- , ._- v ‘- I _. --— ‘1‘ .r '1' . ' 'Y ,Y _ 11:4 ‘7..' P431 El'v' .1 3*.2“ ."-" ' . ,. "~“~I.‘IA .‘ _ 1;"n I, ‘» :.-:t:.. u .. .III I t I n .’.',I ".n" "$2., .‘, ‘ ." , . l in.“ I' ‘7." V. ' I .x h. . I ’ J‘ | u > I Jpn, _I'.' " ‘l' .“ ' . . t , ’. .5: _‘ .. 1.... II ‘.' .L ‘f .." 1 a .,- - . ”I“: . . ml 2%. u ‘ ‘ 'r. -;. .‘ ' ‘ '3 ‘ " ”fiI .. o; n A - - t . -, _, u : .4.‘ \I . .7 ‘3'.- .I ”7“,“; -‘| “I” p: ‘ I ‘ . w- 7 .... _. I ' .'u f; I't‘;'1fir"alfl H" .. "_:d'::-'L}"1TIE%..‘ :I': i" .71 .i‘ffi- ~. 313:, ’va . , T 1!. H“: u'o“ 5‘ yo I. -' | [Why W ! " ’ ‘ I" ‘ 1v ‘ tack? (IV?) ‘| ' :Dg‘\ I I § O J? ‘. ‘ t: . § 1 N. y . . » . 3r 1 1' "I .l.'.‘ ' ‘q '3: 'l «1:5: “I“; M‘ k‘ 2?? j!” .'“ ' +‘ W; L": IE" ;\ \M,‘ If {5 L591} Awywwy-z‘a- .4 p- 'ill'.‘ v:‘{:I\ 'A' ‘ 'x c ""7 T WNW ‘ bi" ' “:é‘flzffl| .itll' Lv‘ktfl 'Iu'? I qu‘I’i'dmg" >5! C I "n tan ' ' 3“ ‘1’? 329,1. ’ “Tia \ ‘6' Iy. ‘ 1 why " W‘i‘ , I ’ .Ir“: 5 3‘“: “N“l’ u '9‘; ' " ‘ "lg‘u: VIM] I t:- ‘I‘U‘SAV :jg'fil : ‘ng'fi; ”:6 ‘ ‘ m1: W "W *si 3“,. '5 1M. ,. I, . . . 3“ w ..I--In,.wy a ‘ Ll‘flwtuh )>l.". ' .I‘I.,' I." ‘ “-13? ‘vu : ' “‘3": 5“?! l " .«m’lwl :3 - ' - ' " .. 3321:?! . . ' ,. ‘ .Wp‘. Mu-I - ~ ~61, : {. c. If“, :3 v I , I I ‘ V {In - ‘ . ll» ' {i T "'1. 1.1 2 Via: .' ' ‘ - " ’1‘ “b ."‘ , I“ :‘f ‘7' ' I , (.1 4. I‘. I 3' . 1' Ln“ ‘ . z'fl'r‘: I I . § . . 9% THESIS This is to certify that the thesis entitled A STUDY OF FACULTY AND ADMINISTRATION PERCEPTIONS REGARDING FACULTY INVOLVEMENT IN REDUCING STUDENT ATTRITION IN SELECTED LOW AND HIGH ATTRITION LIBERAL ARTS COLLEGES presented by Richard H. Mullendore has been accepted towards fulfillment of the requirements for Ph-DL degreein Education Major professor » Date June 24, 1980 0-7639 meal-943033 00‘; OVERDUE FINES: 25¢ per day per item RETUMING LIBRARY MATERIALS: Place in book return to remove charge from clrculetton records A STUDY OF FACULTY AND ADMINISTRATION PERCEPTIONS REGARDING FACULTY INVOLVEMENT IN REDUCING STUDENT ATTRITION IN SELECTED LOW AND HIGH ATTRITION LIBERAL ARTS COLLEGES BY Richard H. Mullendore A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Administration and Higher Education 1980 (A152 053.12 ABSTRACT A STUDY OF FACULTY AND ADMINISTRATION PERCEPTIONS REGARDING FACULTY INVOLVEMENT IN REDUCING STUDENT ATTRITION IN SELECTED LOW AND HIGH ATTRITION LIBERAL ARTS COLLEGES BY Richard H. Mullendore The subject of student attrition has received increased attention recently, due to projections for declining enrollments and shifts in postsecondary enroll- ment patterns. In many colleges, improving retention is being equated with institutional survival. As the pool of traditional college students becomes smaller, the need to create a low attrition environment becomes greater. Several investigators have speculated that faculty can play a major role in student retention programs, however, no studies were located in which faculty have been asked what impact they feel they can have in reducing attrition. The purpose of this study was to ascertain the perceptions of faculty and administrators regarding the factors associated with attrition, the amount of out of class contact that faculty state they have with students, and the impact that faculty and administrators feel that faculty can have in reducing attrition. The perceptions of faculty and administrators from high attrition colleges were compared with those from low attrition colleges. In Richard H. Mullendore addition, faculty perceptions were compared with those of administrators within high attrition and low attrition institutions. For this investigation, a questionnaire was developed and administered to all faculty and administrators in twelve small, private, four-year, liberal arts colleges (eight high attrition and four low attrition). Questionnaire items regarding factors associated with attrition and faculty impact in reducing attrition were factor analyzed into subscales for hypothesis testing. The data were analyzed statistically using one-way, fixed effects analysis of variance. Of the nineteen statistical tests, eight yielded significant findings. In considering the specific factors which affect student attrition in these colleges, most attrition was perceived to be attributable to lack of interest or moti- vation, inadequate academic preparation, financial problems, low academic ability, and lack of personal/emotional adjust— ment to college. Although all respondent groups perceived that factors concerning student ability, interest and preparation were the most important factors in attrition, faculty members in both high and low attrition colleges cited these factors to be of greater importance than did administrators. In the high attrition colleges, there was signifi- cantly more concern about the quality of the academic ser- vices provided to students (such as adequacy of curricular Richard H. Mullendore offerings and quality of academic advising) than in the low attrition schools. Conversely, in the low attrition colleges, social factors (such as insufficient extracur- ricular activities) were perceived as more important causes of attrition than in the high attrition institutions. No differences were found between high and low attrition colleges regarding the amount of contact that faculty state they have with students outside the class- room. In fact, the two groups were virtually identical in all role capacities studied. In this study, faculty and administrators were asked their perceptions of the potential impact that selected faculty-student interactions may have in reducing student attrition. Three role capacities were identified by all respondent groups as having the greatest potential impact in reducing attrition. They were: academic advising, out of class instruction, and career advising. Comparisons of the responses of faculty and administrators from low attrition colleges with those from high attrition schools yielded no significant differences regarding the impact of faculty in reducing attrition. However, comparisons between faculty and administrators disclosed significant differences in both low and high attrition institutions. In each case, administrator ratings of potential faculty impact were higher than those of faculty. Richard H. Mullendore The findings of this study suggest the need for further research regarding the causes of attrition at individual colleges, and the need for more open communi- cation between faculty and administrators concerning mutual expectations in a retention program. TABLE OF CONTENTS Chapter I. II. III. IV. INTRODUCTION 0.00.00.00.00...OOOOOOOOOOOOOOOOOOOO Purpose of the Study .......................... Questions for Investigation ................... Need for the Study ............................ Hypotheses .................................... Abbreviated Summary of Related Literature ..... Definition of Terms ........................... Limitations of the Study ...................... Assumptions ................................... Population of the Study ....................... Methodology ................................... Analysis and Treatment of Data ................ Organization of the Study ..................... REVIEW OF THE LITERATURE ........................ Why Study Attrition? ..................... ..... The Problem of Definitions .................... Criticisms of Attrition Research .............. Factors Associated with Attrition ............. Rates of Attrition ............................ Faculty Impact upon Student Persistence ....... Summary ....................................... MTHODOLOGY OOOOOOOOOOOIOOO..OOOOOOOOOOOOOOOOOOOO Hypotheses .................................... Population .................................... Instrumentation ............................... Use of Instrument and Collection of Data ...... Statistical Analysis .......................... PRESENTATION AND ANALYSIS OF DATA ..... ..... ..... Introduction .................................. Hypotheses to be Tested ....................... Treatment of the Data ......................... Description of the Population--Demographic Information ................................. General Institutional Perceptions--Summary of Responses ................................ ii l'--| Ideahnahahua ~Jaun¢>¢¢ohamm40\mwn 18 19 20 22 24 29 31 38 40 4O 49 52 55 58 58 59 60 64 69 Chapter Factors Associated with Attrition ............. Out of Class Faculty-Student Contact .......... Impact of Faculty Activities in Reducing Student Attrition ........................... Summary ................................. ...... V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ....... Summary of the Development of the Study ....... Findings....................................... Conclusions and Discussion..................... Implications of the Study ..................... Speculations .................................. Recommendations for Further Research .......... Concluding Statement .......................... MPENDIXAO......0.0.00.0.....OOOCOOOOOOOOO0.0.0.0.... APPENDIXB0.0.0....00.0.0...0.0.....OOOOOOOOOOOOOOOOOO BIBLIOGRAPHY iii 74 88 91 99 103 104 107 114 123 127 130 133 135 139 144 4.4. 4.5. 4.6. 4.7. 4.8. LIST OF TABLES Summary of College Characteristics . ...... ...... Response Rates (Usable Questionnaires Returned) OOOOOOOOOOOOCOOCOOO0.00.00...0...... Graduation/Attrition Rates 1972-76, 1973-77, 1974-78, 1975-79 one.ooooooooooooooooooooooooo "Number of years you have held current, primary responsibility at this institution:" (Summary of responses, in percentages) .. ..... "Highest degree you have earned:" (Summary of responses, in percentages) ................ "Faculty: Are you tenured? (If your insti- tution does not grant tenure, please leave this item blank.)" (Summary of responses, in percentages) ...... ........................ “Faculty: Number of students for whom you are serving as academic advisor:" (Summary of responses, in percentageS) O O O O I O O O O O O O O O O O O I 0 "Faculty and Administrators: Your total number of years of full time postsecondary teaching experience:" (Summary of responses, in percentages) ...... ..... ...................... "I consider student attrition to be a problem at this institution." (Summary of responses, in percentages) ................... "I estimate that the percentage of students who leave this college prior to completion of a degree is:" (Summary of responses, in per- centages) .................................... "The primary responsibility for reducing student attrition should rest with administrators." (Summary of responses, in percentages) ....... iv 44 54 56 66 67 67 68 69 7O 71 72 4.10. 4.12. 4.13. 4.14. 4.16. 4.17. "Faculty members can play a role in reducing student attrition without lowering academic standards." (Summary of responses, in per- centageS) 00.0.0.0.....OOOOOOCOOOOOOOOOOO "This institution has a formal program for reducing student attrition." (Summary of responses, in percentages) .............. "The program for reducing student attrition this institution is effective.“ (Summary of responses, in percentages) ........... at "Faculty members are involved in the program for reducing attrition." (Summary of responses, in percentages) .............. Analysis of Variance for Subscale 1, Hypothesis 1 ................ ............ Analysis of Variance for Subscale 1, Hypothesis 2 ......... . .................. Analysis of Variance for Subscale 1, Hypothesis 3 .......... .... ..... . ........ Analysis of Variance for Subscale 2, HypotheSi81 OOOOOOOOOOOOOOOOOOOOO ....... Analysis of Variance for Subscale 2, Hypothesis 2 ..... ......... .... ...... .... Analysis of Variance for Subscale 2, HypotheSis3 ..OOOOOOOCOOOOOOOOOO ........ Analysis of Variance for Subscale 3, HypotheSj-Sl OOOOOOOOOOOOOOOO0.0.0.000... Analysis of Variance for Subscale 3, HypotheSisz .0...OOOOOOOOOOOOOOIOOOOOOO. Analysis of Variance for Subscale 3, HypotheSis3 0.0.0.0......OOOOOOOOOOOOOOO Analysis of Variance for Subscale 4, HypotheSiS]. OOOIOOIOOOOOOOOOOOOOOOO ..... Analysis of Variance for Subscale 4, Hypothesis 2 ........... ................. Analysis of Variance for Subscale 4, HypotheSiSB 0.0.0.000... ......... O ...... 72 73 74 74 76 77 78 79 80 81 83 84 85 86 87 88 Analysis of Variance for Faculty-Student contact 0.0.0.000...OOOOOOOOOOOIOOOOOOOOOOO0.0 Summary of Responses, Faculty Out of Class Activities (in percentages) ........ .......... Analysis of Variance for Subscale 5, HypotheSiss OOOOOOOOOOOOOOOOOOOO 0000000000000 Analysis of Variance for Subscale 5, Hypothesis 6 ........... . ..... . ............. .. Analysis of Variance for Subscale 5, HypotheSiS7 ..... 0...... ..... .0 ...... .0. ..... Analysis of Variance for Subscale 6, Hypothesis 5 ......... ........ .. .............. Analysis of Variance for Subscale 6, HypotheSiSG 00....OOOOOOOOOOOOOOOOOOO ........ Analysis of Variance for Subscale 6, Hypothesis 7 ............ ..... . ............... vi 90 92 94 95 96 97 98 99 CHAPTER I INTRODUCTION Research available on college student attrition is abundant and indicates that the dropout problem has long been a concern in higher education. The topic has received increased attention recently because of projections for declining enrollments and fears for institutional survival.1 Despite the extensive literature on dropouts from higher education, little is actually known about the nature of the drOpout process, and much of the literature to date has been strongly criticized.2 First, investigators have been overly concerned with the characteristics of dropouts, such as personality and attitudinal variables. Studies of this type have been found to be unclear, inconsistent and non-generalizable.3 Second, previous research has been too limited in sc0pe, has failed to develOp a theoretical or conceptual 1John C. Neddy, "A Review of Research Concerning College Student Attrition," Mimeographed for the Retention Improvement Task Force, (State University of New York at Buffalo, 1977). 2Vincent Tinto, "Dropout from Higher Education: A Theoretical Synthesis of Recent Research," Review of Educational Research, 45 (Winter, 1975), pp. 89-125. 3K.A. Feldman and T.M. Newcomb, The Impact of College on Students: An Analysis of Four Decades of Research, Vol. 1 (San Francisco: Jossey-Bass, 1969). 2 perspective to the problem, and has neglected to present solutions.4 In recent years a few investigators have attempted to counteract the criticisms of past research. William Spady has developed a sociological model of college with- drawal based upon Durkheim's theory of suicide. The model emphasizes the notion of congruence between the student and the institution.5 Other investigators have also studied the interaction between the student and the environment of the institution, however, these studies have not revealed conclusively the specific factors which influence students to leave college or remain enrolled. As a result, there is limited knowledge concerning how these factors may be con- trolled by those who have a vested interest in retaining students.6 Interest in, and concern for student attrition is important to colleges for several reasons. First, attrition affects colleges financially, which reduces the efficiency of operation. Whenever students leave the college, dollars 4Alexander W. Astin, Preventing Students from Dropping Out, (San Francisco: Jossey-Bass, 1975). 5William G. Spady, "Dropouts from Higher Education: Toward an Empirical Model," Interchange, 2 (No. 3, 1971), pp. 38-62. 6Robert L. Husband, "Significant Others: A New Look at Attrition,“ Paper presented at Seventh Annual Meeting on Future Solutions to Today's Problems, sponsored by the Association for Innovation in Higher Education, (Philadelphia, February, 1976), ERIC-ED 124 056. 3 leave the income side of the budget in the form of tuition, fees, and housing.7 A second reason for institutional concern is that attrition tends to break down the continuity and level of maturity of the student body, particularly in the small college. As attrition takes its toll, there continues a large population of underclassmen every year as compared to a small population of upperclassmen. This student mix many times impedes the continuity and stability of a student body and militates against a maturing peer influence.8 Third, institutional credibility is also affected by attrition, in that the college with a high attrition rate is often criticized for doing a poor job. Nongrad- uates erode institutional capacity and credibility, while graduates become a credit.9 For those students who drop out of higher edu- cation, the cost is sometimes high, both personally and psychologically. College graduates traditionally have more career opportunities, more job security, better working conditions, and higher job satisfaction than nongraduates. Psychologically, there is much diversity in leavers' feelings toward themselves and the colleges from which they 7John Summerskill, "Dr0pouts from College," in N. Sanford (Ed.), The American College, (New York: John Wiley & Sons, 1962). 8Husband, 22. cit., p. 1. 9Robert G. Cope and William Hannah, Revolving College Doors: The Causes and Consequences of Dropping Out, Stopping Out, and Transferring, (New York: John Wiley & Sons, 1975). 4 withdraw. However, many students who drop out of college feel disappointed, disillusioned and resentful toward them- selves.10 These administrative, economic and psychological concerns about attrition have generated a considerable amount of research, however, most of the studies consist of tabulations and statistical analyses of previous school records and related data. Investigations of attrition rate by type of college, size of high school graduating class, and age of student upon matriculation can be found in large numbers. In addition, there have been concerted efforts to determine the relationship between attrition and high school grades, aptitude tests, family financial status and many other variables. "Research of this type has not been ade- quate to the development of better understanding of college student dropouts nor has it succeeded in substantially reduc- ing high attrition rates."11 In fact, Cope and Hannah report that since the first national study on attrition rates, conducted in the 1930's, the rate of dropout has remained relatively constant.12 Recent research seems to indicate that it may be the degree of lack of fit between the student and the college that accounts for most of the dropping out, stopping out and 10Cope and Hannah, op. cit. 11Summerskill, pp. cit., p. 629. 12Cope and Hannah, op. cit. 5 transferring. The key to reducing attrition, therefore, may be the development of a more responsive relationship between the student and the environment of the institution.13 Purpose of the Study Administrators at many colleges have initiated programs which are designed to reduce student attrition, and several investigators have stated that faculty members can make positive contributions to these programs. If faculty members are to participate in retention efforts, then faculty and administration perceptions concerning faculty involvement in reducing student attrition need to be understood. In this study, the investigator compared the perceptions of faculty and administrators at high attrition colleges with the perceptions of faculty and administrators at low attrition colleges. This comparison was made to determine whether or not there were differences in adminis— trator and faculty perceptions concerning selected aspects of student attrition at high and low attrition institutions. The focus of this study was on: (1) determining the factors, as perceived by faculty and administrators, which affect student attrition; (2) determining the differences in the amount of contact faculty state they have with students outside the classroom; (3) determining the impact, as perceived by faculty and administrators, 13Cope and Hannah, op. cit. 6 that faculty may have in reducing student attrition; and (4) comparing the perceptions of faculty members with those of administrators to determine whether or not faculty and administration share similar perceptions concerning student attrition. Questions for Investigation 1. Do faculty and administrators at colleges with high attrition rates differ from faculty and administrators at colleges with low attrition rates in their perceptions con- cerning the factors which affect student at— trition in their institutions? 2. Do faculty members differ from administrators in their perceptions concerning the factors which affect student attrition in high and low attrition institutions? 3. Do faculty members at colleges with high attrition rates differ from faculty at colleges with low attrition rates in the amount of out of class contact they state they have with students? 4. Do faculty and administrators at colleges with high attrition rates differ from faculty and administrators at colleges with low attrition rates in their perceptions concerning the impact faculty can have in reducing student attrition? 5. Do faculty members differ from administrators in their perceptions concerning the impact which faculty can have in reducing student attrition in high and low attrition colleges? Need for the Study We suspect that persistence in college requires the personal touch that only dedicated professors can give. Evidence, both personal and other, reveals that such dedication exists in abundance across America, but . . . it is largely mis- directed. Teacher roles must become more person- alized in order to satisfy both the academic and personal needs of students, thus relieving the .7 tension between people and institutions that cause separation. The call for faculty involvement in student retention efforts is repeated throughout the literature, but the literature appears to bexvoidflofflresearch studies which assess and compare faculty and administration per- ceptions concerning faculty involvement in reducing student attrition. Since many institutions of higher education are currently engaged in retention efforts, the perceptions of faculty and administrators need to be understood in order to determine the appropriate functions for faculty in a viable retention program. Hypotheses For the purpose of statistical analysis, the questions for investigation are restated in research hypoth- esis form. 1. There are significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. 2. There are significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. 3. There are significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. 14Cope and Hannah, 92. cit., p. 45. 8 4. There are significant differences in the stated amount of out of class student contact that faculty have in high and low attrition colleges. 5. There are significant differences in the per- ceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact fac- ulty can have in reducing student attrition. 6. There are significant differences in the per- ceptions of faculty and those of administra- tors in high attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. 7. There are significant differences in the per- ceptions of faculty and those of administra- tors in low attrition colleges concerning the impact faculty can have in reducing student attrition. Abbreviated Summary of Related Literature Several investigators have suggested that faculty out of class activities may have some impact upon student persistence, however, few studies exist which can empiri- cally support this notion. Although nearly everyone feels that faculty-student interaction is desirable, and although many institutions are making concerted efforts to in- crease faculty student contact . . . little is known about the actual benefits that can be expected to accrue to either faculty or students from increased interaction beyond the classroom.15 Spady suggests that a student's social and aca- demic integration is enhanced through contact with faculty, and therefore, institutional commitment increases.16 15R.C. Wilson, J.G. Gaff, E.R. Dienst, L. Wood, and J.L. Bavry, College Professors and Thei£_Impact on Students, (New York: John Wiley & Sons, 1975), p. 31. 16 Spady, gp. cit. 9 Pascarella and Terenzini support the notion of integration and its relationship with student persistence. "Other things being equal, the higher the levels of academic and social integration the less likely the student is to vol- untarily leave the institution."17 Faculty-student interaction has been investigated in several ways. Vreeland and Bidwell studied the intimacy and frequency of this interaction and found that faculty have a positive impact upon students' attitude and value change.1gmeams n examined the differences between faculty in the natural sciences and those in the social sciences concerning their orientations toward students. She found that those in the social sciences, in general, felt a stronger extracurricular commitment toward students than did natural science faculty members.19 Gekoski and Schwartz compared the attitudes of dropouts and persisters concerning their interactions with faculty. They found that faculty had a favorable impact upon persisters, but received significantly 17E.T. Pascarella and P.T. Terenzini, "Patterns of Student-Faculty Informal Interaction beyond the Classroom and Voluntary Freshman Attrition,“ Jggrnal of Higher Education, 48 (September/October, 1977), p. 541, 18R.S. Vreeland and C.E. Bidwell, "Classifying University Departments: An Approach to the Analysis of Their Effects upon Undergraduates' Values and Attitudes," Sociology of Education, 39 (1966), pp. 237-254. 