A STUDY OF THE RELATIONSHIP BETWEEN AUTHORITARIANISM AND RESIDENT ASSISTANT EFFECTIVENESS Thesis for the Degree of Ed. D. MICHIGAN STATE UNIVERSITY NORMAN E. HEFKE 1968 J' 8‘ mm IIIIIIIIIIIIII IIIIIIII a __.._ _ Unified; This is to certify that the thesis entitled A STUDY OF THE RELATIONSHIP BETWEEN AUTHORITARIANISM AND RESIDENT ASSISTANT EFFECTIVENESS presented by Norman E . Hefke has been accepted towards fulfillment of the requirements for W degree inEanfiIiQfl. érmz/Apamt/m Major professor Date M4168.— 0-169 ~., I. 5,: I ABSTRACT A STUDY OF THE RELATIONSHIP BETWEEN AUTHORITARIANISM AND RESIDENT ASSISTANT EFFECTIVENESS By Norman E. Hefke In this study an attempt was made to determine if authoritarianism is negatively related to the effectiveness of resident assistants employed in college and university residence halls. If a relationship does exist between these two variables, then the addition of an authoritarianism scale to the selection process could improve the selection of resident assistants. Several secondary hypotheses involv- ing variations in sex and raters were also tested. Design: During the winter term of 1968 the Rokeach Dogmatism Scale, Form E was administered at group meetings to 158 male and 170 female undergraduate resident assistants employed at Michigan State University. High and low authoritarian groups were then selected by taking the 25 highest and lowest scores for each sex. Effectiveness ratings were then obtained from two different groups of raters. One group was comprised of the graduate-level advisers who worked in the resident assistant's hall. The second group consisted of 100 samples of six residents selected at random from each resident assistant's living unit. The Resident Adviser Evaluation Form.was used by both groups to assess the resident assistants' effectiveness. Norman E. Hefke Usable: responses were obtained from 570 or 95 per cent of the resident sample. The 98 graduate-level advisers returned 260 ratings. Approximately 86 per cent of the graduate-level advisers responded. Conclusions: As a result of statistical comparisons of the data by four-way analysis of variance, it was found that: 1. There is no difference in effectiveness between male and female resident assistants. L///§. There is no difference in effectiveness between low and high authoritarian resident assistants. 3. There are no interaction effects between the sex of the resident assistants and scores on the Dogmatism Scale. 4. There is a difference between ratings by residents and by graduate-level advisers of resident assistants' effectiveness. 5. There are no interaction effects between the sex of the resident assistants and the position of the rater. 6. There are no interaction effects between resident assistants' scores on the Dogmatism Scale and positions of the raters. 7. There are no interaction effects between the sex of the resident assistants, their scores on the Dogmatism Scale, and positions of the raters. Only one of the hypotheses was found to be significant. Head resident advisers, assistant resident advisers, and graduate resident advisers rated the resident assistants in their hall significantly higher than did residents. Norman E. Hefke Discussion The data collected in the present study do not show that there is a relationship between authoritarianism as measured by the Dogmatism Scale and resident assistant effectiveness as measured by the Resident Adviser Evaluation Form. Failure to obtain significant differences between the high and low authoritarian groups may be due to the fact that the resident assistant population as a whole is extremely non- authoritarian. T test comparisons between the mean scores obtained by resident assistants and a random sample of Michigan State University seniors on the Dogmatism Scale indicated that the resident assistants' scores were significantly lower. The t test values, which were -12.24 for males and -12.90 for females, were significant at the .001 level. The Dogmatism Scale scores obtained by the resident assistants indicate that the subjects in the "high authoritarian" group were actually only "middle authoritarian." It seems reasonable to assume, therefore, that the subjects who were included in the "high authori- tarian" resident assistant group are no more authoritarian in their behavior than are college students in general. Since these resident assistants as a group obtained Dogmatism Scale scores which are only slightly higher than those which would be obtained by the average college senior, it may be assumed that behavior which would be directly affected by this personality variable would vary little between the two groups. The behavior of this particular group of ”high authoritarian" resident assistants is probably different than the behavior of indi- ! viduals who are highly authoritarian. It is, therefore, unlikely that Norman E. Hefke the level of authoritarianism which exists among this group of resident assistants would negatively affect their performance. A STUDY OF THE RELATIONSHIP BETWEEN AUTHORITARIANISM AND RESIDENT ASSISTANT EFFECTIVENESS By Norman E. Hefke A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF EDUCATION Department of Administration and Higher Education 1968 ACKNOWLEDGEMENTS The author of this study wishes to express his appreciation to the members of his guidance committee, Dr. Walter F. Johnson, Dr. John A. Fuzak, Dr. Orden C. Smucker, and especially to the chairman, Dr. Eldon R. Nonnamaker, for the help and encouragement which they gave during the course of the investigation. An expression of appreciation is also extended to Dr. Donald V. Adams and the members of the residence halls staff at Michigan State University for their assistance and understanding while this study was being conducted. Thanks are also extended to Dr. James P. Duncan for permission,to use the Resident Adviser Evaluation Form, to Dr. Milton Rokeach for the use of the Rokeach Dogmatism.Scale, and to Dr. Patrick B. Smith for his assistance and encouragement during the investigation. Finally, the author owes special debts of gratitude to several members of his family. These include his mother and his oldest brother, Harold, for sacrifices which they made so that the writer could pursue a higher education. In addition, the author wishes to express his deep appreciation to his wife, Suzanne, for her encourage- ment, understanding, and for her many hours of assistance during the investigation and writing of the thesis. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . Chapter I. II. III. THE PROBLEM. . . . . . . . . . . Theoretical Basis of the Study . . . . Objectives of the Study . . . . . . Definition of Terms . . . . . . . . . Limitations of the Study Overview of the Study . . . . . REVIEW OF RELATED LITERATURE . . . . . . . . Literature Pertaining to Selection of Resident Assistants . . . . . . . . Literature Pertaining to Resident Assistant Effectiveness . . . . Summary . . . . . . . . . . . INSTRUMENTS AND PROCEDURES USED IN THE S TUDY O O O O O C O O O I O O O O O 0 Selection of the Resident Assistant Population 0 O O O C C O O 'O C C O O 0 Method of Identifying "Low Authoritarian" and ”High Authoritarian" Resident Assistants . . . . . . . . . iii Page vi 10 ll 11 15 22 24 24 25 Chapter Page Method of Evaluating Resident Assistants . . . . . . . . . . . . . . 29 Selection of the Raters . . . . . . . . . 32 Gathering of the Data . . . . . . . . . . 34 Analysis of the Data . . . . . . . . . . 36 Obtaining Estimates of Reliability . . . 38 Comparison of Resident Assistants With seniors O O O O O O O O O O O O O O O 39 Comparison of Resident Assistants From Michigan State University and Indiana University . . . . . . . . . . . . . . 40 Summary . . . . . . . . . . . . . . . . . 40 IV. ANALYSIS OF THE RESULTS . . . . . . . . . . 42 Comparisons of the Ratings . . . . . . . 43 Reliability of the Instruments Used in the StUdy O O O I I O O O O O O O O O 47 Comparison of Resident Assistants' and Seniors' Responses to the D-Scale . . 48 Comparison of Resident Assistants From Michigan State University and Indiana University . . . . . . . . . . . . . . 49 V. SUMMARY, CONCLUSIONS, AND DISCUSSION . . . . 52 Design . . . . . . . . . . . . . . . . . 52 Description of the Instruments . . . . . 53 Reliability of the Instruments . . . . . 54 Research Hypotheses . . . . . . . . . . . 54 Conclusions . . . . . . . . . . . . . . . 54 iv Chapter Page Discussion . . . . . . . . . . . . . . . 55 Recommendations for Further Research . . 59 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 62 APPENDIX A Rokeach Dogmatism Scale . . . . . . . . . 67 Resident Adviser Evaluation Form . . . . 71 APPENDIX B Letter to Head Resident Advisers From the Director of Residence Hall Programs 0 O O O O O O 0 O O O O O O O 78 Letter to Residents . . . . . . . . . . . 79 Letter to Head Resident Advisers . . . . 80 Letter to Assistant Resident Advisers and Graduate Resident Advisers . . . . 81 LIST OF TABLES Table Page 1. Means and Standard Deviations Obtained by Resident Assistants on the D-Scale . . . . 28 2. Mean and Standard Deviation Values for Female and Male Resident Assistants According to Ratings on the Resident Adviser Evaluation Form by Residents and Graduate-Level Advisers . . . . . . . 43 3. Analysis of Variance Table. Responses by Residents and Graduate-Level Advisers to the Resident Adviser Evaluation Form for Female and Male Low and High Authoritarian Resident Assistant Groups . . . . . . . . 44 4. Means, Standard Deviations, and t Test Values for Resident Assistants' and Seniors' Scores on the D-Scale . . . . . . 48 5. Effectiveness Ratings for Resident Assistants at Michigan State University and Indiana University Utilizing the Resident Adviser Evaluation Form . . . . . . . . . . . . . 50 vi CHAPTER I THE PROBLEM Colleges and universities throughout the country employ advisory personnel in campus residences to assist students in their pursuit of learning and personal development. These people live in the residences, .offering guidance to the residents who live there in such areas as .interpersonal relations, study skills, financial aid, and social and emotional problems. These personnel are often under the direction, especially in the larger residences, of a full-time staff member. (31:7) This full—time person, usually called a head resident adviser, is sometimes assisted by one or two graduate students who are employed on a half-time basis. In addition to these graduate-level employees, advisory staffs in most residence halls include a number of undergraduate students. These student assistants, commonly referred to as RAs, are considered part- time employees, and live with a number of residents in a subdivision or corridor of the hall. They are usually in their junior or senior year of college, and carry a full academic course load. The resident assistant position usually includes a wide variety of responsibilities which are designed primarily to promote the general welfare of students in the residence. (l,5,20,23,29,30) Among the more important of these are such functions as orienting students to college life, maintaining an environment conducive to the achievement of academic goals, and aiding students in solving personal problems. A Many resident assistants complete their responsibilities very successfully. They have good rapport with their residents, they are available to help them at their time of need, and they enhance learning by maintaining an atmOSphere conducive to study in the residence. They are effective in integrating such important qualities as understanding, tolerance, permissiveness, and acceptance so that their advisory role I is maximized. Some, however, are not so successful in fulfilling these responsi- bilities. The causes of their ineffectiveness are varied and numerous. Some resident assistants, overly concerned with maintaining order, find themselves engaged primarily in punitive, restrictive, managerial type relationships with their residents. Others, failing to integrate with proper balance all the variables which contribute to effective job performance, find themselves being ignored and made the object of practical jokes. They spend much of their time investigating improper behavior and writing conduct reports. Still others simply shirk their responsibilities by staying out of the hall whenever possible, by keeping the door to their room closed so not to be available, or by simply ignoring their residents' behavior. Serious attempts are made by supervisory personnel in many colleges and universities to select those students who are most likely to be effective resident assistants. (3,8,35,40) Unfortunately, however, selection processes are usually limited to global subjective judgments about the applicant's leadership qualities (3,8,12,40) and his ability to get along well with people. Considering the possibilities of errors involved in selection decisions based upon subjective criteria, the resident selection process could be greatly improved by supplementing present selection criteria with criteria which are quantitative in nature. ‘ Psychometric devices which validly and reliably measure factors found to be related to resident assistant effectiveness could provide this kind of data. The present study was undertaken to test certain quantitative criteria which, when added to the subjective criteria being used at present, could possibly improve the selection of effec- tive resident assistants. Theoretical Basis for the Study A review of the literature indicates that studies which are reported have contributed little to the solution of this problem. Simons (39), comparing high and low male resident assistant performance groups on four personality appraisal instruments, found only three of twenty-eight subscales which differentiated significantly between the two groups. Using the Minnesota Multiphasic Personality Inventory (MMPI), the Guilford-Zimmerman Temperament Survey (G-ZTS), the Allport- Vernon Study of Values (A-VSV), and the Index of Adjustment and Values (IAV), Simons found that the resident assistants who were rated as high performers by their head resident advisers and graduate resident advisers scored significantly above low performers on the Emotional Stability scale of the G-ZTS and on the Religious value scale on the A-VSV. The low performance group, however, exceeded the high group on the Theoretical value scale of the A-VSV. No significant differences were found on the remaining twenty-five subscales. (39:123) In studies of high and low performance groups where other instruments were uti- lized, no differences of any real significance could be identified in either group. (17,25,27) Hoyt and Davidson (9), however, found that authoritarian attitudes, 1 were negatively associated with as measured by the California F Scale, head counselors' (head advisers') ratings of resident assistant effec- tiveness. While the reliability of the rating scale used by Hoyt and Davidson is seriously questionable, the results of the study indicate that a measure of authoritarianism warrants further investigation in the search to find additional criteria on which resident assistant selection decisions can be based. Analysis of the authoritarian personality lends further support to this consideration. Authoritarian personalities have been described as "distrustful and suspicious" (36:40), lacking insight into the emotional and social attitudes of others, rigid, and as ". . .conforming, stereo- typical in their thinking, and intolerant of ambiguity." (34:181) Rokeach