4:50-5:20 7:40-8:50 8:50-9:40 Name (Please Print) Address PHONE Student Number 89 90 In order to make descriptive comments about our sample we need to know a few simple facts about our respondents' backgrounds. PLEASE CIRCLE 1. Your sex: 5, Your year: Male. ° ° -1 Freshman. . . . .1 Female. ° '2 Sophomore . . . .2 Junior. . . . . .5 2. Your age: Senior. . . . . . .4 Other . . . . . . .5 17. O O O .1 . 18, , , , ,2 6. In whlch social class 19. , , , ,3 would you say you and 20. . . , ,4 your family belong: 21. O O O .5 22 o o o o o 6 Upper Class 0 O O 25. . . . .7 Upper middle. . . 24. . . , .8 Middle-middle . . Lower-middle. . 5. Your major: Working class . Lower class . . GUI-L‘UJNA 7. Are you: 4. Marital status: White . . . . . Single. . .1 Black . . . . . Married . .2 Mexican-American Divorced. .5 American Indian Japanese. . . . Chinese . Filipino. Hawaiian. Other . . \OCIJQONUI-DKNN-A APPENDIX F TIME SCHEDULE FOR SUBJECTS Time Date: Subjects Phone Condition Time Subjects Phone Condition Time Subjects Phone Condition 91 I? APPENDIX G REPORT REQUEST If you would like a written report concerning this research, please fill out the information below. NAME (PLEASE PRINT) ADDRESS (Where you can be reached in July) 92 ‘9».- APPENDIX H RECEIPT FORM I acknowledge receipt of two dollars ($2) for partici- pation in the Fischer-Ewens Sociology experiment. Signed Date __\ 0 Ha: OKOCDNONU‘I-L-‘KNN A ._\ .2. o ._\. N 13. 14. 15. 16. 17. 93