19Z. Gamson, "Utilitarian and Normative Orientations toward Education," Sociology of Education, 39 (1966). PP. 46-73. 10 lower ratings from the withdrawal group.20 Faculty have many types of interactions with students in and out of the classroom. Webb suggests that . . . within an interactional framework, there exists a set of behaviors or roles that faculty are expected to perform in their interactions with students. To the extent that the faculty does not fulfill these roles, they contribute to the problem of dropout.21 One faculty role often mentioned in the literature is academic advising. Studies report differences in grade point averages, study skills, retention, and satisfaction with college as a result of special advising programs.22'23'24 Like most available research on retention, however, the results of these studies are inconclusive and somewhat contradictory. Investigators have suggested other roles for faculty which may have some impact on student persistence 20N. Gekoski and S. Schwartz, "Student Mortality and Related Factors," Journal of Educational Research, 54 (January, 1961), pp. 192-194. 21S.C. Webb, "Faculty Contributions to Dropouts," in 0. Milton (Ed.), Proceedings: A Conference on Student Retention in Tennessee Colleges and Universities, (Univer- sity of Tennessee at Knoxville, March, 1966), ERIC-ED 044 084, p. 11. 22Jack E. Rossmann, "Released Time for Faculty Advising: The Impact on Freshmen," Personnel and Guidance Journal, 47 (December, 1968), pp. 358-363. 23Jack E. Rossmann, "An Experimental Study of Faculty Advising," Personnel and Guidance Journal, 46 (October, 1967), pp. 160-164. 24C.G. Morehead and J.C. Johnson, "Some Effects of a Faculty Advising Program," Personnel and Guidance Journal, 43 (October, 1964), pp. 139-144. 11 including, curriculum development, out of class instruction, career counseling, personal counseling, social interaction, 25,26 The and practicum and independent study supervision. extent to which these activities affect student retention efforts, however, remains unclear. Do faculty members agree that they should become involved in retention efforts? In what ways should faculty become involved? What faculty functions do administrators perceive as important in reducing student attrition? Although an extensive review of the related literature will follow, no studies were found which directly relate to these questions. Definition of Terms The following terms are defined by the investigator for use in this study: Administrator. Any full-time employee of a given college whose current, primary responsibility is to perform one or more executive or managerial functions (a list of administrator functions is included in Appendix A). Attrition. Any loss of students from a given college as a result of dropping out, stopping out, trans- ferring, academic failure, or disciplinary suspension or expulsion. Nonpersistence. 25Webb, 22. cit. 26Pascarella and Terenzini, op. cit. 12 Attrition rate. The percentage of full-time freshmen who enroll in a given college, but who do not graduate from that college four years later. Dropout. Any student who has, at some time, enrolled in a given college, but has permanently failed to either continue to enroll or obtain a degree from that college. Faculty member. Any full time employee of a given college whose current, primary responsibility is classroom instruction. High attrition college. A college whose mean attrition rate is greater than 60 percent. Low attrition college. A college whose mean at- trition rate is less than 50 percent. Perception. Awareness, discernment, insight or feeling achieved through intellectual means. Retention. A college's intentional or unin- tentional retainment of students through continuous enroll- ment. The absence of attrition. Persistence. Limitations of the Study The investigator recognizes that certain limi- tations exist in most research. In this study, the reader should be aware of the limitations listed below: 1. Only selected aspects of the problem of student attrition are investigated in this study. The field of attrition/retention research is quite broad, and no attempt has been made to investigate the entire field. This study 13 is concerned with faculty and administrator perceptions concerning the factors which influence attrition in their institutions, the amount of out of class contact faculty have with students, and the potential impact faculty may have in helping reduce attrition. Comparisons are made between faculty and administrators in high attrition col- leges, and between faculty and administrators in low attrition colleges. Comparisons are also made between high and low attrition colleges considering both faculty and administrator responses to the survey instrument. 2. The availability of reliable, longitudinal attrition data from the participating institutions is limited due to limited institutional record-keeping pro- cedures, lack of retrievable data, and differing institutional definitions concerning classifications of students. 3. Anonymity has been guaranteed for all respon- dents and all institutions; therefore, descriptions of each institution must be somewhat restricted. 4. Although admissions standards and certain other factors concerning each college in this study are similar, differences are likely to exist in the composition of each student body and the nature of each institution. 5. The institutions included in this study were not selected randomly. Therefore, generalizations from the results presented in Chapter IV should be made only to the extent that individual institutions closely resemble the 14 colleges in this study. 6. In order to accommodate for the limited avail- ability of reliable institutional data, very narrow defi- nitions of "dropout" and "attrition rate" have been utilized for this study. Assumptions For the purposes of this study, the following assumptions have been made: 1. Each participating college was a member of the Strengthening Institutions Through Improving Retention (SITIR) project at the time of this study. As a result of their participation in this project, it is assumed that some of the administrators at these colleges may be more aware of attrition causes, problems, and solutions than faculty and other administrators at the same institutions. 2. Each participating college had a Student Retention Officer (SRO) who was instructed to administer the questionnaire. It is assumed that each SRO administered the instrument to the appropriate persons and followed up as instructed in order to maximize the response rate. 3. The results of this study are predicated upon the honesty and candidness of the respondents. Population of the Study The population for this study included faculty and administrators from twelve small, private, four-year, liberal arts colleges. Each of these colleges was a 15 participant in the Strengthening Institutions Through Improv- ing Retention (SITIR) project at the time of the study.27 Based upon institutional data provided for this study, four -of these colleges were defined as low attrition institutions and eight were defined as high attrition institutions. Due to the small size of these colleges (enrollments under 1000 students), all faculty and administrators in each institution were surveyed. A brief description of each college is found in Chapter III. Methodology To obtain the necessary data for this study, a questionnaire (Faculty and Administration Perceptions Concerning Faculty Involvement in Reducing Student Attrition) was developed. This instrument was pilot-tested by faculty and administrators in October, 1979, at the University of Charleston (West Virginia), and apprOpriate modifications were made as a result of the pilot test. The questionnaires were administered in November and December, 1979, to all faculty and administrators at each of the twelve colleges by the Student Retention Officer 8 (SRO).2 Completed questionnaires were returned to the 2‘7The SITIR project was funded under Title III of the Higher Education Act of 1965, as amended. Title III is also known as the Strengthening Developing Institutions Program. The SITIR project began in the fall of 1978, under the direction of Dr. Samuel Barnett, Deputy Director of the Conference of Small Private Colleges, Princeton, New Jersey. 28The Student Retention Officer (SRO) was a desig- nated representative from each college, who was responsible for coordination of the on-campus retention efforts for the SITIR project. 16 SITIR office in late December, 1979. For ease in responding, the instrument was divided into the following parts: (1) general institutional per- ceptions, (2) student-related factors in attrition, (3) institutional factors in attrition, (4) faculty activ- ities outside the classroom, (5) impact of faculty activities in reducing attrition, and (6) demographic information. Analysis and Treatment of Data For statistical analysis of the data, the follow- ing techniques were used: 1. Exploratory factor analysis. Uses Fortran program PACKAGE, subprogram FACTOR, for partitioning selected questionnaire items into clusters.29 2. Multiple groups confirmatory factor analysis. Uses Fortran program PACKAGE, subprogram MGRP, to analyze 30 user-defined cluster solution. 3. Condescriptives. SPSS program used to provide descriptive information, including measures of central tendency.31 29J.E. Hunter and D.W. Gerbing, "Unidimensional Measurement and Confirmatory Factor Analysis," Occasional Paper No. 20, (East Lansing, Michigan: Institute for Research on Teaching, May, 1979). BOIbid. 31N.H. Nie, D.H. Bent, and C.H. Hull, Statistical Package for the Social Sciences, (New York: McGraw-Hill Book Co., 1970). 17 4. Crosstabulations. SPSS program used to pro- vide descriptive information, including number and percen- tage of responses for each questionnaire item.32 5. One-way analysis of variance. SPSS program used to determine F-ratios, significance, and strength of association.33 A comprehensive review of the design and method- ology is found in Chapter III. Organization of the Study Chapter II is devoted to a review of pertinent literature concerning selected aspects of student attrition and the potential for faculty impact upon retention. The design and methodology used in collecting_and analyzing the data is presented in Chapter III. The results of the analysis of the data are presented in Chapter IV. Chapter V contains the summary, conclusions, discussion of results, and implications for further study. 32Nie, Bent and Hull, pp. cit. 331bid. CHAPTER II REVIEW OF THE LITERATURE The problem of college student attrition has long been of concern to American educators, and has been a sub- ject of continual rediscovery in the research literature. Out of every ten students who enroll in a college, only four will graduate from that college four years later. Only six out of ten students will eventually complete a degree at some baccalaureate-granting institution.1 In terms of sheer numbers, the attrition prob- lem deserves the attention of those interested in and affiliated with institutions of higher edu- cation. From the institutional point of view, attrition has a heavy impact on institutional operations and finance. From the student's point of view, the effect of dropping out, although difficult to guage, is also another important aspect of the attrition problem. The need to understand this phenomenon becomes more urgent every day. The purpose of this chapter is to report previous related and relevant research; thus, a review of the liter- ature related to the following areas of college student attrition is presented: (1) why study attrition, (2) the 1Robert G. Cope, "Why Students Stay, Why They Leave," in L. Noel (Ed.), Reducing the Dropout Rate, New Directions for Student Services 3 (San Francisco: Jossey- Bass, 1978). 2T.J. Pantages and C.F. Creedon, "Studies of College Attrition: 1950-1975," Review of Educational Research, 48 (Winter, 1978), p. 49. 18 19 problem of definitions, (3) criticisms of attrition research, (4) factors associated with attrition, (5) rates of attrition, and (6) faculty impact upon student persis— tence. Other areas of research which are not directly related to the purposes of this study have not been included. Why Study Attrition? With downward or stable enrollment trends fore- cast for the future, the continuing problem of attrition has become a more pressing concern for educators because of its relationship to enroll- ment problems. Colleges face a decline in the number of potential freshmen and therefore need to retain those students already enrolled to maintain their financial stability and the breadth and strength of their academic programs. The problem of attrition is not_a new one for colleges. Research on college dropouts has a history of at least fifty years. "It has gained a position of promi- nence in recent years, however, because the loss of stu- dents is now critically linked to the issue of survival for most colleges and universities."4 Neddy also suggests that survival is a genuine concern. "Coupled with pro- jected declining enrollments and fears for institutional survival, attrition and retention are becoming issues of concern to chief campus administrators."5 3Carol H. Shulman, "Recent Trends in Student Retention," ERIC Higher Education Currents, (May, 1976), p. 1. 