has described an authoritarian person as one who uncritically accepts authority and rejects or is intolerent of those who disagree with him. (32:5, 32:72) 1The California F Scale is a measure of prejudice and underlying personality predispositions toward a fascistic outlook on life. High F Scale scorers are found to be generally ethnocentric, anti-semetic, and politically conservative. (32:17) Rokeach and others have pointed out quite conclusively that, because it primarily measures "right" authoritarianism (those who are bigots on the political right), the F Scale should not be used as a measure of general authoritarianism. (32:1-70) In contrast, nonauthoritarian people have been described as a type whose evaluations are more objective and who are ". . .flexible. . .and have greater capacity for intense interpersonal relationships." (34:181) In general, a nonauthoritarian person may be described as one who is tolerant, permissive, open to criticism and suggestion, and is considerate of the concerns of others. These descriptions suggest that resident assistants with authori- tarian tendencies would have difficulty in establishing and maintaining positive relationships with the members of their living units. More specifically, they suggest that the high authoritarian resident assistant is not the kind of person that a resident would seek out for guidance or discussion of his problems. As an example, a resident who perceives his resident assistant to be distrustful and suspicious is unlikely to look to him as one who can help him to think through his ‘gyocational, educational, and personal problems. Because he gives the student the impression that he is incapable of being insightful, under- standing, and accepting, it is unlikely that he would be sought out for help. Where such perceptions exist, the resident assistant is not fulfilling his maximum potential in promoting the overall development of his residents. In addition, it seems unlikely that a person who is intolerant, insensitive, and inflexible would be capable of dealing adequately with the needs and problems of individual residents with which a resident assistant is typically confronted. In short, a resident assistant who is unable to develop positive, constructive, helpingrelationships with his residents and who is incapable of dealing with a variety of problems and personalities is less effective than he is generally expected to be. Objectives of the Study In view of the previous discussions about the problems of select- ing effective resident assistants and the characteristics of the authoritarian personality, the objectives of this investigation were to gather empirical evidence supporting the hypothesis that there is‘a negative relationship between authoritarianism and resident assistant effectiveness. More specifically, an attempt was made to determine if authoritarianism as measured by the Rokeach Dogmatism Scale2 is nega- tively related to resident assistant effectiveness as measured by the Resident Adviser Evaluation Form.3 The first three hypotheses stated below were developed to test this relationship. Since male residents may react differently than female residents to resident assistants who tend to be authoritarian or closed in their mode of thought and belief, comparisons of effectiveness based on residents' ratings between the high and low authoritarian groups were made separately for men and women. 2The Rokeach Dogmatism Scale measures open and closed mindedness, general authoritarianism, and general intolerance. Authoritarian personalities (persons who have closed belief systems) are associated with high scores on this scale. It is described in further detail in Chapter III of this study. 3The Resident Adviser Evaluation Form measures the resident assistant's success in establishing rapport with the students in his living unit. It was obtained from James P. Duncan of Oregon State University who constructed it as a doctoral dissertation study at Indiana University in 1965. Resident assistant effectiveness is associated with high scores on this instrument. It is described in further detail in Chapter III of this study. Hypothesis No. 1: Sub-hypothesis a. Sub-hypothesis b. For the purpose of this study, exploration questions are described Research indicates differ significantly in High authoritarian resident assistants will receive lower scores on the Resident Adviser Evaluation Form than will low authoritarian resident assistants. High authoritarian female resident assistants will receive lower scores on the Resident Adviser Evaluation Form than will low authoritarian female resident assistants. High authoritarian male resident assistants will receive lower scores on the Resident Adviser Evaluation Form than will low authoritarian male resident assistants. making optimum use of all data collected in of additional questions was undertaken. These below. 4 that graduate-level advisers and residents their expectations for the resident assistant's role. (18) Research carried out by Ortenzi (25) indicates that their respective evaluations of resident assistant performance are also different. Further analysis may reveal how authoritarianism exhibited by resident assistants is related to performance evaluations by graduate- level advisers and residents. In pursuit of information relative to 4 For the purposes of this study, the term "graduate-level advisers" will be used in a collective sense to include head resident advisers, assistant resident advisers, and graduate resident advisers. these questions, hypothesis No. 2 was developed for testing. Responses relating to male and female resident assistants as well as high and low authoritarian groups were tested separately. Hypothesis No. 2: There will be differences in effectiveness of the resident assistant groups between ratings by graduate-level advisers and ratings by residents. Definition of Terms The terms given below are defined in the way that they were used for the purpdses of this study. Graduate-level Adviser: This term is used in this study in a collective sense to include head resident advisers, assistant resident advisers, and graduate resident advisers who were employed as Residence Hall Programs Office staff members at Michigan State University during the winter term of 1968. Resident Assistant: This term refers to a male or female under- graduate student who was employed as a Residence Hall Programs Office staff member at Michigan State University during the winter term of 1968. Resident: This term refers to an undergraduate student at Mich- igan State University who lives in a residence hall living unit where a resident assistant is employed. Resident Assistant Effectiveness: This term refers to the results of ratings of resident assistants which were made by residents and graduate-level resident advisers using the Resident Adviser Evaluation Form. The resident assistant's obtained score represents the respon- dent's assessment of how many of a given number of behavior statements that particular resident assistant exhibits. Only those behavior statements which were found by Duncan (7) to discriminate between effective and ineffective resident assistants are counted in obtaining the resident assistant's score. The Resident Adviser Evaluation Form is described in greater detail in Chapter III of this study. Authoritarianism: For the purposes of this study, authoritarianism is measured by the Rokeach Dogmatism Scale. Persons who receive high scores on this scale are considered to be "dogmatic" or "closed" in their modes of thought and belief. Persons who receive low scores are considered to be "open minded" or tolerant, flexible, and non- stereotypical in their thinking. The D-Scale is described in greater detail in Chapter III of this study. Limitations of the Study The following limitations were apparent in this investigation: 1. The use of instruments which have less than perfect reliability and on which norms have not yet been well established presented a built-in limitation of this study. Although the instru- ments were among the best available for their particular purposes, the above weakness must be considered when studying and using the results. 10 2. Since this investigation is delimited to resident assistants employed at Michigan State University, precautions should be taken in generalizing the results to resident assistants at other institutions. Overview of the Study This study is designed to investigate the relationship of authori- tarianism to resident assistant effectiveness. Chapter I provides a description of the problem and the theoretical basis for the hypotheses which were tested. Chapter II contains a review of the literature pertaining to the selection and evaluation of residence hall under- graduate student assistants. Chapter III presents the instrumentation and the procedures used in collecting and analyzing the data used in the study. The fourth chapter contains the analysis of the data. The summary, conclusions, discussion of the results, and implications for further research are stated in the final chapter. CHAPTER II REVIEW OF RELATED LITERATURE Although a great many colleges and universities use undergraduate student assistants as staff members in their residence halls, little information is available in the literature concerning the procedures which are followed in the selection process or the characteristics which seem to be related to effectiveness or ineffectiveness in these positions. This chapter is devoted to a review of literature pertain- ing tprthe selection and evaluation of residence hall undergraduate student assistants. The review will be presented in two parts. The first section will include a review of articles and studies pertaining to resident assistant selection. This section will be followed by a review of studies pertaining to the relation between various personal characteristics and effectiveness in the resident assistant position. Literature Pertaining to Selection of Resident Assistants Williamson (42:396-397) has emphasized the need to staff residence halls with student assistants who have high academic and intellectual qualifications. He suggested that residence halls cannot perform an educational function until they are staffed by individuals who are 1For the purposes of this review, where the terms "resident counselor or "student counselor" were used in the literature in reference to undergraduate student assistants, the term "resident assistant" was substituted. 11 12 committed to the advancement of education and who are skilled in the techniques of learning. He stated that effective student residence staff members must be academically oriented and must believe in the academic mission of the institution. They should also be committed and reasonably skilled in using informal and casual relationships in the lounge, the library, the dining hall and student rooms to feed intel- lectual content into what often may be rather superficial and meaning- less conversations. In a survey of the methods which were used by 60 selected colleges and universities in the selection of male resident assistants, Enos (8:39) found that each of the 60 institutions conducted interview sessions with the applicants and that these sessions were considered to be the most important of the selection techniques reported. Enos also attempted to identify the qualities which were desired in resident assistants at these institutions. He found that interest and concern for others, leadership ability, and maturity were considered to be the most important personal qualities for resident assistants to possess. Other favorable characteristics frequently mentioned include the ability to work effectively with others, common sense, a pleasing personality, high scholarship, loyalty, integrity, and the ability to gain respect. (8:40) Regarding the qualities that are desirable in resident assistants, Sifferd has stated that this person should have . . .a genuine interest in the problems of others. Without such an interest, many of the (resident assistant's) fine qualities may never be put to use for the residents. The (resident assistant) will be more effective if he also possesses a quality of friendliness and geniality. People will turn more quickly for advice and help to a (resident assistant) who is approachable and who makes confiding easy. (38:58) l3 Sifferd also identified several other qualities which he felt that the resident assistant should possess. These include neat and attrac- tive appearance, social ease and poise, emotional stability, a cooperative attitude, alertness and energy, and experience in various kinds of extracurricular and collegiate activities. (38:40) Brady (3) suggested the use of free discussion as a technique in the selection of resident assistants. He developed a selection pro- cedure in which applicants were involved in group discussions covering situations with which resident assistants are frequently confronted. Brady utilized a selection committee to observe the groups' discussions and to evaluate each applicant. Selection of individuals for resident assistant positions were based on the committee's evaluation of the information contained in the individual's personal folder and his performance in the group discussions. Sheeder (35), in an evaluation of a resident assistant screening program developed for the purpose of overcoming the weaknesses of interview procedures, found that role playing as a method of selecting residence hall staff members received favorable responses from applicants and staff members. The role playing procedure involved having the experienced resident assistants play the role of students while the applicants assumed the role of resident assistants. Each applicant played three different roles: I) conducting a floor meeting with a group of residents; 2) talking to a resident about his grades; and 3) handling a disciplinary situation. The applicant was judged on the basis of poise, clarity of expression, approach, directions given, ability to command respect, appropriateness of response, and general impression. In addition to these role playing situations, the applicant 14 was interviewed by a student leader and by the head resident, his assistant, and an experienced resident assistant. At the conclusion of the sessions, candidates and staff members were asked to submit written critiques of the role playing procedure. All applicants and staff members involved responded positively to this screening process. Sheeder stated that: The very positive reaction which was experienced at Ohio University to the use of role playing as a (resident assistant) selection technique indicates that such a system may warrant the careful consideration of other college and university housing personnel. (35:158) Kidd (12:360), in an attempt to improve the selection of resident assistants, asked 525 male residents of a residence hall at Michigan State University to list students whom they would prefer as resident assistants. Each resident was asked to name someone whom he would like as his resident assistant, and to explain why. An identical procedure was followed in identifying those residents who were least preferred and the reasons why they were least preferred. Kidd concluded as a result of this study that residents prefer resident assistants who are l) friendly, cooperative, and pleasant; 2) responsible; 3) mature and respected; 4) intellectual and capable; 5) considerate; 6) moral; and 7) quiet. Those students whom they would not like as their resi- dent