4Husband, pp. cit., p. 1. 5Neddy, pp. cit., p. l. 20 Currently, there is anxiety not only about declining enrollments, but also for recent shifts in post- secondary enrollment patterns. According to Noel, The most significant shifts have been from pri- vate to public institutions, from four-year to two-year institutions, from baccalaureate to vocational/technical institutions, from liberal arts to specialized majors, and from full-time to part-time enrollment. If Noel's assessment of the shifts in enrollment patterns is accurate, the need for this study becomes increasingly apparent, since it focuses upon selected four- year, private, liberal arts, baccalaureate institutions. The Problem of Definitions Although a considerable amount of research con- cerning dropouts has been generated, it is difficult to compare studies or to determine the actual causes and extent of the problem because the term "dropout" has been so variously defined.7 Panos and Astin state that "the results of many attrition studies are not comparable because they in fact deal with different phenomena."8 "The major definitional problem is the tempor- ariness of any classification of a student as a drOpout. 6Lee Noel, "College Student Retention--A Campus- Wide Responsibility," Journal of the National Association of College Admissions Counselors, 21 (July, 1976), P. 33. 7Shulman, pp. cit. 8R.J. Panos and A.W. Astin, "Attrition Among College Students," Amegican Educational Research Journal, 5 (January, 1968), p. 70. 21 No categorization will be wholly satisfactory until all students either obtain their degrees or die without "9 In addition, it is difficult to approp- receiving them. riately define the problem because leaving college before graduation takes on a number of different, occasionally distinct forms. For example, . . . some students may enter college with no intention of completing the baccalaureate pro- gram; for them dropping out is an expression of an original plan. Among the students who originally plan to complete their academic pro- gram, the reasons for dropping out are complex and overlapping. "In its strictest definition, from the institution's perspective, a dropout is a student who does not graduate from the college in which he enrolled as a 11 This definition, however, is very narrow and freshman." does not account for those students who take longer than four years to complete a degree or those who transfer to another institution; nor does it consider differences between voluntary and nonvoluntary withdrawal. In an attempt to provide some resolution to the definitional problem, Panos and Astin suggest that ". . . it is important in any research on dropouts that 'dropout' be unambiguously defined, and that the definition make sense with regard to the problem being investigated 9Astin, pp. cit., 1975, pp. 5-6. 10Cope, pp. cit., p. 2. 11Shulman, pp. cit., p. 1. 22 and to the possible applications of the findings."12 Criticisms of Attrition Research A considerable amount of criticism has been directed toward previous attrition/retention research. "Retention research over the past fifty years has not been productive for the most part. We haven't learned anything new and we haven't used what we know."13 Part of the prob- lem appears to be that " . . . the disappointing status of the knowledge and understanding produced by attrition research undoubtedly reflects the large number of relevant variables, their complexity, and their interdependence."14 Several types of criticism about research on dropouts are mentioned in the literature. First, the nature of the dropout phenomenon is somewhat unclear. Marks states that ". . . with few exceptions the research on college dropouts is characterized by a lack of an adequate concep— tual base and a reliance on ex post facto methodology."15 Cope and Hannah suggest that the problem with previous research is that it has not penetrated beyond the collection 12Panos and Astin, pp. cit., p. 70. 13Noel, pp. cit., p. 34. 14Donald P. Hoyt, "A Retrospective and Prospective Examination of Retention-Attrition Research, " in L. Noel (Ed.), _p. cit., p. 78. 15Edmond Marks, "Student Perceptions of College Persistence, and Their Intellective, Personality and Performance Correlates," Journal of Educational Psychology, 58 (No. 4, 1967), P. 210. 23 of simple demographic data, and, as such, has provided an oversimplification of the dynamics of attrition.16 A second criticism of the literature on dropouts is that there has been a lack of emphasis on developing models or finding methods to explain and reduce attrition rates. Astin indicates that most of the research has con- sisted of counting, describing and classifying dropouts 17 Tinto rather than seeking solutions to the problem. advocates the ". . . development of theoretical models that seek to explain, not simply to describe, the processes that bring individuals to leave institutions of higher edu- cation."18 Knoell suggests that existing research is microcosmic in nature, and that it fails to look at more general phenomena, such as the initial decision to attend college. By looking at more general considerations, ". . . attrition may then be viewed as one type of resul- tant of the interaction of student, institution, and system variables."19 Third, there is criticism of the emphasis upon the notion that dropping out is a negative behavior. 16Cope and Hannah, pp. cit. 17Astin, pp. cit., 1975. 18Tinto, pp. cit., p. 89. 19Dorothy M. Knoell, "A Critical Review of Research on the College Dropout," in Pervin, Reik, and Dalrymple (Eds.), The College Dropout and the Utilization of Talent, (Princeton, N.J.: Princeton University Press, 1966), p. 63. 24 Dropouts are often called "casualties" or "non-survivors." The rate of attrition is often referred to as the "mortality rate." "These negative attitudes have not only had the effect of placing the dropout under increased pressure but have also often determined policies, particularly those concerning readmission and even concerning transfer stu- dents."20 Factors Associated with Attrition "Most colleges know very little about why their students withdraw."21 In many colleges, records concerning reasons for withdrawal are nonexistent. In others, the records indicate only superficial causes of attrition with categories such as academic, personal and financial. In reviewing research studies which were specifically devel- oped to identify the factors associated with attrition, one often finds inconclusive and contradictory results.22 In a study of student perceptions of college persistence, Marks found that: When referring to themselves (the subjects) spoke mostly of external or personally acceptable causes, e.g., poor preparation, lack of funds, present school not first choice. Reasons given for the dropout of another person were imputed almost 20L.A. Pervin, L.E. Reik, and W. Dalrymple (Eds.), "The Dropout in Conflict with Society," pp. cit., p. 18. 21Cope and Hannah, pp. cit., p. 7. 22Ibid. 25 entirely to personal weaknesses, e.g., immaturity, maladjustment, lack of self-discipline. Students report many reasons for leaving college prior to receiving a degree. Alexander Astin conducted a major study of the characteristics of students who enrolled in college in 1968. In 1972, he conducted a follow-up on both persisters and non-persisters. Students who had drOpped out of school were asked to check up to three of the twelve listed reasons for dropping out. Astin found that ". . . the most frequent reasons for dropping out for both men and women are boredom with courses, financial difficul- ties, dissatisfaction with requirements or regulations, and change in career goals."24 It should be noted that twenty- eight percent of the respondents checked “some other reason," which suggests that certain important items may have been missing from the list.25 In a study of freshmen who left Temple University after their first semester, Gekoski and Schwartz found the chief reasons for withdrawing to be: change of interest or plans, dissatisfaction with courses and/or university, financial difficulties, college adjustment problems, and job interference.26 23Marks, pp. cit., p. 214. 24Astin, pp. cit., 1975, p. 15. 25Ibid. 26Gekoski and Schwartz, pp. cit. 26 The actual reasons for dropping out generally are quite complicated in that several factors are often involved simultaneously. For example, academic difficulty has been found in approximately one-third of college dropouts, and motivational problems have been ascribed to an even larger number. It has been found, however, that these problems often interplay with other problems which eventually lead to withdrawal.27 Not only do students leave college due to their own inadequacies or problems, but also because of factors within the college environment. Cope and Hannah show, through their research, how the feelings, experiences and perceptions of dropouts influence decisions to withdraw. Students leave for many reasons; insufficient intellectual challenge, major program not offered, location of college, problems with faculty and administrators, inadequate financial aid, stifling social climate, change in personal goals, value conflicts, impersonal academic environment, and housing arrangements are some of the reasons given.28 Astin's research has been oriented toward predict- ing a student's chances of persistence based on background, ability, lifestyle, goals and more. He has found students' goals and expectations concerning their own persistence play a moderate role in predicting which students will 27Summerski11, pp. cit. 28Cope and Hannah, pp. ci . 27 stay.29 In another study of student perceptions, Marks found a much stronger relationship between expectations of dropping out and the actual behavior. He concludes that ". . . if you want to know whether a student is a potential college dropout, a good starting place is to ask him."30 Currently, investigators appear to be advocating the notion of "fit" or congruence between student and institution in their search for ways to reduce student attrition. Should the student attend a large college or a small one? Should the student attend a single-sex or coeducational institution? Should the student live on campus, at home or in an apartment? Should the student attend a college which is associated with his religious faith? Should the student attend a public institution, or the more expensive private one? What effect do these choices have upon the student's chances of dropping out? Although each of the above questions may have some influence upon a student's decision to persist or leave, it is important to note that the best predictors of probable retention are the high school grade point averages and scores on standardized tests.31 In considering the colleges which participated in this study, some differences do exist among these 29Astin, 92. cit., 1975. 30Marks, pp. cit., p. 219. 31Cope, pp. cit. 28 institutions which should be noted and related to previous research. Two of the colleges, for example, are classified as women's colleges (although a small percentage of males do attend as commuters). According to Astin, there is some difference in dropout rates for women attending women's colleges versus those in coeducational institutions. This difference may be attributable partially to the fact that women's colleges tend to be more selective in their admissions policies. Also, there are greater opportunities for marriage in coeducational colleges, which is associated with an increase in the expected dropout rate for women.32 All of the colleges in this study still maintain, at least, informal ties with some religious denomination, even though a few classify themselves as independent institutions. Dropout rates are expected to be lower at these institutions, especially those associated with the Catholic Church.33 A student's chances of dropping out are decreased if he lives in a residence hall, particularly during the 34 In three of the colleges in this study, freshman year. less than thirty percent of the students reside on campus, so the dropout rate would be expected to be somewhat higher, other factors being equal, at these colleges. 32Astin, 92. cit., 1975. 331bid. 34Ibid. 29 Despite the extensive amount of research about the causes of student attrition, very little is known that is applicable to each individual campus. "Studies employing different definitions of the dropout, employing divergent variables, carried out in diverse institutions, and utilizing dissimilar samples and research techniques are 35 virtually impossible to synthesize." Besides, " . . . a retention agent on one campus might be an attrition agent on another."36 In this study, the investigator ascertained faculty and administrator perceptions of the causes of attrition on their respective campuses in order to see if any patterns exist within similar types of institutions. Rates of Attrition "American higher education's dropout rate is 37 In the persistent, serious, and difficult to reduce." first national study concerning the rate of attrition in American colleges, Iffert found that 40 percent of all students graduate from the institution of original enroll- ment in four years, 51 percent graduate from some institution 35Cope, pp. cit., p. 7. 36Noel, pp. cit., p. 34. 37Shulman, pp. cit., p. 1. 30 in four years, and 60 percent eventually graduate.38 It appears that different types of colleges have varying average graduation rates. Cope and Hannah found that less selective institutions have the highest rates of attrition, and that attrition rates are generally higher at city and state supported schools than at private colleges. They also report that of those students who leave, over half do so before their second year.39 In a follow-up of students who left the University of Illinois, Eckland suggests that estimates of attrition rates have been exaggerated. He found that, after ten years, approximately 70 percent of the students in his study had either graduated or were potential graduates.40 Since this study is concerned with small, private, church-related, liberal arts colleges, it may be helpful to report the findings of Cope and Hannah regarding attrition rates at these institutions. They found the first year attrition rate to be 26.8 percent of all full time freshmen, and a 51.7 percent attrition rate after four years. They also note that of the twelve colleges studied, 38Robert E. Iffert, Retention and Withdrawal of College Students, U.S. Office of Education Bulletin 1958, No. 1 (Washington, D.C.: U.S. Government Printing Office, 1957). 39Cope and Hannah, pp. cit. 4OBruce K. Eckland, "College Dropouts Who Came Back," Harvard Educational Review, 34 (No. 3, 1964), pp. 402-420. 31 the one with the highest rate of attrition has recently closed.41 Faculty Impact upon Student Persistence Student retention should be looked upon as a by-product of quality programs and services, not as a goal in itself. However, all faculty and staff on the campus have a responsibility to contribute in some way to retention efforts.42 A few investigators have suggested that faculty play a vital role in increasing students' satisfaction with and commitment to college, and thus aid retention. Student-faculty interaction has a stronger relationship to student satisfaction with the college experience than any other involvement variable or, indeed, any other student or institutional characteristic. Students who interact frequently with faculty are more satis- fied with all aspects of their institutional experience, including student friendships, variety of courses, intellectual environment, and even administration of the institution. Finding ways to encourage greater personal con- tact between faculty and students might increase students' satisfaction with their college experiences.43 Katz and Sanford, in interviews with students, report that ". . . students almost universally link their most significant educational experiences to teachers with 41C0pe and Hannah, pp. cit. 42Noel, pp. cit. 43Alexander W. Astin, Four Critical Years: Effects of College on Beliefs, Attitudes, and Knowledge, (San Francisco: Jossey-Bass, 1977), p. 223. 32 whom they have had some personal relation in and out of the classroom."44 In the development of an empirical model of the drOpout process, Spady found that outside contacts with faculty are very important to the intellectual and cultural growth of students, which aids in the development of greater institutional commitment.45 Panos and Astin studied freshmen at 248 colleges and followed up four years later in order to examine the impact of several aspects of the college environment on student persistence. They found that . . . Students are less likely to drop out if they attend colleges where the classroom environment is characterized by a high level of personal involvement on the part of the instruc- tors and students, and where there is a hi h degree of familiarity with the instructor. It becomes increasingly apparent that faculty can, and do, have impact upon those students with whom they interact. There are, however, two conceptual problems which tend to cloud the findings of those studies which show a relationship between faculty-student interaction and persistence. 44Joseph Katz and Nevitt Sanford, "Curriculum and Personality,“ in N. Sanford (Ed.), Collpge and Character, (New York: John Wiley & Sons, 1964), pp. 123- 124. 45Spady, pp. cit. 46Panos and Astin, pp. cit., p. 66. 33 First, Wilson, pp pl. report that the nature and frequency of faculty-student interactions are, to a great extent, a function of the characteristics of those people involved in the interaction. As a result, the finding that persisters have more out of class contact with faculty than dropouts may be due as much to the particular character- istics which the student brings to the college as the actual experience of college itself.47 Second, out of class student-faculty contact has generally been treated as a global phenomenon. There is very little research available which examines the types of interaction between students and faculty with respect to persistence in college. In one recent study, however, it was hypothesized that different types of faculty-student contact are not of equal importance in fostering the persistence of students. The purpose of this study was to compare the pattern of relationships between types of faculty-student contact and retention. After controlling for entering student charac- teristics, the investigators found that persisters had significantly more contact with faculty than nonpersisters concerning intellectual or course-related matters. Dis- cussions concerning career concerns and academic advising also yielded significant results. The investigators suggest 47Wilson, pp 21., pp. cit. 34 that colleges take steps to find ways to increase the frequency of faculty-student interaction.48 It appears that whenever students do hold dis- cussions with faculty, they find these discussions valuable. However, prevailing institutional climates and student- faculty relationships lead to minimal interaction. "Thus the potential benefits are realized by but a few students. The answer . . . may lie with increased faculty sensitivity and accessibility."49 The call for increased faculty involvement in retention efforts appears to be a popular notion in the recent literature. Burton and Johnson suggest that a large number of promising students, who may have survived if they had received proper guidance and attention from faculty, have been lost forever.50 The literature supports the claim that faculty can and do have impact on individual students, but persis- tence is a result of students' interaction and integration with the total campus environment. What impact do faculty have in shaping the environment of the institution? Although the peer group has generally been seen as the major influence on student satisfaction with college, 48Pascarella and Terenzini, pp. cit. 49Cope and Hannah, pp. cit., p. 58. 50Robin Burton and Philip Johnson, "Identifying Potential Dropouts with Class Lists," Improving Cpllegp and University Teaching, 17 (Summer, 1969), pp. 178-179. 35 there exists a considerable amount of evidence that faculty attitudes and behaviors do shape the college milieu.51 Pascarella studied student perceptions of the institutional environment and found that ". . . the distinctive environ- ment of an institution is shaped more by the policies and actions of faculty and administrators than by the students it admits . "52 One could argue that the amount of faculty impact and faculty-student contact on a campus are a function not only of the type of people involved, but that they are also related to the size of the institution. Kamens reports that freshmen at small schools have more out of class dis- cussions with faculty than those at large schools. According to Kamens, ". . . these data suggest that faculty at small colleges may attach more value to undergraduates as students and, therefore, be more willing to interact with them informally or take an interest in them."53 On the other hand, in a survey of students who withdrew from thirteen small colleges, Chickering found 51Joan S. Stark, "The Relation of Disparity in Student and Faculty Educational Attitudes to Early Student Transfer from College," Research in Higher Education, 3 (1975), PP. 329-344. 52Ernest T. Pascarella, "Student-Institutional Congruence: A Student Perspective," College Student Journal, 8 (1974), p. 86. 53David H. Kamens, "The College Charter and College Size: Effects on Occupational Choice and College Attrition," Sociology of Education, 44 (Summer, 1971), p. 281. 36 that there is a limited amount of communication between students and faculty outside of class. This communication was . . . Limited in the range of different faculty members seen, limited in the numbers of conver- sations, limited in the amount of time spent-- even on academic and educational planning, matters central to the purpose of the institution and to the prime function of the faculty.54 To this point, the review of existing literature concerning faculty involvement in student persistence has focused upon the concepts of contact and impact. If faculty have been, or are to be involved in programs to improve student retention, it is necessary to look at specific, suggested roles that faculty may perform. According to Webb, there is little factual mater- ial available concerning faculty contributions to dropouts. However, in considering the ways that faculty might behave in interacting with students, it is possible to suggest certain families of behaviors or roles performed by faculty that seem on a logical basis related to student dropout.55 These roles include: curriculum developer, instructor, evaluator, maintainer of standards, counselor or advisor, and stimulator or supporter. Webb argues that, to the extent that faculty do not perform these roles, they 54Arthur W. Chickering, "Student-Faculty Relationships: Bedrock for College Governance," Revised version of comments addressed to the 15th Annual Institute on College and University Administration, (University of Michigan: Ann Arbor, June, 1969), ERIC-ED 038 910, p. 3. 55Webb, pp. cit., p. 3. 37 contribute to the problem of attrition.56 The roles suggested above include both in and out of class interactions and functions of faculty. Since this study is primarily concerned with the out of class impact of faculty, a separate set of roles will be investigated. Wilson, pp pl. developed six role capacities which deal with formal and informal teaching activities that take place outside the classroom. They include: Instructor. To discuss intellectual or academic matters with a student. Educational advisor. To give a student basic information or advice about his academic program. Career advisor. To help a student consider matters related to his future career. Friend. To socialize informally with a student. Counselor. To help a student resolve a disturbing personal problem. Campus citizen. To discuss a campus issue or prob- 57 lem with a student. The investigators utilized these roles to deter- mine the frequency of out of class interaction between faculty and students by asking faculty how many contacts they had in each of these capacities in a two week period of time. These results were compared with students' 56Webb, pp. cit. 57Wilson, pp pl., pp. cit., p. 32. C ..e. I.“ . 