assistants were described as 1) unfriendly; 2) carefree; 3) immature; 4) incapable; 5) inconsiderate; 6) immoral; 7) loud; 8) conceited; 9) deviant; and 10) unstable. (12:361) Kidd credited this method of selecting resident assistants with having high reliability and validity. He based this conclusion on the fact that the 16 resident assistants who were employed at the time of the study were among the highest 20 scorers in the group. Kidd's 15 conclusion is open to considerable question, however, in light of the fact that the resident assistants had definite advantages over other participants in the study. For one, the resident assistants were older students (the residents consisted primarily of freshmen) who might be more mature and respected for that reason alone. In addition, resident assistants might be known better and by more students than most of the other residents due to the responsibilities, authority, and status associated with the resident assistant position. The author gave no consideration of these factors in the report of this selection procedure. Literature Pertaining to Resident Assistant Effectiveness Ortenzi (25) attempted to determine the relationship between the rated effectiveness of resident assistants at The Pennsylvania State University and selected personality characteristics. Two standardized instruments, the Strong Vocational Interest Blank and the Edwards Personal Preference Schedule, were administered to 93 resident assistants. Later, these assistants were evaluated by their students and their immediate supervisors. The evaluation forms used in this study were developed by the investigator for use in the study. On the basis of the combined evaluations, the subjects were divided into three groups-~successful, middle and less successful. The middle group was not utilized in the analysis of the data. Ortenzi found that a moderate correlation existed between the students' evaluations of their resident assistant and the evaluation by the resident assistant's immediate supervisor. The author stated that "Such correlations suggest that students perceive an effective resident l6 assistant model in a somewhat different way than does the immediate supervisor." (25:69) Furthermore, Ortenzi concluded that: The Strong Vocational Interest Blank was found to be of little value as a predictor of success as a resident (assistant). . . Though the analysis of the Edwards Personal Preference Schedule met with moderately more success than the Strong Vocational Interest Blank, it also appeared to have little value as a predictor of performance as a resident (assistant). . . (25:69) Peterson (27) dealt with the evaluation of performance and pre- diction of success of 89 resident assistants employed in the freshman residence hall program at the University of Kansas between 1951 and 1956. The resident assistants were evaluated by their staff super- visors and divided into three groups--most successful, middle, and least successful. The study was designed to investigate the relation- ship between success as a resident assistant and the results of the Kuder Preference Record-Vocational, the Guilford Zimmerman Temperament Survey, the Minnesota Multiphasic Personality Inventory, the Strong Vocational Interest Blank for Women, and other personal data. This study did not produce any results which would be useful in the selection of resident assistants. In a similar study Simons (39) explored the relationship between resident assistant job performance and selected personality character- istics. Sixty-eight resident assistants from eight men's residence halls at Michigan State University formed the population for the study. High and low performance groups were established by means of a rating form consisting of 100 items judged by head resident advisers and graduate resident advisers to be descriptive of successful resident l7 assistants. Head resident advisers and graduate resident advisers rated the job performance of the resident assistants with whom they worked. The performance groups were compared on individual scales included in the Minnesota Multiphasic Personality Inventory (MMPI), the Guilford- Zimmerman Temperament Survey (G-ZTS), the Allport-Vernon Study of Values (A-VSV), and the Index of Adjustment and Values (IAV). The following is a summary of the findings when-high and low job performance groups were compared on the personality characteristics included in these instruments: 1. No significant differences were found between the high and low performance groups on any of the MMPI scales. 2. The high group significantly exceeded the low group on only the Emotional Stability scale of the G-ZTS. 3. The A-VSV showed the high performance group to be significantly higher on the Religious values scale, but the low group was signifi- cantly higher on the Theoretical values scale. 4. There was no evidence of a significant relationship between the high and low groups and the classifications obtained from the IAV. In the final phase of his study, Simons compared the personality characteristic of the 68 resident assistants with those of selected general male undergraduate college student groups. A comparison on the results of the MMPI revealed that the resident assistant group was significantly higher on the Hy, Mf, and Pa scales, and significantly lower on the Hs, Pt, Sc, and Ma scales. In the comparison with the G-ZTS, the resident assistant group scored significantly higher on the R, A, S, E, O, F, T, and P scales. Finally, a comparison of the 18 results of the Index of Adjustment and Values indicated that the resident assistant group contained a disproportionate number of people who (I) accept themselves but who believe that other people in their peer group are not as accepting of themselves, and (2) accept them- selves and who believe that other people in their peer group are equally or more accepting of themselves. Simons concluded as a result of his research that the MMPI and the G-ZTS showed the most promise as instruments for utilization in the resident assistant selection process. It is the view of this writer, however, that the results of Simons' study offer little evidence to support this conclusion. I'" cerning the evaluation of resident adviser effectiveness. The study was based on a master's thesis by the junior author and was carried out at the University of Iowa during the 1962-63 academic year. The sample consisted of the 32 graduate and undergraduate resident advisers employed in the men's residence hall system at the University of Iowa. Ratings on overall effectiveness were obtained for each of the resident assistants by the use of a short rating form. The seven items on the form involved leadership, loyalty and cooperation, order and discipline, dealing with individual students, identifying problems and taking action, progress and development, and desirability for rehiring. The authors found that there was close agreement between head resident advisers' and house managers' ratings of the 32 resident advisers included in the study. A correlation of .81 was obtained between rank orders of the overall ratings of effectiveness by the two groups of raters. 19 To determine the degree to which the resident advisers in this study characteristically played one or more of four roles, a sample of 446 of their residents who were selected at random responded to an inventory having 37 items. The authors found significant relationships- between effectiveness and tendencies to play the "interest” role or theltf "competence" role. No significant relationships were found between y \ effectiveness and a tendency to play the ”Buddy” role or the "Authori- h / tarian" role. .x/’ -r" .. A third aspect of Hoyt and Davidson's study involved a test of the relationship between authoritarianism and effectiveness. For this purpose, effective and ineffective resident advisers (upper and lower halves of the total group of resident advisers based upon head resident adviser ratings) were compared on the basis of their scores on the California F Scale. High F Scale scores were found to be negatively associated with effectiveness. Hoyt and Davidson also attempted to determine if effective and ineffective advisers could be differentiated on the basis of their opinions about residence hall advisers and advising. They concluded as a result of their study that an instrument could be constructed that could predict the degree of adviser effectiveness. They suggested that items that reflect "activity oriented" and "power oriented" concepts2 should be eSpecially effective as discriminators, and that item content H might well be concentrated on . . .realistic aspects of residence hall living." (9:256) These concepts are defined by Osgood, Suci, and Tannenbaum in reference No. 26. 20 Madson (17) studied the relationship between resident assistant effectiveness and the following factors: scores on the Ohio State University Psychological Test; scores on the College Student Question- naire, Part Two; accumulated grade point; and length of time served as a resident assistant. Subjects for the study were 145 male under- graduate resident assistants employed at Ohio University in the fall of 1965. "Most effective" and "least effective" resident assistant groups were selected according to ratings by head resident advisers and graduate assistants. The items on which the resident assistants were rated dealt with five categories of job responsibilities. Significant differences between the two groups of resident assistants were found in only one set of comparisons. These differ- ences were found on eight items included in the College Student Questionnaires, Part Two. Madson suggested as a result of his study that the following description of resident assistants who were rated as "most effective" as opposed to resident assistants who were rated as "least effective" could be supported: 1. The "most effective” resident assistants are more definite about their major field and vocational decisions. 2. The "most effective" resident assistants perceive their college teachers and student personnel deans in a more favorable manner. 3. The "most effective" resident assistants are more satisfied 5 with their academic standing in their major field. f 4. The "most effective" resident assistants own more books. \\ 5. The ”most effective" resident assistants are more concerned I; with achieving a sense of identity. 21 The studies which have been reported thus far show little or no evidence to support the theory that specific personality character- istics influence the performance of resident assistants. Close analysis of the studies, however, reveal several very plausable explanations for the failure of these studies to make contributions to knowledge in this area. One very likely cause of failure in past investigations is due to homogeneity within the resident assistant populations which were studied. Due to self-selection and screening by personnel responsible for the selection of resident assistants, the personalities of effec- tive and ineffective resident assistants are alike in many ways. The data collected in Simons' study provide a good example of the narrowness of score distributions of the groups taken as a whole on various personality factors. Unfortunately, existing psychological tests are ineffective in identifying personality factors which are associated with effectiveness or ineffectiveness when the personalities of all subjects under consideration are so similar. A second failure of research in this area is due to weaknesses in the instruments which have been used. Performance rating scales which are developed for the purposes of a particular study by the person who is carrying out the research are often extremely short and are, there- fore, weak with respect to reliability. The performance rating instrument used in the study which was reported by Hoyt and Davidson presents a good example of this weakness. An additional cause for failure in past studies is due to major weaknesses in regard to underlying theory and design. All of them seem to lack sound theory tying Specific personality variables to high 22 or low performance. An increase in emphasis in regard to theoretical support for the selection of the instruments to be used in a particular study should result in the exclusion of those instruments such as the MMPI which are inappropriate due to basic assumptions and theories on which they are based.3 Finally, it seems that, in general, the approach which has been taken in past studies has been too global. Consideration of more than two or three personality characteristics in any one study allows too little control of the variables under consideration. In addition, concentration on fewer variables would not only allow more careful collection of the data, but would also provide more time for the pur- pose of careful analysis. Summary A review of the literature resulted in the identification of a small number of articles pertaining to the qualities which are thought to be desirable in undergraduate resident assistants. The qualities frequently mentioned include leadership ability, interest in helping fellow students, and the ability to establish rapport with students. The assessment of qualities such as these has been done primarily through personal interviews. The questionable use of the MMPI in discriminating between individuals within a normal population is suggested by Lingoes (15:318) who, in reviewing the MMPI, states that "For assessing personalities within the normal range of adjustment, however, it (the MMPI) will be found wanting" and ". . .as a general test of personality, the MMPI has a number of weaknesses. . ." Lingoes supported the use of the MMPI for the purpose for which it was developed--for discriminating between normal and abnormal personalities. 