38 responses concerning the impact of faculty upon their college experience. Their findings ". . . suggest that the relationship that faculty and students develop outside the classroom may well be the part of teaching which has the greatest impact on students."58 Summary In this chapter, the investigator reviewed the literature concerning selected aspects of the problem of student attrition. Although a tremendous amount of research exists concerning dropouts from higher education, there are a number of problems with the existing litera- ture. First, there did not appear to be a universal set of definitions of the terms relevant to the problem of attrition. Most studies, therefore, seemed to continually re-invent the wheel. Second, much of the literature concerning student attrition was contradictory in nature. There was little agreement concerning the factors which cause separation of the student from the institution, the impact of various environments upon students, or ways to reduce attrition rates. Part of the problem appeared to be the lack of an adequate conceptual base upon which new research can build. Third, most of the research has been concerned with descriptions of dropouts or predictions of which 58Wilson, _e_t_:_ §_1_., 92. cit., p. 151. 39 students will leave, rather than on the development of the elements of a staying environment. It was shown that attrition rates have been con- sistent for many years and involve a large percentage of those students who initially enter postsecondary institutions. Concern for attrition will continue to grow as the pool of traditional students decreases. In many colleges, at the present time, concern for attrition is being equated with survival of the institution. It was apparent that faculty have some impact on students, and several investigators have suggested further research into the impact that faculty may have in campus retention efforts. No studies were located, however, in which faculty had been asked what impact they felt they could and should have in assisting these efforts. Despite the acknowledged importance of college professors for instructing the youth of the nation and for governing colleges and universities, reliable information about faculty members' activities, attitudes, and values is surprisingly limited. Relatively few empirical studies of faculty have been conducted. Of the studies that have been made, only a few have obtained data directly from faculty members themselves, and only a handful have included more than one institution.59 The following chapter is concerned with the methodology and procedures used for this study. 59Wilson, pp pl., pp. cit., p. 4. CHAPTER III METHODOLOGY The investigator's purpose in this study was to elicit, describe, compare and evaluate the perceptions of faculty and administrators from eight high attrition and four low attrition colleges concerning selected aspects of student attrition in these institutions. The twelve schools utilized in this study were participants in the Strengthen- ing Institutions Through Improving Retention (SITIR) pro- ject, which is funded through the Office of Education under Title III.1 This chapter describes the hypotheses tested, the population sampled, the instrument used, the adminis- tration of the instrument, and the statistical methods used to test the hypotheses. Hypotheses The basic hypotheses of this study were stated in Chapter I. They are restated here as null hypotheses: 1Permission to conduct this study was received in accordance with the "General Provisions for Programs," (45 CFR 100A.263) Office of Education, Washington, D.C. In addition, each of the participating institutions agreed to become involved in this project upon a guarantee of anonymity for both respondents and institutions. Separate reports have been made available to the appropriate offi- cials at each participating institution and the SITIR office. 40 POpulation 41 There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. There are no significant differences in the stated amount of out of class student contact that faculty have in high and low attrition colleges. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact faculty can have in reducing student attrition. There are no significant differences in the perceptions of faculty and those of adminis- trators in high attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. There are no significant differences in the perceptions of faculty and those of adminis- trators in low attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. The population for this study consisted of 544 respondents (391 faculty and 153 administrators) from twelve of the twenty colleges which participate in the SITIR pro- ject. Eight of the colleges in the SITIR project were excluded from this study for one or more of the following reasons: 42 1. Data on attrition rates were not available. 2. Admissions standards differed considerably from those colleges included in the study. 3. The college does not emphasize a four year program for graduation, thus attrition rates were skewed. 4. The college could not be classified as either a high or low attrition institution, because attrition rate was within national average range. Each of the colleges in this study is classified as a small, private, four-year, liberal arts school. In addition, each of these institutions qualifies for, and receives, funds for participation in the SITIR project from the Strengthening Developing Institutions Program.2 Due to the small size of these colleges, it was propitious and appropriate to include all faculty and administrators of each institution in the survey. Because anonymity was guaranteed for the colleges in this study, each school was assigned a letter designation so that colleges could be distinguished from one another. Institutions A,B,C, and D, are the four low attrition colleges. Institutions S, T, U, V, W, X, Y, and Z, are the eight high attrition colleges in the study; Descriptions of each college are limited in order to protect anonymity. In these descriptions, the percentage of resident students is provided because living on campus has been shown to be related to persistence in college. 2Authority: Title III, Higher Education Act, 1965, P.L. 89-329, as amended. 43 Church affiliation is provided because church-related colleges, especially Catholic colleges, appear to have greater holding power than other schools. The single sex/ coeducational distinction has been made because single sex colleges tend to have greater rates of persistence than coeducational institutions. The literature indicates that the scores on standardized tests and high school grade point averages are the best predictors of success in college. If compari- sons are to be made between colleges in this study, it is important that information concerning the previous perfor- mance of students in these colleges be provided. Infor- mation is provided in Table 3.1 concerning the 1978-79 incoming freshman average scores on either the ACT or SAT tests for each institution. In addition, a summary of admissions standards for each college has been included, since these standards are often a major determinant of the academic quality of the student body. Admissions standards, however, are generally quite vague in order to allow for institutional flexibility in the selection of students. ,The total institutional enrollment figures provided are based upon Full Time Equivalency (FTE) figures for Fall, 1978. The information about each institution was gathered from the college catalogs, a SITIR project data survey, and specific requests from the investigator. Descriptions of each college are provided below, and a summary table of information is found in Table 3.1. Amaomeo>m pone <\z«« mhma .HHmm .cmfinmmhm ocflumgco new omuoasoamo monoom mmoum>< AmHBHm. sumo mhlmhma com: compo cowuocfiumwo oooo\xom onCHmv mmonuoo wmmHHoo Ga coaumEHOMcfl coo: comma scanneawmwm msoflmwaomm AmHBHmV sumo mulmhmfl coma comma wmmucmoumm ucoosum ucoowmomm mucoosum mEHuluumm maflnmfiflum .mHmE we moon Pompoum och mo wm wanes 44 m.mH mmh amoo qeonadommozoo mam u m.mH mom omoo zmHmmewmmmmm .wmm » A.GH «\z omoo z m.na 4mm nmoo ngomedo me o «\z omm omou ezmozmdmozH was a «H «\z omoo zeHmmewmmmmm mmm m «\2 sea zmzoz quome wo.m~ hmfi\mm hNH\mm ema\wm ovH\mm > wh.mo wm.mm mmH\mw me\Hm oma\vw mom\hm D w¢.on mm.mm moN\mm mNN\vm mmm\mh mmm\mh B wm.bh wH.NN mmH\Hm mNH\vm mvH\Nm mma\mm m wo.mm wo.mm mmH\mNH m5H\ONH mmH\MOH Nom\mmH D wo.mm wv.om vma\Hh NHH\vw mvH\Hm mHH\ow U wfi.mm mm.om N¢H\Nh mNH\om NmH\ooa mNH\Nm m wh.hv wm.~m mmm\bNH NNN\hHH oNN\MNH mom\mMH d ZOHBHmBB< madmo m5. madmo m5. mnémw vb. madam oh. madmw .U>d .0>¢ ms. mmomh on. mmomm m5. mmomm «Nb. mmomm mwmAAOU mhlmhmfi .mhlvbma .bhlmnma .whlmhmfi mmfimm GOfluHHuu¢\GOflHMDUMHU .m.m mnmda 57 determine whether or not interactions occurred between the variables. Where interactions do not occur, the program tests for significance of main effects. Since significance is merely an indication that some relationship exists, an "eta squared" correlation ratio was applied to determine the prOportion of variance accounted for by the treatment of the data.8 In addition, SPSS Crosstabulations and Condescriptives were used in order to display all the data, including the cluster groupings, in descriptive form. One way analysis of variance was selected for use in this study because it produces F-ratios which are used to compare means in order to determine the existence of a statistical relationship. An important feature of ANOVA is that it permits the separation of all potential infor- mation in the data into distinct and nonoverlapping portions, each reflecting only certain aspects of the experiment.9 Chapter IV includes the presentation and analysis of the data. 8Nie, Bent, and Hull, pp. cit. 9W.L. Hays, Statistics for the Social_§ciences, (New York: Holt, Rinehart and Winston, Inc., 1973). CHAPTER IV PRESENTATION AND ANALYSIS OF DATA Introduction This chapter is devoted to a presentation and analysis of the data collected for this study. The purpose of this research was to determine whether or not differ- ences exist in the perceptions of faculty and adminis— trators at high attrition colleges and those at low attrition colleges concerning faculty involvement in reducing student attrition. Comparisons were also made between faculty and administrators to see if differences exist between these two groups in both high attrition and low attrition colleges. The population for this study consisted of all faculty and administrators in twelve small, private, four- year, liberal arts colleges. Each of these colleges was a participant in the Strengthening Institutions Through Improving Retention (SITIR) project at the time of the study. Based upon institutional data provided for this study, eight of these colleges were considered to be high attrition schools and four were considered to be low attrition schools. Following an extensive review of the literature, a questionnaire was developed to elicit the perceptions of faculty and administrators concerning the factors associated 58 59 with student attrition, the amount of out-of—class contact that faculty state they have with students, and the poten- tial impact which faculty can have in helping reduce student attrition. The instrument was pilot-tested in October, 1979, and appr0priate modifications were made as a result of the pilot test. The questionnaires were administered in November and December, 1979, on each campus by the campus representative to the SITIR project. Of the 659 question- naires administered, 544 (82.6%) were returned in usable form (391 faculty and 153 administrators). The responses to the questionnaires were trans- posed onto data processing cards and computer tape to accommodate computer-assisted statistical analysis of the data. The data were analyzed through the use of both descriptive and inferential statistical techniques, including exploratory and confirmatory factor analysis, crosstabulations, condescriptives, and one way (fixed- effects) analysis of variance. Hypotheses to be Tested For the purpose of this study, the following null hypotheses were develOped: 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. 2. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. 60 3. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. 4. There are no significant differences in the stated amount of out of class student contact that faculty have in high and low attrition colleges. 5. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact faculty can have in reducing student attrition. 6. There are no significant differences in the perceptions of faculty and those of adminis- trators in high attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. 