23 Five studies were reviewed in which effective and ineffective resident assistants were compared on various factors. The purpose behind each of these comparisons was to identify personality variables on which these groups differed so that these variables could be measured and used as criteria for selection. Personality assessment instruments such as the MMPI, the G-ZTS, and the SVIB have been utilized in attempts to improve the resident assistant selection process. At the present time, however, largely due to weaknesses in the theory and design of past research, there seems to be no significant evidence to support the use of quantitative criteria in the selection of resident assistants. In Chapter III a description of the instruments and the procedures used in collecting the data for the present study are presented. CHAPTER III INSTRUMENTS AND PROCEDURES USED IN THE STUDY This chapter presents a description of the instruments and procedures used in collecting data for the present study. First, the method of selecting the resident assistants for the study is presented. Second, a description of the instrument used to obtain ratings of the resident assistants is presented. Third, the selection of the raters and the method of obtaining their responses is described. Fourth, descriptions of how the data were analyzed and how estimates or reli- ability were obtained are presented. Finally, descriptions are presented of how data which were obtained in the present study were selected and compared with data obtained in other studies. Selection of the Resident Assistant Population In the early stages of the study it was decided by the writer that the population of resident assistants employed at Michigan State University was large enough to provide a fairly adequate number of subjects for the purposes of the study. In addition, it was felt that the use of a local population would greatly facilitate the collection of all data needed for the study. For these reasons all of the sub- jects included in the study were students who were employed as resident assistants at Michigan State University during'the second term of the 24 25 1967-68 academic year. The names of these students were obtained from a roster which was distributed at the beginning of the second term by the Division of Residence Hall Programs at Michigan State University. The total number of resident assistants at that time was 357. One hundred seventy-five of these resident assistants were male, and 182 were female. All of the subjects were undergraduates, and most were in their junior or senior year of college. Each of these resident assistants lived on a corridor in a residence hall where he was responsible for working in a "helping" capacity with between 45 and 60 residents. Method of Identifying ”Low Authoritarian" and ”High Authoritarian" Resident Assistants One of the first tasks in carrying out this study was the selection of a method of identifying the low authoritarian and the high authori- tarian subjects among the members of the resident assistant population. After careful consideration of the most well known measures of authori- tarianism, the California F Scale and the Rokeach Dogmatism Scale (D-Scale),1 the latter was selected because it was considered by the writer to be the best measure of "general" authoritarianism. The California F Scale was rejected because it is considered to be a measure of political conservatism, ethnocentrism, and anti-semitism which are ”right" authoritarian dispositions only and which therefore exclude other forms of authoritarianism. (33:11-15) 1A copy of the D-Scale is included in Appendix A. 26 The D—Scale consists of 40 statements with which subjects are asked to agree or disagree along a 6-point scale which has a range of -3 to +3. The 0 point is excluded in order to force responses toward disagreement or agreement. This scale is subsequently converted for scoring purposes to a scale having a range of l to 7, where -3 responses are given a value of 1 and +3 responses are given a value of 7.2 The total score is derived by summing the responses to the 40 items included in the instrument. Respondents who receive high scores are considered dogmatic and unreceptive to new ideas; those who score low are considered flexible, adaptive, and receptive to new ideas. (33) Reliability coefficients have been calculated on the D-Scale based on responses by college students of both sexes. Odd-even and test- retest reliability values based upon eight groups of students ranged from .68 to .93. Lehmann, using D-Scale data obtained from 2,746 freshmen at Michigan State University in 1958, obtained a reliability coefficient of .76. (14:22) The method of "known groups” (high and low dogmatic persons as judged by peers) was used by Rokeach in establish- ing validity for the Scale. Normative data obtained by Rokeach from 288 students at Ohio State University and Michigan State University resulted in mean values ranging from 141.8 to 143.8, and standard deviation values ranging from 22.1 to 28.1 (33:90) Lehmann found a mean value of 165.98 and a standard deviation value of 25.70 from responses by 2,746 freshmen in 1958. (14:27) 2The scoring procedures which were followed in this study were based upon suggestions made by Dr. Irvin J. Lehmann of the Office of Evaluation Services at Michigan State University. 27 The next step in selecting the low and high authoritarian subjects was the administration of the D-Scale to the resident assistants. On February 1, 1968, a letter was sent by the Director of the Division of Residence Hall Programs to all head resident advisers employed at that time at Michigan State University.3 The letter introduced the writer and the study, described the design of the study, emphasized its signi- ficance, and asked for staff support and assistance. Prior to this time, the investigator had personally described and discussed this study with the Director of the Division of Residence Hall Programs. During the following week the head resident advisers were con- tacted by telephone for the purpose of asking for their support of and participation in the study. Arrangements were made with each of the head resident advisers for the administration of the D-Scale to the members of their resident assistant staff. Folders were then prepared for each head resident adviser. Each folder included the correct number of D-Scales for the resident assistants employed in a particular residence hall. Separate answer sheets were not necessary because the response scales were provided at the left margin on the D-Scale form. Each D-Scale contained a number corresponding to a number which had been previously assigned to that particular resident assistant. The system of using numbers instead of names was utilized for the purpose of maintaining privacy as much as was possible. The folders were delivered to each of the head resident advisers according to the arrangements which had been made by telephone. 3A copy of the letter to head resident advisers is included in Appendix B. 28 The D-Scales were then administered by the head resident advisers to the resident assistants at their next regularly scheduled staff meeting. Of the 357 resident assistants employed in the residence halls, 328 responses were obtained. One hundred fifty-eight of these reSpon- dents were male, and 170 were female. Fifteen of the nonrespondents preferred not to be included in the study. The remainder were either unable to be contacted or responded erroneously to the D-Scale. The responses to the D-Scale were scored according to procedures described previously in this chapter. The responses were scored by hand and punched by hand onto computer data cards. Assistance was then gained from a Control Data 3600 Computer in obtaining mean and standard deviation values. The mean of the D-Scale scores for the total group of 328 resident assistants was 125.40. The standard deviation of the total group was 21.53. Mean and standard deviation values for the male resident assistants were 128.43 and 23.31 respectively. The female resident assistant group received a mean value of 122.58 and a standard deviation value of 19.39. Table 1 shows the mean and standard deviation values for resident assistants based on their scores on the D-Scale. Table 1: Means and Standard Deviations Obtained by Resident Assistants on the D-Scale Class Number Mean Standard Deviation Males 158 128.43 23.31 Females 170 122.58 19.39 Total 328 125.40 21.53 29 The next step in the study was the selection of the resident assistants who would be identified as either low authoritarian or high authoritarian. It was the opinion of the writer that consideration should be given to the requirements of the statistics which were to be used in analyzing the data. The design of the study allowed four-way analysis of variance, and it was felt by the writer that this kind of statistical treatment would provide the most thorough examination of the data. Therefore, because it is most appropriate to have N3 of 'equal size when comparing groups by analysis of variance, it was decided by the writer after consultation with a consultant in the Office of Research Consultation that the size of the low and high authoritarian groups for both sexes should be equal.4 It was also decided at this time that groups consisting of 25 subjects would result in representative samples of ratings of the high and low D-Scale scores. Subjects were then selected for the low and high authoritarian groups. The male resident assistants who received the lowest 25 scores were selected for the male low authoritarian group. The male resident assistants who received the highest 25 scores were selected for the male high authoritarian group. Identical procedures were followed in selecting the female low and high authoritarian groups. Method of Evaluating Resident Assistants The next step in the study required the use of an additional instrument. This instrument was needed for the purpose of rating the 4This decision was made during consultation with Dr. Andrew Porter of the Office of Research Consultation, College of Education, Michigan State University. 30 effectiveness of the resident assistants who were included in the low and high authoritarian groups. An attempt was made to identify a rating form.which had reasonably high reliability and which also applied well to the resident assistant positions as they are defined at Michigan State University. In addition, it was hoped that an instrument could be found which could be utilized by both residents and graduate-level advisers. A review of the literature resulted in the identification of several different rating forms which had been used in rating resident assistants by residents (17,37,25) and graduate—leve1.advisers. (17,27, 25) One of the studies which was reviewed included separate ratings by head resident advisers and hall managers. (9) From still another study, it was found that students agreed with resident directors' and graduate assistants' ratings of resident assistants' effectiveness. (17:91) Double ratings were desired in the present study for the purpose of studying how the position of the rater affects the scores obtained by male, female, low, and high authoritarian resident assistants. It was felt that, due to frequent contact in the residence hall, subjects from both groups should know the resident assistant who lives in his residence hall well enough to make an evaluation of him. The Resident Adviser Evaluation Form.was selected because it seemed to be best in terms of the criteria established for selection.5 This instrument was constructed by Dr. James P. Duncan at Indiana University in 1965. It was utilized in Duncan's study by residents who 5A copy of the Resident Adviser Evaluation Form is included in Appendix A. 31 rated graduate student resident advisers. In essence the instrument measures the resident assistant's success in establishing rapport with the students in his living unit. The forced-choice technique is utilized in this instrument to avoid weaknesses such as "judgments of " "rater bias," and "halo effect" which are found in many leniency, rating scales. (22) In developing the instrument, opinions from undergraduate students living in residence balls at Indiana University were utilized in identifying 416 favorable and unfavorable statements concerning the behavior and activities of resident assistants. A panel of 12 resi- dence hall staff supervisors then judged the relatedness of these statements to 10 categories of resident assistant job functions. As a result of this procedure the number of statements was reduced to 176. One hundred student judges were then asked to compare the items on a l-to-5 scale of favorability for the purpose of establishing the preference index of each item. A second group of 100 student judges who had lived in residence halls for at least two years were asked to evaluate the 176 items in terms of their applicability to the "best" (most effective) and ”worst" (least effective) resident assistant they had known. Those items which discriminated between effective and ineffective resident assistants at the one per cent level of signifi- cance were identified for inclusion in the final form of the instrument. These procedures resulted in the retention of 96 items for inclu- sion in the final form of the rating scale. These items are arranged in 32 triads or blocks of three statements. Each triad contains items with equal or comparable preference ratings. However, two of the three 32 statements possess a high discrimination index while the third state- ment in the triad has a low discrimination index. The rater is instructed to consider carefully the statements in each block and select the statement(s) that are most applicable to the resident assistant he is rating. He is instructed to pick at least one statement but no more than two statements in each block. The score for the subject being rated consists of the total of those statements selected which have a high discrimination index. These are identified by the answer key for the instrument. Effective resident assistants are associated with high scores. Reliability estimates on this instrument were obtained at Indiana University. A split-half reliability on the odd versus even items using the Pearson r and Spearman-Brown formulas resulted in reliability coefficients of .70 and .74. The mean and standard deviation values for the 1,058 ratings were 30.86 and 3.32 respectively. (6) Selection of the Raters The resident raters were selected according to the following procedures. First, the number of residents who were to rate each resident assistant had to be decided upon. It was the writer's opinion that a number which would represent at least ten per cent of the stu- dents in each resident assistant's living unit would probably provide a fairly good general impression of the effectiveness of the resident assistant. Since the number of residents in a particular living unit during the second term of the 1968 academic year was found to be approx- imately 50, it was felt that ratings by six of these residents would be sufficient. 33 Therefore, for each resident assistant who was in one of the low or high authoritarian groups, six residents were selected by random procedure from living unit rosters which were available at the recep- tion desk in each residence hall. The rosters include the students' room and phone numbers, and therefore these were also obtained at that time. Since the study included four groups with 25 resident assistants in each, a total sample of 600 resident raters was selected. The selection of the graduate-level adviser raters was accomplished by a very simple procedure. A list of all head resident advisers, assistant resident advisers and graduate resident advisers was obtained from the Residence Hall Programs Office. From this list all those who worked in a residence hall where at least one member of one of the low or high authoritarian resident assistant groups also worked were identi- fied and assigned a number from 1 to 98. The numbers were assigned for the purpose of identifying the graduate-level advisers' responses by number rather than by name. It was felt by the writer that using numbers rather than names on the rating response forms would allow for identification of non-respondents and at the same time maintain a certain degree of privacy. It was found that, in most residence halls, three graduate-level advisers were employed. Therefore most resident assistants selected for the study could receive ratings by three graduate-level advisers who worked with them in the residence halls. The total number of raters in the graduate-level adviser sample was 98. 34 Gathering of the Data Data for this study were collected according to two different sets of procedures. To obtain residents' evaluations, a combination of letters, phone calls, and group meetings were utilized. These proce- dures seemed necessary in order to obtain an adequate number of responses from.the resident sample. To obtain graduate-level advisers' evaluations, only letters were utilized. It was felt that an initial letter and a follow-up letter would probably be sufficient to induce responses from an adequately large number of the graduate-level advisers. The ratings of effectiveness by residents were obtained in the following manner. Residents who were selected for the study were asked by letter6 and by telephone to attend group meetings which were sched- uled in the residence hall in which the student lived. The meetings took place during the latter part of winter term and early part of Spring term of the 1967-1968 academic year. No meetings were scheduled for the period starting one week prior to winter term final exams through the period of registration for the spring term. Friends and colleagues assisted the writer by attending these meetings and adminis- tering the rating forms. At the group meetings the residents received several instructions. First, they were instructed to place on an IBM answer sheet the name of their resident assistant. They were then asked to read carefully the instructions on the Resident Adviser Evaluation Form. Assurance was given that the information being gathered was for research purposes 6Copies of the letters which were sent to residents are included in Appendix B. 35 only and that it would in no way affect them or their resident assis- tant. Ample time was allowed for questions and completion of the Evaluation Form. The residents were asked to check their names off a roster as they left the meeting so that those who did not attend could be identified. Members of the resident sample sometimes asked about the purpose of the resident assistant evaluation. They were told that the evalua- tion was part of a doctoral research project in which an association between a certain personality variable and resident assistant perfor- mance was being studied. The name of the variable was withheld for the purpose of minimizing possible reaction to it. Additional group meetings were scheduled until at least five responses were obtained for each resident assistant. Attempts to obtain responses from non-respondents were discontinued at the end of the third week of spring term. A total of 570 or 95 per cent of the resident sample responded to the Evaluation Form. The ratings of effectiveness by graduate-level advisers were obtained according to the following procedures. Head resident advisers, assistant resident advisers, and graduate-resident advisers in whose hall resident assistants were selected for the study were asked by mail on March 8, 1968, to rate the selected resident assistants on their staff by responding to the Resident Adviser Evaluation Form. Each member of this group received a packet which contained a letter, a Resident Adviser Evaluation Form, and an IBM answer sheet. The letter reminded them of the purpose and the significance of the study and 36 asked for additional assistance.7 The answer sheet was coded to identify respondents and at the same time provide a certain degree of privacy. The name of the resident assistant who was to be evaluated was also included on the answer sheet. A follow-up letter was sent early in April to inform non-respondents of the graduate-level adviser group that their responses had not been received. By the end of the third week of the spring term, a total of 84 or 85.7 per cent of the group of graduate-level advisers responded. At least one reSponse from a graduate-level adviser was returned for every resident assistant included in the study. The total number of graduate—level responses was 260. Analysis of the Data Responses to the Resident Adviser Evaluation Form were scored according to procedures which were suggested by its author.8 After completion of this step, arithmetic mean scores for the reSponses by residents and by graduate-level advisers were calculated separately for each resident assistant. The method of analyzing the data was then worked out with assis- tance from personnel in the Office of Research Consultation, College of Education, at Michigan State University. Following suggestions given in those discussions, a four-way analysis of variance program was 7Copies of the letters which were sent to graduate-level advisers are included in Appendix B. 8Procedures for scoring the Resident Adviser Evaluation Form are described on pages 31 and 32 of this report. 37 utilized to determine if differences between the groups of resident assistants and the raters were statistically significant. The data were treated by the program in such a way that F ratios were calculated between all groups for which the design of the study allowed comparison. A Control Data 3600 Computer was utilized in carrying out this analysis. A total of seven hypotheses was developed for the purpose of test- ing the relations between the three independent variables included in the study. The first three of these hypotheses deal with what are considered by Kerlinger (11:228) to be "main effects." Based on these hypotheses, F ratios are calculated for the purpose of testing sepa- rately the differences between the mean scores obtained by males and females, by low and high authoritarians, and also according to who rated these groups--residents or graduate-level advisers. Hypotheses four through seven deal with what may be considered to be "interaction effects." (11:219-228) These hypotheses form the basis for the calculation of F ratios which illustrate the effect of two or more of the independent variables on the dependent variable when the independent variables interact or "work together." The seven hypotheses are stated below. 1. H1: A difference in effectiveness will be found between male and female resident assistants. 2. H1: A difference in effectiveness will be found between high and low authoritarian resident assistants. 3. H1: There are interaction effects between the sex of the resident assistant and scores on the D-Scale. 4. H : A difference will be found between residents' and graduate- 1 level advisers' ratings of resident assistant effectiveness. 38 5. H1: There are interaction effects between the sex of the resident assistant and the position of the raters. 6. H1: There are interaction effects between the position of the rater and the resident assistant's score on the D-Scale. 7. H1: There are interaction effects between the sex of the resident assistant, his scores on the D-Scale, and the position of the raters. The probability of 0.05 was used to determine the region of signi- ficance for all of the statistical tests used in this study. Obtaining Estimates of Reliability For the purpose of obtaining estimates of the reliability of the instruments used in the study, split-half reliability coefficients using Pearson's product-moment coefficient of correlation procedure and the Spearman-Brown correction formula were obtained on the responses to the D-Scale and the Resident Adviser Evaluation Form. The odd and even items on all response sheets were scored separately and then punched onto computer data cards. Reliability values were then obtained from the odd and even scores by using the Pearson product-moment correlation formula. This procedure yielded correlation coefficients which were then corrected by using the Spearman-Brown prophecy formula. The prophecy formula merely adjusts the correlation coefficient to what the coefficient would be if the measuring instrument was of the original length. A Control Data 3600 Computer was utilized in computing the reliability coefficients. 39 Comparison of Resident Assistants with Seniors During the course of the investigation the writer raised the question of whether or not the resident assistant population involved in the study was less authoritarian than college students in general. Such a difference could be possible due to self selection by students who have an interest in helping students and who enjoy working with people on an individual and group basis. It also seemed possible that flexible, open-minded, tolerant applicants would be selected by resi- dent assistant selection committees because these characteristics may be seen as being necessary for success in the position. For this reason it was decided that a comparison should be made between the total group of resident assistants included in this study and a group of college students who were not resident assistants. An attempt was made to find available data on students who were somewhat similar to the resident assistants in regard to age and which were also differentiated according to sex. A sample of 590 men and 461 women who were seniors at Michigan State University responded to the D-Scale near the end of the 1961-62 academic year. (14:52) The results of this study of seniors were compared with the data obtained in the present study. The t statistic was utilized for this comparison. Since the direction of the difference between resident assistants and seniors was predicted in advance, one- tailed t tests of significance were employed. 40 Comparison of Resident Assistants from Michigan State University and Indiana University It was the opinion of the writer that consideration should be given to how the effectiveness ratings which were obtained in the present study compared with those obtained in previous studies. Such a comparison could indicate whether the resident assistants in the present study were rated comparatively high or low by both residents and graduate-level advisers. For this purpose it was decided that the ratings obtained at Indiana University where the Resident Adviser Evaluation Form.was developed would be presented in a table with the ratings obtained at Michigan State University. The data which were selected for this comparison were reported by James P. Duncan. The study was conducted at Indiana University in 1965. A copy of the complete report was obtained for this purpose from University Microfilms at Ann Arbor, Michigan. (7:53) The writer could find no reports of other studies in which the Resident Adviser Evalua- tion Form was utilized. Summar y The D-Scale was administered to 328 undergraduate resident assis- tants employed at Michigan State University. High and low authori- tarian groups for males and females were then selected. Ninety~eight graduate-level advisers and 600 residents selected at random from each resident assistant's corridor were then asked to rate the selected resident assistants with whom they were associated by reSponding to the Resident Adviser Evaluation Form. The total number of responses from 41 graduate-level advisers was 260. Residents completed 570 responses. Reliability coefficients were computed from responses to the two instruments included in the study. For the purposes of additional analysis, a comparison was made between resident assistants' and seniors' responses to the D-Scale. Finally, the data obtained from ratings of student assistants at Indiana University were selected for presentation with the data obtained in the present study. A four-way analysis of variance program was utilized in analyzing the ratings. The results of the analysis of the data are presented in Chapter IV. CHAPTER IV ANALYSIS OF THE RESULTS This chapter presents an analysis of the data collected in the study. First, an analysis of the data gathered from the responses to the Resident Adviser Evaluation Form is presented. Second, estimates of reliability for the D-Scale and the Resident Adviser Evaluation Form are presented. Third, a comparison of resident assistants' responses to the D-Scale with seniors' responses is shown. Finally, mean and standard deviation scores for Michigan State University resident assistants' responses to the Resident Adviser Evaluation Form are shown with those obtained at Indiana University. In this study 100 resident assistants who were employed in the residence balls at Michigan State University were rated in regard to their effectiveness as resident assistants. The ratings were made by 570 residents and 260 graduate-level advisers in response to the 96 behavior statements included in the Resident Adviser Evaluation Form. The resident assistants were selected according to their scores on the Rokeach Dogmatism Scale (D-Scale). Male and female resident assis- tants were included. Four groups of 25 subjects were selected. The female resident assistants who received the 25 lowest scores on the D-Scale were assigned to the "female low authoritarian" group. Females who received the 25 highest scores were assigned to the "female high 42 43 authoritarian" group. Identical procedures were followed in assigning male resident assistants to low and high authoritarian groups. Means and standard deviations were calculated from the effec- tiveness ratings received by the resident assistants included in the high and low authoritarian groups. These are presented in Table 2. It can be readily seen that the resident assistants are quite consistently rated higher by graduate-level advisers than they are by residents. It is also obvious that there is little variation from one group of resi- dent assistants to another within the columns of ratings made by the two groups of raters. Table 2: Mean and Standard Deviation Values for Female and Male Resi- dent Assistants According to Ratings on the Resident Adviser Evaluation Form by Residents and Graduate-Level Advisers. Means Standard Deviations Graduate- Graduate- D-Scale Level Level Sex N Scores Residents Advisers Residents Advisers Female 25 low 33.650 37.038 5.269 7.313 Female 25 high 33.574 36.771 4.398 6.759 Male 25 low 33.720 37.064 5.573 9.177 Male 25 high 32.871 36.592 4.434 8.547 Totals 100 33.454 36.866 4.918 7.949 Comparisons of the Ratings To study the differences between the ratings by residents and graduate-level advisers which were received by the four groups of resi- dent assistants, analysis of variance was used. The analysis of variance program provided comparisons between the following variables: 44 male and female resident assistants, low and high authoritarian resi- dent assistants, and residents' and graduate-level advisers' ratings. In addition, the analysis of variance provides a description of the effects of interactions between all possible combinations of these factors. Table 3: Analysis of Variance Table.1 Responses by Residents and Graduate-Level Advisers to the Resident Adviser Evaluation Form for Female and Male Low and High Authoritarian Resident Assistant Groups. Sum of Mean Source of Variance d.f. Squares Squares F Statistic P Sex (between male and female groups) 1 1.927 1.927 .033 n.s. Authoritarianism (between low and high groups) 1 8.657 8.657 .149 n.s. Interactions (between sex and authoritarianism) 1 2.994 2.994 .051 n.s. Within Grdups (error variance) 96 5547.508 57.786 Raters (between resident and graduate-level adviser groups) 1 582.258 582.258 18.916 .0005 Interactions (between sex and raters) l .719 .719 .023 n.s. Interactions (between authoritarianism and raters) l .106 .106 .003 n.s. Interactions (between sex, authoritarianism, and raters) 1 1.007 1.007 .032 n.s. Remaining Within Groups (error variance) 96 2954.981 30.781 Total 100 1Table 3 is an adaptation of the analysis of variance table which is found on page 228 of Kerlinger, Fred N. Foundations of Behavioral Research, Holt, Rinehart and Winston, Inc., New York, 1964. 