7. There are no significant differences in the perceptions of faculty and those of adminis- trators in low attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. Treatment of the Data The questionnaire used for this study was divided into six parts: general institutional perceptions, student- related factors in attrition, institutional factors in attrition, faculty activities outside the classroom, impact of faculty activities in reducing attrition, and demographic information. A variety of statistical methods were employed in order to adequately analyze the resultant data. For ease in responding, the questionnaire items concerning the factors associated with attrition were divided into two sections, student-related factors and institutional factors. Exploratory and confirmatory factor analyses were used to group these questionnaire items into clusters 61 (subscales) for measurement purposes. The reason for using factor analysis was that producing a cluster of items which measure a single factor provides a more reliable measure of the factor than one would get from measurement of a single 1 item. As a result of the factor analyses, four subscales were produced: Subscale 1: Counseling and Academic Services. This subscale included nine items from the list of student- related and institutional factors in attrition: lack of academic challenge, dissatisfaction with academic require- ments or regulations, curricular offerings too narrow, cpurses or majors eliminated, inadequate academic advising, lack of flexibility in planning personalized academic pro- grams, low instructional quality, inadequate personal coun- seling, and inadequate career counseling. Subscale 2: Student Problems or Goals. The items in this subscale included financial difficulties, lack of personal/emotional adjustment to college, change in per- sonal or career goals, conflict with full-time job, personal problems, illness or accident, lack of peer group identi- fication, and student originally planned to dr0p out or transfer. Subscale 3: Nonacademic Institutional Services. This subscale included dissatisfaction with social require— ments or regulations, social policies too restrictive, lack 1Hunter and Gerbing, pp. cit. 62 of extracurricular activities, inadequate student financial aid program, and poorly maintained physical facilities. Subscale 4: Student Ability, Interest and Preparation. The items in this subscale were low academic ability, inadequate academic preparation for college, lack of interest or motivation, and low academic ability of stu- dents admitted to this college. One questionnaire item concerning factors in attrition was eliminated from the study as a result of the process of factor analysis. The item, "social policies too liberal," did not correlate with the other items in any of the above subscales. One way, fixed-effects analysis of variance was used to test the hypotheses concerning the perceptions of the factors associated with attrition. These results will be reported later in this chapter. In the fourth part of the questionnaire, faculty were asked to indicate the number of contacts they had with students in a two week period of time in each of the following capacities: academic advisor, campus citizen, career advisor, out of class instructor, participant, per- sonal counselor, referral agent, and social interactor. Only contacts which were "substantive in time or content" were to be counted. The hypothesis concerning faculty- student contact outside the classroom was tested using analysis of variance. In addition, faculty were asked if they were involved with students in any of the following activities: as a sponsor or advisor to a student club or 63 organization, as a member of a campus committee that has student representation, as a practicum/internship super- visor, and/or as an independent study supervisor. These responses were analyzed descriptively. In order to test the hypotheses concerning the impact of faculty activities in reducing student attrition, the twelve activities were factor analyzed into two sub- scales: Subscale 5: Faculty Out of Class Impact-Advice, Instruction and Supprvision. This subscale included five faculty activities: academic advisor, career advisor, independent study supervisor, out of class instructor, and practicum supervisor. Subscale 6: Faculty Out of Class Impact-PersonalJ Social and Special Interest Interactions. This subscale included the following items: campus citizen, club sponsor, committee member, participant, social interactor, and per- sonal counselor. One item, "referral agent," was eliminated from the study as a result of the factor analysis. One way analysis of variance was used to test each subscale. Questionnaire items concerning general institu- tional perceptions and the demographic information were analyzed descriptively using SPSS Condescriptives and Crosstabulations. In this study, the null hypotheses were not accepted if the F-ratios were significant at the .05 level. One of 64 the assumptions underlying the F-test is that samples under comparison are approximately the same size. This assumption could not be met due to the relatively small number of administrators compared with the number of faculty in each institution, and because the number of low attrition col— leges was smaller than the number of high attrition colleges. Because of the problem of unequal sample sizes, the .05 level of significance was chosen in order to allow for identification of trends and differences within the data, while prohibiting the rejection of a true null hypothesis. Although the decision to set a significance level at a certain value is arbitrary, the .05 level is small enough to keep from making a Type I error; that is, falsely rejecting the tested hypothesis, while maintinaing enough flexibility to identify differences between populations.2 Description of the Population--Demographic Information The population for this study was comprised of 544 respondents (391 faculty and 153 administrators) from eight high attrition colleges and four low attrition colleges. The colleges in this study are quite small and anonymity was guaranteed for all respondents; therefore, the description of the population was necessarily limited. Faculty members were asked to indicate their teach- ing field ("Your current, primary responsibility"). The choices provided and number of faculty in each field were: 2Hays, pp. cit. 65 HUMANITIES ----------------------------------- 142 SOCIAL SCIENCES ------------------------------ 73 NATURAL SCIENCES ------------------------- _-_- 79 PROFESSIONAL-APPLIED FIELDS ------------------ 31 OTHER ---------------------------------------- 16 The responses to the remaining demographic items (Tables 4.1-4.5) are provided in percentages and are sep- arated by faculty responses, administrator responses, re- sponses from high attrition colleges, and responses from low attrition colleges. Table 4.1 provides a summary of responses regard- ing the number of years the subjects in this study had held their "current" positions. Almost 60 percent of the respondents from the low attrition colleges indicated that they had been in the same position for seven or more years, while 40 percent of those from high attrition colleges had seven or more years of experience in their positions at those institutions. Differences between faculty members and administrators were also noted. Of the faculty respondents, 55.9 percent had been in their "current" positions seven years or more, whereas 22.5 percent of the administrators indicated seven or more years of experience in their positions at the same institutions. In fact, 47 percent of the administrator respondents had held their positions two years or less at the time of the study. TABLE 4.1. "Number of years you have held current, primary responsibility at this institution:" in percentages) (Summary of responses, HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Less than one year 11.5 26.5 17.4 12.6 1-2 years 10.5 20.5 14.0 12.0 3-6 years 22.1 30.5 29.1 15.7 7-12 years 27.7 12.6 23.4 23.6 More than 12 years 28.2 9.9 16.0 36.1 As indicated in Table 4.2, 47.4 percent of the faculty respondents and 21.7 percent of the administrators held doctorate degrees at the time of this study. Very little difference existed, however, between low and high attrition colleges in the levels of educational attainment of the respondents. Over two-thirds of the faculty respondents from low attrition colleges were tenured at the time of this study, while less than half of the faculty respondents from the high attrition institutions held tenure. These results were not surprising, since a large percentage of the sub- jects from the low attrition colleges had held their "current" positions longer than those from the high attrition schools. Table 4.3 provides the percentages of tenured and non-tenured faculty respondents. 67 TABLE 4.2. "Highest degree you have earned:" (Summary of responses, in percentages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRIT ION ATTRITION Bachelor's .5 25.7 8.9 5.2 Master's 49.0 51.3 47.6 53.4 Doctorate 47.4 21.7 41.2 38.3 Other 3.1 1.3 2.3 3.1 TABLE 4.3. "Faculty: Are you tenured? (If your insti- tution does not grant tenure, please leave this item blank.)" (Summary of responses, in percentages) RESPONSE FACULTY HIGH ATTRITION LOW ATTRITION Yes 53.6 44.8 67.1 No 46.4 55.2 32.9 Faculty members were asked to indicate the num- ber of students for whom they were serving as academic advi- sors at the time of the study. It was interesting to note that 15.8 percent of the faculty respondents had no advisees, while 13 percent had more than 30 advisees. These results are presented in Table 4.4. 68 TABLE 4.4. "Faculty: Number of students for whom you are serving as academic advisor:" (Summary of responses, in percentages) RESPONSE FACULTY HIGH ATTRITION LOW ATTRITION None 15.8 13.3 22.7 1-10 28.5 34.1 17.7 11-20 30.3 29.7 30.5 21-30 12.4 11.6 13.5 More than 30 13.0 11.2 15.6 Table 4.5 provides a summary of responses from faculty and administrators regarding the total number of years of college teaching experience each group had at the time of the study. Almost half (48 percent) of the respon- dents from the low attrition colleges indicated that they had taught full time at the college level for over twelve years, while less than one-third (29.7 percent) of those from the high attrition colleges had twelve or more years of college teaching experience. TABLE 4.5. "Faculty and Administrators: Your total number of years of full time postsecondary teaching experience:" (Summary of responses, in percentages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Less than one year 5.1 27.6 12.1 8.6 1-2 years 8.1 10.2 8.7 8.6 3-6 years 17.8 21.3 20.1 16.0 7-12 years 28.8 16.5 29.4 18.9 More than 12 years 40.2 24.4 29.7 48.0 General Institutional Perceptions--Summary of Responses The first section of the instrument used for this study was designed to elicit faculty and administration perceptions regarding the extent of student attrition, the locus of responsibility for reducing attrition, and the current status of institutional programs for reducing attrition. The responses to these items (Tables 4.6-4.12) are provided in percentages and are separated by faculty responses, administrator responses, responses from high attrition colleges, and responses from low attrition col- leges. The subjects were asked to respond to the following statement: "I consider student attrition to be a problem at this institution." In the high attrition colleges, 43.4 percent of the respondents indicated that 70 they "strongly agree" that attrition is a problem, while 17.8 percent of those in the low attrition colleges marked "strongly agree." Similarly, when asked to estimate the percentage of students who leave these colleges prior to completion of a degree, the respondents from the low attrition colleges generally perceived their attrition rate to be lower than those in the high attrition insti— tutions. Tables 4.6 and 4.7 provide a summary of responses to these items. It appears that faculty and administrators from both low and high attrition colleges were aware of the relative extent of the problem of student attrition in their respective institutions. It should be noted, how- ever, that a large percentage of the respondents from both high and low attrition colleges underestimated the rate of attrition in their institutions. TABLE 4.6. "I consider student attrition to be a problem at this institution." (Summary of responses, in percent- ages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Strongly disagree 1.8 .7 .3 3.7 Disagree 8.0 7.2 2.6 17.3 Neither agree nor disagree 12.4 8.5 10.1 13.6 Agree 47.7 38.6 43.7 47.6 Strongly agree 30.1 45.1 43.4 17.8 71 TABLE 4.7. "I estimate that the percentage of students who leave this college prior to completion of a degree is:" (Summary of responses, in percentages) HIGH LOW RESPONSES FACULTY ADMINISTRATORS ATTRITION ATTRITION Less than 20% 16.8 8.6 8.1 26.1 21-40% 48.0 47.4 44.9 53.2 41-60% 31.0 36.8 39.1 20.7 61-80% 4.2 7.2 7.8 0 More than 80% 0 0 0 0 Tables 4.8 and 4.9 provide a summary of responses to two items regarding the locus of responsibility for reducing student attrition. First, the subjects were asked whether or not the primary responsibility for reducing student attrition should rest with administrators. Eighty percent of the administrators and 76.1 percent of the faculty respondents did not attribute the primary respon- sibility for reducing attrition to administrators. In fact, when asked if faculty members can play a role in reducing student attrition without lowering academic stand- ards, 89.6 percent of the faculty respondents either agreed or strongly agreed that they can play a role in reducing attrition. Although the actual locus of responsibility in programs for reducing attrition remains unclear, it was apparent that faculty perceived the existence of a role for themselves in these programs. 