45 According to Kerlinger, part of the total variance which is associated with a set of data may be due to the joint action, or inter- action, of the independent variables involved. (11:218-219) In the present study these independent variables include sex, authoritarianism, and raters' position. In Table 3 the results of the analysis of variance due to sex, authoritarianism, and raters' position as well as the interactions of these three variables are presented. In regard to the first hypothesis, that there is no difference in effectiveness between male and female resident assistants, an F ratio of .033 was obtained. This F ratio is not significant at the .05 level of significance because it is smaller than the necessary critical ratio of 3.96 with l and 96 degrees of freedom. Therefore the null hypothesis of no difference between male and female resident assistants is accepted. The second hypothesis tested was that there is no difference in effectiveness between low and high authoritarian resident assistants. The analysis of variance yielded an F ratio of .149 which, given 1 and 96 degrees of freedom, is not significant at the .05 level. This hypothesis of no difference between low and high authoritarian resident assistants is therefore, accepted. Hypothesis number 3 stated that there are no interaction effects between the sex of the resident assistant and his score on the D-Scale. The results of the analysis of variance show that neither the resident assistant's sex nor his score on the D-Scale affect positively or negatively his effectiveness ratings. The F ratio with 1 and 96 degrees of freedom is not significant at the .05 level. The third null hypothesis is, therefore, accepted. 46 In hypothesis number 4 it was stated that there is no difference between ratings by residents and by graduate-level advisers of resident assistants' effectiveness. Table 3 shows that an F ratio of 18.916 was obtained from an analysis of the variance due to different raters. It is readily apparent that this F ratio is statistically significant. With 1 and 96 degrees of freedom, F ratios of 3.96 are statistically significant at the .0005 level. Table 2 shows that the comparatively large amount of variance is due to the fact that graduate-level advisers quite consistently rated all subgroups of resident assistants about 3 points higher than residents rated them. The null hypothesis of no difference between the two groups of raters is rejected. Hypothesis number 5 stated that there are no interaction effects between the sex of the resident assistant and positions of the raters. The results of the analysis of variance show that neither of these sources of variance affects positively or negatively the resident assistants' effectiveness ratings. The obtained F ratio, .023, is not significant. Therefore null hypothesis number 5 is accepted. The sixth hypothesis which was tested stated that there are no interaction effects between resident assistants' scores on the D-Scale and positions of the raters. An F ratio of .003 was obtained from the analysis of these two sources of variance. It is readily apparent that neither of these factors affects positively or negatively the effec- tiveness ratings which the resident assistant receives from residents and graduate-resident advisers. The F ratio is not significant, and .null hypothesis number 6 is, therefore, accepted. Finally, hypothesis number 7 stated that there are no interaction effects between the sex of the resident assistants, their scores on the 47 D-Scale, and positions of the rater. Again, a very small F ratio, .032 was obtained. It is obvious that there are no interaction effects between these factors. Since the F ratio is not significant, null hypothesis number 7 is accepted. Reliability of the Instruments Used in the Study For the purpose of obtaining estimates of the reliability of the instruments which were used in the study, split-half reliability coefficients were computed on the responses to the D-Scale and the Resident Adviser Evaluation Form. To obtain the reliability estimates, a Pearson product moment (r) correlation coefficient was first computed between the odd and even scores. Reliability coefficients were then computed by using the Spearman-Brown correction formula. When corrected using the Spearman—Brown formula, the estimated reliability of the D-Scale as a whole based on the 328 responses by resident assistants was r11=.73. The estimate of reliability for the 570 responses to the Resident Adviser Evaluation Form by residents when corrected was r11=.88. The estimated reliability for the Resident Adviser Evaluation based on 260 graduate-level advisers' responses was r11=.8l. The effectiveness scores on which the four groups of resident assistants were compared were mean scores which were usually derived from several independent ratings. As an example, the mean scores for resident assistants based on residents' ratings were derived from ratings by 5 or 6 residents. It may therefore be concluded that the 48 effectiveness ratings on which the estimates or reliability were computed are somewhat more reliable than they indicate.2 Comparison of Resident Assistants' and Seniors' Responses to the D-Scale It was the Opinion of the writer from the very beginning of the present study that the possibility existed that resident assistants at Michigan State University were less authoritarian than their classmates in general. For this reason a comparison was made between resident assistants' responses to the D-Scale and data which were collected on a group of Michigan State University seniors. (14:52) Table 4: Means, Standard Deviations, and t Test Values for Resident Assistants' and Seniors' Scores on the D-Scale. Resident Assistants Seniors Sex N Mean S.D. N Mean S.D. d.f. t Statistic P M 158 128.43 23.31 590 153.98 22.94 746 -12.24 .001 F 170 122.58 19.39 461 146.69 24.16 629 -12.90 .001 Totals 328 125.40 21.53 1051 150.33 23.55 In Table 4 the data relative to the comparison of D-Scale scores between resident assistants and seniors at Michigan State University are presented. It is readily apparent that both male and female 2Appreciation is granted to Dr. Andrew Porter of the Office of Research Consultation, College of Education, Michigan State University for his consultation concerning this matter. 49 resident assistants' scores are significantly lower than the seniors'. The t test of differences between the means of the groups was signi- ficant at the .001 level. In summary, the data show that the total resident assistant group is much less authoritarian than the senior group. Similar to what has been found in studies of other personality characteristics among resident assistant groups (39,17), the female resident assistant population at Michigan State University is rela- tively homogeneous in regard to their scores on the D-Scale. Homo- geneity is shown in Table 4 by the fact that the standard deviation for female resident assistants is 19.39 as compared to a standard deviation for female seniors of 24.16. Homogeneity was not found, however, among the male resident assistants as is shown by the fact that they received a larger standard deviation value than did the male senior group. Comparison of Resident Assistants at Michigan State University and Indiana University In Table 5 the ratings of resident assistants' effectiveness which were obtained in studies at Indiana University and Michigan State University are presented. This information is presented for the pur- pose of examining the ratings obtained in the present study compared to ratings obtained in other studies in which the same rating instru- ment was utilized. There are several differences between the studies done at Michigan State University and at Indiana University which should be pointed out. In the present study the resident assistants were junior and senior 50 undergraduates. However, in the Indiana study the student assistant sample consisted of graduate students as well as undergraduates. These two studies also differed in that, in the Indiana study, only one group of raters, residents, was involved. In the present study graduate- level advisers as well as residents rated the resident assistants. Finally, an additional factor, sex, was included in the present study. Male and female resident assistants rather than males only were rated at Michigan State University. Table 5: Effectiveness Ratings for Resident Assistants at Michigan State University and Indiana University Utilizing the Resident Adviser Evaluation Form. Number of Number of Resident Source Respondents Respondents Assistants Rated Mean S.D. MSU Residents 570 100 33.45 4.918 MSU G-LA 280 100 36.86 7.949 IU Residents 1058 41 30.86 3.32 As the table shows, the ratings obtained at Michigan State University by both residents and graduate-level advisers for both sexes were considerably higher than the ratings obtained at Indiana University. It is also evident that the standard deviations for all groups at Mich- igan State University were much larger. There are many possible causes for such differences in ratings. For one, due to a better selection process, a better training program, or both of these in combination, the resident assistants at Michigan State University could simply be more effective than those at Indiana University. 51 Second, differences in the ratings at the two institutions may be due to differences in the way that the Evaluation Form was administered. At Indiana, resident assistants were involved in distributing the Evaluation Forms and collecting the responses, whereas in the present study, distribution and collection were accomplished through the use of the campus mail service or by selected student personnel workers. A third cause of differences between the ratings at the two insti- tutions may be the fact that the percentage of returns were not at all equal. The per cent of returns from 1,928 residents at Indiana Univer- sity was only 54, as compared to a response from the resident sample at Michigan State University of 95 per cent. It is possible that non- respondents at Indiana University would have rated their residents higher than did the respondents,making the responses at Indiana more similar to those obtained in the present study. In Chapter IV the results of the data collected in the study were analyzed. In the next chapter the summary, conclusions, and discussion of the results will be presented, followed by some suggestions for further research. CHAPTER V SUMMARY, CONCLUSIONS, AND DISCUSSION Resident assistants are employed in college and university residence halls throughout the country to assist students in making the most of their college experience. Increasingly, as the students' residence is viewed as an integral part Of the college learning experience, the demand for highly effective resident assistants is becoming greater. Unfortunately, because present selection processes rely heavily on subjective judgments, the identification and subsequent rejection of potentially ineffective resident assistants is difficult. The purpose Of this study was to identify quantitative criteria which, when added to the subjective criteria being used at present, could improve the selection Of effective resident assistants. In this study an attempt was made tO determine if authoritarianism is negatively related to resident assistant effectiveness. If a rela- tionship does exist between these two variables, then the addition of an authoritarianism scale to the selection process could improve the selection of resident assistants. Several secondary hypotheses involv- ing variations in sex and raters were also tested. Design: During the winter term Of 1968 the Rokeach Dogmatism Scale, Form E was administered at group meetings to 158 male and 170 52 53 female undergraduate resident assistants employed at Michigan State University. High and low authoritarian groups were then selected by taking the 25 highest and lowest scores for each sex. Effectiveness ratings were then Obtained from two different groups of raters. One group was comprised of the graduate-level advisers who worked in the resident assistant's hall. The second group consisted Of 100 samples of six residents selected at random from each resident assistant's living unit. The Resident Adviser Evaluation Form was used by both groups to assess the resident assistants' effectiveness. Usable responses were Obtained from 570 or 95 per cent of the resident sample. The 98 graduate-level advisers returned 260 ratings. Approximately 86 per cent of the graduate-level advisers responded. Four-way analysis Of variance was used to study differences in the ratings. Description of Instruments: The Rokeach Dogmatism Scale consists of 40 statements with which subjects are asked to agree or disagree along a 6-point continuum. The score Obtained is a measure of general authoritarianism. The Resident Adviser Evaluation Form consists of 32 blocks of three statements each from which subjects are asked to select at least one but not more than two statements which are most descriptive of his resident assistant. Two Of the statements in each block have a high index for discriminating between effective and ineffective resident assistants. The third statement has a low discrimination index. Only high index items are tallied, and high scores are associated with effectiveness. 54 Reliability of the Instruments: The estimated reliability of the D-Scale as a whole based on the 328 responses by resident assistants was r11=.73. Based on the 570 responses to the Resident Adviser Evaluation Form by residents, the estimate of reliability was r11=.88. The estimated reliability based on 260 graduate-level advisers' responses to the rating form was r .81. 11= Research Hypotheses: It was hypothesized that differences in effectiveness would be found between the resident assistant groups according to variations in sex, authoritarianism, and raters. Further- more, it was hypothesized that joint actions of these three independent variables would result in other differences in effectiveness. Conclusions: Seven F ratios were obtained by four-way analysis of variance of the effectiveness scores obtained from residents and graduate- 1evel advisers. The results of these statistical comparisons are sum- marized below with the hypotheses from which they were derived. 1. There is no difference in effectiveness between male and female resident assistants. An F ratio of only .033 was obtained. Therefore the null hypothesis of no difference between male and female resident assistants is aécepted. 2. There is no difference in effectiveness between low and high authoritarian resident assistants. The analysis of variance yielded an F ratio of .149. Since this ratio was smaller than the necessary ratio of 3.96, the null hypothesis of no difference between high and low authoritarian resident assistants is accepted. 3. There are no interaction effects between the sex of the resident assistants and scores on the D-Scale. An F ratio of .051 was obtained between these two sources Of variance. Since this ratio 55 is not significant at the .05 level, this null hypothesis is accepted. 4. There is no difference between ratings by residents and by graduate-level advisers of resident assistants' effectiveness. An F ratio of 18.916 was obtained which is significant at the .001 level. Graduate-level advisers rated resident assistants significantly higher than did residents. Therefore the null hypothesis of no difference between the two groups of raters is rejected. 5. There are no interaction effects between the sex of the resident assistants and the position of the rater. The F ratio which was Obtained is .023. Since this ratio is not significant, this null hypothesis is accepted. 6. There are no interaction effects between resident assistants' scores on the D-Scale and positions Of the raters. An F ratio of .003 was obtained which is not significant at the .05 level. Therefore this null hypothesis is accepted. 7. There are no interaction effects between the sex of the resident assistants, their scores on the D-Scale, and positions of the raters. An F ratio of .032 was obtained. Since this ratio is not significant at the .05 level, the null hypothesis is accepted. Discussion The theory on which the primary hypothesis in this study was based was that resident assistants who obtain high scores on the Rokeach Dogmatism Scale are less effective than resident assistants who Obtain low scores on this scale. As the F ratios included in Table 2 show, this theory was not supported by the data collected in the study. 56 It may very well be that a resident assistant's effectiveness as measured by the Resident Adviser Evaluation Form is not affected by how dogmatic or authoritarian he happens to be. If however, authoritarian resident assistants are in actuality less effective than nonauthori- tarian resident assistants, then the fact that such a relationship was not supported in the present study may be due to weaknesses in the study. One very possible reason why significant differences between the high and low authoritarian groups were not obtained could be due to the fact that the resident assistant population as a whole is extremely nonauthoritarian. This fact is shown by the data presented in Table 4. As the mean scores for both males and females show, the resident assistants obtained much lower scores on the D-Scale than did a random sample of Michigan State University seniors. The standard deviation shown in Table 4 also is smaller for female resident assis- tants. Data obtained in other studies also support the conclusion that the resident assistants who were included in the present study as a group are nonauthoritarian. Rokeach, for example, found mean scores which ranged from 141.8 to 143.8 from 288 sUbjects of both sexes tested in studies at Ohio State University and Michigan State University prior to 1958. (33:90) These mean scores are much higher than those obtained in the present study. The D-Scale scores Obtained by the resident assistants indicate that the subjects in the "high authoritarian" group were actually only "middle authoritarian.” It seems reasonable to assume, therefore, that the subjects who were included in the "high authoritarian" resident assistant group are no more authoritarian in their behavior than are 57 college students in general. Since these resident assistants as a group Obtained D-Scale scores which are only slightly higher than those which would be obtained by the average college senior, it may be assumed that behavior which would be affected by this personality variable would vary little between the two groups. The behavior of this group of "high authoritarian" resident assistants is probably different than the behavior of authoritarian individuals, and it is, therefore, unlikely that this particular personality characteristic would affect the resident assistant's performance. There is, Of cOurse, the possibility that failure to Obtain signi- ficant differences in the present study is due to weaknesses in the instruments which were used. From the information which was obtained in this regard, however, it does not seem likely that this is the cause of failure. The reliability of the D-Scale, for example, which is .73 in this study, is somewhat lower than is desirable, but, in the opinion of the writer, it is within the range of acceptability. In regard to the Resident Adviser Evaluation Form, the two estimates of reliability which were Obtained were relatively high. These were .88 for the responses by residents, and .81 for responses by graduate-level advisers. Unfortunately, little information was obtained in regard to the validity of the instruments which were used. Comments made during the investi- gation by respondents Of the resident sample verbally and in writing, however, seem to indicate that the validity of the Resident Adviser Evaluation Form was high. High validity was demonstrated by the fact that students who felt that their resident assistant was ineffective generally rated him low on the rating form. Those who felt that their resident assistant was effective generally rated him high on the rating 58 form. Validity of this type may also have occured among the subjects in the graduate-level adviser sample. It was not Observed, however, due to the fact that their responses were obtained by mail. Several secondary hypotheses were tested in this study. These pertained to differences in effectiveness between the high and low authoritarian groups due to variations in the sex of the resident assistant and the position of the rater. None of these secondary hypotheses were theoretically supported, and were merely included for the purpose Of trying to find out if relationships did in fact exist between the factors involved. Only one of these secondary hypotheses was found to be signifi- cant. This one pertained tO differences in effectiveness between ratings by residents and ratings by graduate-level advisers. The results of the study show that head resident advisers, assistant resi- dent advisers, and graduate resident advisers rate the resident assis- tants in their hall significantly higher than do the residents who live in the resident assistant's living unit. There are several plausable explanations as to why these two groups of raters rated the resident assistants differently. One possibility is that the resident assistant may play one role with residents and another with graduate-level advisers. This possibility is supported by a study in which it was found that resident advisers, resident assistants, and students all differ significantly in their expectations Of the resident assistant's role. (18) Because of differences in rewards and expectations associated with either of the rater groups or with the resident assistant himself, a resident assistant may present himself to one group of raters in a way that is different than the way he presents 59 himself to his residents. As a result the two groups of raters may actually see different behavior exhibited by the resident assistant. Another explanation for higher ratings by graduate-level advisers may be due to the fact that graduate-level advisers are involved in the selection of resident assistants. Because they are responsible for selecting resident assistants, they may see them as being a "select" group of students who have many strengths, particularly the ones which they had looked for during the selection period, but relatively few weaknesses. They may therefore unconsciously look for and see exhibited certain preferred behaviors which are not seen by residents. Recommendations for Further Research There is need for further research concerning the characteristics which affect the performance of resident assistants. Specifically, further research should be conducted regarding the effect of authori- tarianism on the effectiveness of resident assistants. It is reasonable to speculate that the failure to find significant differences in effectiveness between the low and high authoritarian groups in the present study is due to the fact that the resident assistant population as a whole was nonauthoritarian. Therefore it is the opinion of the writer that a replication of the present study or variations of it would be worthy contributions to the improvement of the resident assis- tant selection process. It may very well be possible that effective selection procedures result in the elimination of "high authoritarian" students. The writer suggests, therefore, that research of this kind be conducted in 60 institutions where those who are responsible for hiring resident assistants are not highly trained in the processes and procedures Of personnel selection. During the Course of this investigation it became apparent to the writer that many of the resident assistants had good rapport, or were effective, with a large number of their residents, but at the same time also had poor rapport with many of them. It seemed that resident assistants who were excessively permissive were criticized as seriously as those who set up undue restrictions or who controlled residents' behavior too rigidly. In the Opinion of the writer, an investigation into how the expectations Of each resident and the behavior of his resident assistant could be balanced to Obtain an optimal "fit" could make a valuable contribution to the improvement of relationships between resident assistants and their residents. BIBLIOGRAPHY 10. ll. BIBLIOGRAPHY Aceto, Thomas D. "Students in Pre-Professional Staff Roles," The Journal gf College Student Personnel. Vol. IV, No. 1, October, 1962, 23-27. Adorno, T. W., Frankel-Brunswick, Else, Levinson, D. J. and Sanford, R. N. The Authoritarian Personality. New York, Harper, 1950. Brady, Marna V. "Student Counselor Selection," Personnel and Guidance Journal. Vol. XXXIII, NO. 5, January, 1955, 289-292. Christie, Robert and Cook, Peggy. "A Guide to Published Literature Relating to the Authoritarian Personality Through 1956," Journal of Psychology. Vol. XLV, 1958, 171-199. Crane, William J. "Administrative Practices in Men's Housing," The Journal 2f College Student Personnel. Vol. III, No. 2, December, 1961, 70-76. Duncan, James P. "A Rating Scale for Student Evaluation of Resi- dence Hall Counselors,” Personnel and Guidance Journal. Vol. XLV, NO. 5, January, 1967, 452-454. "Construction of a Forced-Choice Rating Scale for Student Evaluation of Residence Hall Counselors," unpublished Doctoral thesis, Indiana University, Bloomington, 1965. Enos, Brian R. "Selection of Residence Hall Counselors at a Group Of Colleges and Universities," unpublished study Obtained from the author, completed May 23, 1964. Hoyt, Donald P. and Davidson, Alexander. "Evaluating Residence Hall Advisers," The Journal of College Student Personnel. Vol. VIII, NO. 4, July, 1967, 251-256. Kemp, C. Gratton. "Influence Of Dogmatism on the Training of Counselors," Journal 2: Counseling Psychology. Vol. IX, Summer, 1962, 155-157. Kerlinger, Fred N. Foundations 9f Behavioral Research. Holt, Rinehart, and Winston, Inc., New York, 1964. 62 12. 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 63 Kidd, John W. ”Positive and Negative Leadership Traits in a College Men's Residence Hall," North Central Association Quarterly. Vol. XXIX, NO. 4, April, 1955, 360-362. Krathwohl, David R. How tg Prepare a Research Proposal. 1965. Pamphlets available from the Syracuse University Bookstore, Syracuse, New York. Lehmann, Irvin J. and Dressel, Paul L. Critical Thinking, Attitudes, and Values in Higher Education: Final Report 9f Cogperative Research Project No. 590. East Lansing: Michigan State University, 1962. Lingoes, James C. A Review of the Minnesota Multiphasic Person- ality Inventory, in The Sixth Mental Measurements Yearbook. O. K. Buros, Editor. The Gryphan Press, Highland Park, New Jersey. 1965, 317-319. McNemar, Quinn. Psychological Statistics. John Wiley and Sons Inc., New York, 1962. Madson, Dennis Leroy. "Factors Associated with Effectiveness as a Male Undergraduate Residence Hall Counselor," unpublished Doctoral thesis, Ohio University, Athens, 1966. Marquardt, Harold Roy. "The Role of the Resident Assistant in the Men's Residence Halls at Michigan State University," unpublished Doctoral thesis, Michigan State University, East Lansing, 1961. Milliken, Robert L. and Paterson, John S. "Relationship of Dogmatism and Prejudice to Counseling Effectiveness," Counselor Education and Supervision. Vol. VI, Winter, 1967, 125-129. Mueller, Kate Hevner. Student Personnel Work in Higher Education. Houghton Mifflin Company, Boston, 1961. Murphy, Raymond 0. "Administrative Practices in Utilizing Students as Staff in Residence Halls," Journal 2f College Student Personnel. Vol. VI, December, 1964, 109-113. Noll, Victor H. Introduction £9 Educational Measurement. Houghton Mifflin Company, Boston, 1957. Olsen, Merle M. "Developments in Residence Hall Counseling," Educational and PsyChological Measurements. Vol. X, No. 3, Autumn, 1950, 455-464. ”Evaluation of Dormitory Counselors' Services," Educational and Psychological Measurements. Vol. XI, NO. 3, Autumn, 1951, 419-426. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 64 Ortenzi, Angelo. "Establishment and Cross-Validation of Selection Criteria for Resident Counselors at the Pennsylvania State UniverSity," unpublished Doctoral thesis, The Pennsylvania State University, University Park, 1965. Osgood, C. E. Suci, G. J. and Tannenbaum, P. H. The Measurement 2f Meaning. University of Illinois Press, Urbana, 1957. Peterson, Martha E. "An Evaluation of Relationships Between Test Data and Success as a Residence Hall Counselor," unpublished Doctoral thesis, University of Kansas, Lawrence, 1959. Pettigrew, T. F. ”The Measurement and Correlates of Category Width as a Cognitive Variable," Journal gf Personality. Vol. XXVI, 1958, 532-544. Raines, Max Reid. "A survey of the Counseling and Activities Programs Within the Men's Residence Halls of the Big Ten Universities," unpublished Master's thesis, Indiana University, Bloomington, 1949. "The Role of the Part-Time Student Assistant in the Men's Residence Halls of the Big Ten Universities,” unpublished Doctoral thesis, Michigan State University, East Lansing, 1952. Riker, Harold C. College Housing gs Learning Centers. Vol. III in the Student Personnel Series, American Personnel and Guidance Association, Washington, D. C., November, 1965. Rokeach, Milton. "Political and Religious Dogmatism: An Alterna- tive to the Authoritarian Personality," Psychological Monographs, Vol. LXX, No. 18, 1956. The Open and Closed Mind. Basic Books, Inc., New York, 1960. Scodel, Alvin and Mussen, Paul. "Social Perceptions of Authori- tarians and Nonauthoritarians," Journal 2f Abnormal and Social Psychology, Vol. XLVIII, 1952, 181-184. Sheeder, William B. ”Role Playing as a Method of Selecting Dormitory Counselors," Journal 2: College Student Personnel. Vol. IV, March, 1963, 354-358. Shils, Edward A. "Authoritarianism: Right and Left," in Studies 33 the Scope and Method 2: the Authoritarian Personaliyy. Christie and Johoda, Editors, The Free Press, Glencoe, 1954. Sifferd, Calvin S. "Evaluation of a Residence Hall Counseling Program," Collegg and University. Vol. XXV, No. 3, April, 1950, 445-447. 38. 39. 40. 41. 42. 65 Residence Hall Counseling. McKnight and McKnight Publishing Co., 1954. Simons, Wesley S. "The Personality Characteristics of the Residence Hall Assistant as Related to Job Performance," unpublished Doctoral thesis, Michigan State University, East Lansing, 1957. Stark, Matthew. "Pointers on Selecting and Training Residence Hall Staff," College and University Business. Vol. XXXVI, No. 5, May, 1959, 50-52. Thorndike, Robert L. and Hagen, Elizabeth. Measurement and Evaluation in Psychology and Education. John Wiley and Sons Inc., New York, 1955. Williamson, E. G. ”Student Residences: Shelter or Education," The Personnel and Guidance Journal. Vol. XXXVI, February, 1958, 392-398. APPENDIX A ‘k ROKEACH DOGMATISM SCALE INSTRUCTIONS: The following items pertain to what the general public thinks and feels about a number of important social and personal questions. The best answer to each statement below is your personal opinion. Many different and opposing points Of view are included; you may find yourself agreeing strongly with some of the statements, disagreeing just as strongly with others, and perhaps uncertain about others. Whether you agree or dis- agree with any statement, you can be sure that many other people feel the same as you do. Mark each of the following statements according to how much you agree or disagree with it. Please mark evegy statement. Blacken spaces +3, +2, +1, -1, -2, -3 depending on how you feel in each case. +3 I agree very much. -3 I disagree very much. +2 I agree on the whole. -2 I disagree on the whole. +1 I agree a little. -1 I disagree a little. +3 +2 +1 -1 —2 -3 1. In times like these it is often necessary to be more on guard against ideas put out by people or groups in one's own camp than by those in the Opposing camp. +3 +2 +1 -1 -2 -3 2. In times like these, a person must be pretty selfish if he considers primarily his own happiness. +3 +2 +1 -1 -2 -3 3. The worst crime a person could commit is to attack publicly the people who believe in the same thing he does. +3 +2 +1 -1 -2 -3 4. If a man is to accomplish his mission in life it is sometimes necessary to gamble ”all or nothing at all." * The name of this instrument was not included on the copies which were administered to the resident assistants in the present study. 67 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 5. 10. 11. 12. 13. 14. 15. 16. 17. 18. 68 A group which tolerates too much difference of Opinion among its own members cannot exist for long. . While I don't like to admit this even to myself, my secret ambition is to become a great man like Einstein, or Beethoven, or Shakespeare. . When it comes to differences of opinion in religion we must be careful not to compro- mise with those who believe differently than the way we do. In this complicated world of ours the only way we can know what is going on is to rely on leaders or experts who can be trusted. . If given the chance I would do something of great benefit to the world. Of all the different philosophies which exist in this world there is probably only one which is correct. It is only natural for a person to be rather fearful of the future. There are two kinds of people in the world; those who are for truth and those who are against the truth. There is so much to be done and so little time to do it in. There are a number of persons I have come to hate because of the things they stand for. Fundamentally, the world we live in is a pretty lonesome place. Unfortunately, a good many people with whom I have discussed important social and moral problems don't really understand what's going on. A person who thinks primarily of his own happiness is beneath contempt. Once I get wound up in a heated discussion I just can't stop. +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +3 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 69 In the long run the best way to live is to pick friends and associates whose tastes and beliefs are the same as one's own. It is only natural that a person should have a much better acquaintance with ideas he believes in than with ideas he opposes. Man on his own is a helpless and miserable creature. I'd like it if I could find someone who would tell me how to solve my personal problems. A man who does not believe in some great cause has not really lived. In a discussion I Often find it necessary to repeat myself several times to make sure I am being understood. Even though freedom of speech for all groups is a worthwhile goal, it is unfortunately necessary to restrict the freedom of certain political groups. Most Of the ideas which get printed nowa- days aren't worth the paper they are on. The present is all too often full of unhappiness. It is only the future that counts. The United States and Russia have just about nothing in common. It is only when a person devotes himself to an ideal or cause that life becomes meaningful. ‘It is often desirable to reserve judgment about what's going on until one has had a chance to hear the Opinions of those one respects. The highest form of government is a democracy and the highest form of democ- racy is a government run by those who are most intelligent. +3 +3 +3 +3 +3 +3 +3 +3 +3 +2 +2 +2 +2 +2 +2 +2 +2 +2 +1 +1 +1 +1 +1 +1 +1 +1 +1 -1 -2 -3 32. 33. 34. 35. 36. 37. 38. 39. 40. 70 In the history of mankind there have probably been just a handful of really great thinkers. To compromise with our political opponents is dangerous because it usually leads to the betrayal Of our own side. A person who gets enthusiastic about too many causes is likely to be a pretty "wishy-washy" sort of a person. The main thing in life is for a person to want to do something important. Most people just don't know what's good for them. In a heated discussion I generally become so absorbed in what I am going to say that I forget to listen to what the others are saying. My blood boils whenever a person stubbornly refuses to admit he's wrong. It is better to be a dead hero than a live coward. Most people just don't give a "damn" for others. RESIDENT ADVISER EVALUATION FORM Instructions This is a rating scale for student evaluation of resident assis- tants. The scale consists of thirty-two groups or blocks containing three statements each. The statements are descriptive Of resident assistants (RAS) and their behavior. Please consider carefully the statements in each block and select the statement(s) that are mpg; applicable to 1225 RA. Be certain to pick fig least one statement but 29 more than two statements in each block. Mark your choice of statement or statements from each block of items 13 the proper brackets gn_the IBM answer sheet as shown in the "Sample” below. SAMPLE: IBM ANSWER SHEET 1-A Is Sincere A B C D E 1-B Speaks Well at Unit Meetings 1-C Is Easy to Get to Know Use alternatives A, B, and C only on the answer sheet. Check to see that you have chosen 3; least one statement but not more than two for each of the thirty-two sets of items. PLEASE LEAVE NONE BLANK. 71 72 Select one but not more than two statements from each block. l-A Dresses Neatly And Attractively 1-B Exhibits Self Confidence l-C Is Fair In All Dealings With Students 2-A Is Mature 2-B Has Above Average Intelligence 2-C Has Good Common Sense 3-A Has A Warm And Pleasing Personality 3-B Is Easy To Get Along With 3-C Does Not Coddle The Students 4-A Sets A Good Example For Students 4-B Respects The Feelings And Opinions Of Students 4-C Exhibits Good Study Habits 5-A Is Easy To Talk To Or Approach 5-B Does Not Use Position For Personal Gain 5-C Is Willing To Admit A Mistake 6-A Has A Background In Psychology 6-B Has A Basic Knowledge Of The Psychological Make-Up Of People 6-C Is Able To Talk With Students About Personal Problems 7-A Possesses Good Table Manners 7-B Has Ability To Mingle With Students Is Poised And Well-Mannered 73 8-A Treats Students According To Their Maturity 8-B Helps Those Who Have Violated Rules 8-C Considers Situation And Does Not GO Strictly By The Book (Flexible) 9-A Does Not Drink 9-B Is Able To Be Strict When And If Necessary 9-C Is Willing To Cooperate With Men And Officers Of The Unit lO-A Is Firm But Understanding In Matters Of Discipline 10-B Is Able To Exert Authority Without Creating Ill-Feelings lO-C Is Not A Cop ll-A Has Had Wide Experience In Activities And Can Offer Sound Advice ll-B Is Able To Recognize His Own Limitations ll-C Has Had Experience In Advising Students 12-A Takes Time To Know Names Of All His Students 12-B Has Lived In Residence Halls Before 12-C Accepts Criticism In Good Grace l3-A Has Relatively High Moral Standards l3-B Has Strength Of Character 13-C Is Tolerant Of Other People's Views l4-A Does Not Hold Grudges 14-B Has Good Rapport With Students 14-C Is A Friend Of The Student 74 15-A Commands Respect Of Students 15-B Is Able To Control Students Without Use Of Threats 15-C Obeys All Rules Himself l6-A Has Social Graces l6-B Knows What It Is Like To Be An Undergraduate l6-C Is Physically Clean 17-A Is Efficient In Handling Administrative Responsibilities l7-B Does Not Associate Only With Resident Assistants 17-C Is Relatively Attractive l8-A Enjoys Being A Resident Assistant 18-B Understands Well His Duties And Purpose 18-C Has Spent One Or More Semesters On Campus l9-A Eats Meals With Students 19-B Is On Hand To Enforce Quiet Hours l9-C Is Available A Reasonable Amount Of The Time 20-A Is Well-Informed Concerning University Policies, Procedures, And Activities 20-B Has Experience In Working With People Of All Races And Backgrounds 20-C Is Able To Secure The Cooperation Of Students 21-A Has A Sense Of Humor 21-B Has A Great Deal Of Patience 21-C Is Neat And Well Groomed 75 22-A Follows Through With Action Promised 22-B Is Lenient - Allows Some Rules To Be Broken 22-C Has Courage Of Convictions 23-A Dates Members Of Opposite Sex 23-B Has An Open-Door Policy 23-C Visits And Talks With Students In Their Rooms 24-A Has Good Personal Habits 24-B Is Even-Tempered From Day To Day 24-C Keeps Confidence 25-A Does Not Act Superior To His Students 25-B Is Natural - Unaffected By Authority 25-C Treats Residents With Respect 26-A Is Congenial - Easy Going 26-B Has An Out Going Personality - Aggressive 26-C Always Has A Pleasing Smile And Greeting 27-A Is Able To Converse On Students Level 27-B Is Interested In The Academic Well-Being Of His Residents 27-C Is Academically Oriented (Encourages Scholarship) 28-A Is Self-Disciplined 28-B Is Dependable 28-C Is Able To Remain Calm And Act Competently In Emergencies 76 29-A Is Willing To Sacrifice Time To Help Students 29-B Has A Real Devotion To The Job 29-C Has A Strong Sense Of Responsibility 30-A Establishes Good Communications With Unit Officers 30-B Attends All Unit Meetings 30-C Has Genuine Interest In Students And Their Problems 3l—A Places Trust And Confidence In Students 31-B Is Discreet In Handling Problems 31-C Does Not Pry Into Students' Private Life 32-A Is Fairly Strict But Not Sticky On Minor Rules 32-B Upholds University And Residence Hall Regulations 32-C Avoids Being A Disciplinarian Or Dictator Type APPENDIX B 78 Letter sent by the Director of Residence Hall Programs for the purpose of informing head resident advisers Of the study. DATE: February 1, 1968 TO: All Head Resident Advisers FROM: Don Adams SUBJECT: Participation in Research Study Mr. Norman Hefke, a doctoral student in student personnel work, has developed a study which he believes can make a significant contribution to the improvement Of the resident assistant selection process. I have examined the research proposal and believe it to be worthy Of whatever assistance we can provide. I am therefore asking each advisory staff member to cooperate in whatever way he can while this study is in progress. The study requires the administration of two instruments. The first one in the series is to be administered to our total resident assistant staff. It requires only 15 to 20 minutes to complete, and may be admin- istered in a group situation such as a staff meeting. The second instrument is to be administered at a later date to two different popu- lations. One population will consist of selected head resident advisers, assistant resident advisers and graduate resident advisers. The other population will consist of selected residents from residence halls throughout the campus. This second instrument can be taken by staff members at their convenience. Students will probably be asked to respond in group meetings arranged by Mr. Hefke. This instrument takes about 15 minutes to complete. Because of the nature of the factors being investigated, it is necessary that the instruments be discussed as little as possible among staff members as well as between staff and residents. Mr. Hefke or an assistant will contact you soon regarding the adminis- tration of the first instrument. I sincerely hope that staff members will do their best to aid him in completing this research. 79 Letter sent to selected residents asking them to attend a meeting where the Resident Adviser Evaluation Form was administered. Dear I am engaged in a doctoral thesis concerning the relationship of a selected personality variable to resident assistants' performance. The purpose Of this study is to gather information which could prove helpful in improving the selection of effective resident assistants. I would appreciate very much your assistance in this matter. I would like you to attend a group meeting in Rm. Of Hall on , February at p.m. At this time you will be asked to complete a short evaluation form which should take you no longer than fifteen minutes to complete. If you have any questions regarding the study, please feel free to call me at 3-3860 or 372-2895. Thank you very much for this assistance. Sincerely, Norman E. Hefke Assistant Director Student Activities Division 80 Letter sent to head resident advisers asking them to respond to the Resident Adviser Evaluation Form. Dear I am now in the final phase of my study concerning the relationship of a selected personality variable to resident assistants' performance. I want to thank you for the assistance you have given me in carrying out this research. The last part of this study requires ratings of effectiveness of selected RAS by head resident advisers, assistant resident advisers, and graduate resident advisers. These will take approximately fifteen minutes each to complete. I would appreciate very much your assis- tance in this matter. You have received an evaluation form and answer sheets for one or more RAS in your ball. These RAS are in either the high group or the low group on the personality variable with which I am dealing. Please rate each of these RAS by responding to the items in the Resident Assistant Evaluation Form. Read the instructions carefully. Please respond in pencil. The ratings of each RA by supervisory personnel in each hall will be averaged and then grouped with ratings by advisers in other halls. Once the ratings have been recorded your name will not be associated with them. I hope to start analyzing the data by March 29. Therefore, I would appreciate it very much if you could send your responses to me by that time by campus mail. If you have any questions, please feel free to call me at 3-3860. Sincerely, Norman E. Hefke 107 Student Services Building Campus 81 Letter sent to assistant resident advisers and graduate resident advisers asking them to respond to the Resident Adviser Evaluation Form. Dear I am engaged in a study concerning the relationship of a selected personality variable to resident assistants' performance. The purpose of this study is to gather information which could prove helpful in improving the selection of effective resident assistants. Part of this study requires ratings of effectiveness of selected RAs by head resident advisers, assistant resident advisers, and graduate resident advisers. These will take approximately fifteen minutes each to complete. I would appreciate very much your assistance in this matter. You have received an evaluation form and answer sheets for one or more RAs in your hall. These RAs are in either the high group or the low group on the personality variable with which I am dealing. Please rate each of these RAs by responding to the items in the Resident Assistant Evaluation Form. Read the instructions carefully. Please respond in pencil. The ratings of each RA by supervisory personnel in each hall will be averaged and then grouped with ratings by advisers in other halls. Once the ratings have been recorded your name will not be associated with them. I hope to start analyzing the data by March 29. Therefore, I would appreciate it very much if you could send your responses to me by that time by campus mail. If you have any questions, please feel free to call me at 3-3860. Sincerely, Norman E. Hefke 107 Student Services Building Campus GRN STATE UNIV LIBR IIIIIIIIIIIIIIII IIIIIIII III IIIIII2 IIIIIIIIIIIIIIES