72 TABLE 4.8. "The primary responsibility for reducing student attrition should rest with administrators." (Summary of responses, in percentages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Strongly disagree 15.3 24.0 16.5 20.0 Disagree 35.1 42.0 35.7 39.5 Neither agree nor disagree 25.7 14.0 24.6 18.4 Agree 20.3 16.0 19.1 18.9 Strongly agree 3.6 4.0 4.1 3.2 TABLE 4.9. "Faculty members can play a role in reducing student attrition without lowering academic standards." (Summary of responses, in percentages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Strongly disagree 1.6 0 .9 1.6 Disagree 2.9 .7 1.5 3.7 Neither agree nor disagree 6.0 .7 5.2 3.1 Agree 51.2 32.9 42.4 52.4 Strongly agree 38.4 65.8 50.0 39.3 As concern for retention has increased, many colleges have developed formal programs in an attempt to reduce student attrition. The subjects were asked whether 73 or not a formal program existed in their institutions at the time of this study. In the high attrition colleges, 34.3 percent of the respondents indicated that their institutions had a retention program, while 67 percent of those from low attrition colleges indicated that a program existed in their institutions. (See Table 4.10.) Of the subjects who stated that a formal program for reducing attrition existed, over half did not know if the program was effective in either the high or low attrition colleges. (See Table 4.11.) Respondents from low attrition colleges, however, perceived their programs to be effective consider— ably more often than did those from the high attrition col- leges (46.5 percent to 23.7 percent). These subjects were also asked whether or not faculty members were involved in their retention programs. Over 80 percent of the respondents in both high and low attrition colleges indicated that faculty were involved in the formal programs for reducing student attrition. Table 4.12 provides a summary of the responses to this item. TABLE 4.10. "This institution has a formal program for reducing student attrition." (Summary of responses, in percentages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Yes 43.5 52.0 34.3 67.0 No 28.8 34.9 38.9 15.2 Do not know 27.7 13.2 26.8 17.8 74 TABLE 4.11. "The program for reducing student attrition at this institution is effective." (Summary of responses, in percentages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Yes 32.3 42.3 23.7 46.5 No 12.0 10.3 20.3 3.1 Do not know 55.7 47.4 55.9 50.4 TABLE 4.12. "Faculty members are involved in the program for reducing attrition." (Summary of responses, in per- centages) HIGH LOW RESPONSE FACULTY ADMINISTRATORS ATTRITION ATTRITION Yes 85.6 83.3 85.6 84.3 No 4.2 3.8 5.1 3.1 Do not know 10.2 12.8 9.3 12.6 Factors Associated with Attrition Twenty-seven institutional and student-related factors in attrition were identified from previous research. The subjects were asked to report their perceptions of the importance of each item as a cause of student attrition in their institutions, according to the following choices: Of Very Low Importance Of Low Importance Of Moderate Importance (AND-l II II II 75 4 5 Of High Importance Of Very High Importance The items concerning the factors associated with attrition were grouped into four subscales. Null hypotheses 1, 2, and 3 were tested for each subscale: 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. 2. There are no significant differences in the perceptions of faculty and those of adminis— trators concerning the factors which affect student attrition in high attrition colleges. 3. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. Analysis of Subscale 1: Counseling and Academic Services. Subscale 1 included the following items: lack of academic challenge, dissatisfaction with academic require- ments or regulations, curricular offerings too narrow, courses or majors eliminated, inadequate academic advising, lack of flexibility in planning personalized academic pro- grams, low instructional quality, inadequate personal coun- seling, and inadequate career counseling. The mean response for faculty and administrators at high attrition colleges on Subscale l was 2.62. The mean response for faculty and administrators at low attrition colleges was 2.32. Table 4.13 presents the results of the analysis of variance between the two groups. Null hypothesis 1 for Subscale 1 was not accepted. 76 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. These results indicated that significant differ- ences did exist between the high and low attrition colleges regarding the adequacy of selected counseling and academic services in these institutions. Both faculty members and administrators in the high attrition colleges perceived the items in Subscale 1 to be more important as factors in attrition in their institutions than did the respondents from the low attrition schools. TABLE 4.13. Analysis of Variance for Subscale 1, Hypothesis 1 SOURCE S.S. D.F. M.S. F Between Groups 10.567 1 10.567 26.450 Within Groups 201.360 504 .400 Total 211.927 505 src. = .000 ETA2 = .050 The mean response for faculty at high attrition colleges on Subscale 1 was 2.60. The mean response for administrators at high attrition colleges was 2.68. Table 4.14 presents the results of the analysis of variance between the two groups. The results support null hypothesis 77 2 for Subscale 1. 2. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. The data indicated no significant differences between faculty members and administrators in the high attrition colleges regarding the items in Subscale 1. These groups expressed similar concerns about the quality and quantity of programs and services in their institutions. Because the mean responses for both groups were relatively low, it was apparent that the respondents from the high attrition colleges perceived that factors other than those in Subscale 1 contributed to most of the student attrition in these colleges. TABLE 4.14. Analysis of Variance for Subscale 1, Hypothesis 2 SOURCE S.S. D.F. M.S. F Between Groups .465 l .465 1.202 Within Groups 125.707 325 .387 Total 126.172 326 src. = .274 ETA2 = .004 Faculty members at low attrition colleges had a 2.28 mean response on Subscale 1, while administrators at the same colleges had a mean response of 2.41. The results 78 of the analysis of variance between the two groups are presented in Table 4.15. The results support null hypothe— sis 3 for Subscale 1. 3. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. Although differences existed in the mean responses of faculty and administrators at the low attrition colleges, these differences were not statistically significant. Neither the faculty nor the administrators appeared to attach much importance to the items in Subscale 1 as fac— tors in attrition in the low attrition schools. TABLE 4.15. Analysis of Variance for Subscale 1, Hypothesis 3 SOURCE S.S. D.F. M.S. F Between Groups .531 1 .531 1.259 Within Groups 74.657 177 .422 Total 75.188 178 src. = .263 ETA2 = .007 Analysis of Subscale 2: Student Problems or gpppp. Subscale 2 included the following items: financial difficulties, lack of personal/emotional adjustment to college, change in personal or career goals, conflict with full time job, personal problems, illness or accident, 79 lack of peer group identification, and student originally planned to drop out or transfer. The mean response for faculty and administrators at high attrition colleges on Subscale 2 was 2.75, while the mean for faculty and administrators from low attrition colleges was 2.64. Table 4.16 presents the results of the analysis of variance between the two groups. Null hypo- thesis 1 for Subscale 2 was not accepted. 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. The data revealed the existence of significant differences between the low and high attrition colleges regarding the items in Subscale 2. Faculty and adminis- trators in the high attrition schools indicated that these items were of greater importance as factors in attrition in their institutions than did the respondents from the low attrition colleges. TABLE 4.16. Analysis of Variance for Subscale 2, Hypothesis 1 SOURCE S.S. D.F. M.S. F Between Groups 1.367 1 1.367 4.568 Within Groups 155.985 508 .299 Total 153.352 509 2 SIG. = .033 ETA = .009 80 Faculty members from high attrition colleges had a mean response of 2.78 on Subscale 2, while administrators at the same colleges had a mean of 2.69. The results of the analysis of variance between the two groups are pre- sented in Table 4.17. The results support null hypothesis 2 for Subscale 2. 2. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. These results indicated no significant differences between faculty members and administrators in the high attrition institutions regarding the importance of student problems or goals as factors in attrition. Although the mean responses on Subscales l and 2 for these administrators were quite similar, faculty members in these colleges attached considerably greater importance to the items in Subscale 2 as factors in attrition than they did those in Subscale 1. TABLE 4.17. Analysis of Variance for Subscale 2, Hypothesis 2 SOURCE S.S. D.F. M.S. F Between Groups .548 1 .548 2.015 Within Groups 88.686 326 .272 Total 89.234 327 SIG. = .157 ETA = .006 81 Table 4.18 presents the results of the analysis of variance for Subscale 2 for faculty and administrators at low attrition colleges. For faculty, the mean response was 2.67; for administrators, it was 2.57. The results support null hypothesis 3 for Subscale 2. 3. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. Although no significant differences existed between faculty and administrators in the low attrition colleges regarding the items in Subscale 2, both groups attached greater importance to these items as factors in attrition than they did the items in Subscale 1. As a result, it was apparent that items concerning student problems or goals were perceived to cause more student attrition than items regarding the adequacy of counseling and academic services in these colleges. TABLE 4.18. Analysis of Variance for Subscale 2, Hypothesis 3 SOURCE S.S. D.F. M.S. F Between Groups .323 1 .323 .930 Within Groups 62.429 180 .347 Total 62.752 181 SIG. = .336 ETA = .005 82 Analysis of Subscale 3: Nonacademic Institutional Services. This subscale included the following items: dissatisfaction with social requirements or regulations, social policies too restrictive, lack of extracurricular activities, inadequate student financial aid program, and poorly maintained physical facilities. On Subscale 3, faculty and administrators from high attrition colleges had a mean response of 2.31, while those at low attrition colleges had a mean of 2.47. The results of the analysis of variance between the two groups indicate that null hypothesis 1 for Subscale 3 was not accepted. These results are presented in Table 4.19. 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. These data indicate that significant differences existed between the low and high attrition colleges regard- ing the items in Subscale 3. In contrast to the signifi- cant differences noted earlier regarding Subscales 1 and 2, however, the respondents from the ppg attrition colleges indicated that the items on Subscale 3 were of greater importance as factors in attrition than did the respon- dents from the high attrition colleges. 83 TABLE 4.19. Analysis of Variance for Subscale 3, Hypothesis 1 SOURCE S.S. D.F. M.S. F Between Groups 3.195 1 3.195 6.270 Within Groups 261.404 513 .510 Total 264.599 514 src. = .013 ETA2 = .012 The mean response for faculty at high attrition colleges on Subscale 3 was 2.32, and for administrators it was 2.29. Table 4.20 presents the results of the analysis of variance between the two groups. The results support null hypothesis 2 for Subscale 3. 2. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. These results revealed no significant differences between faculty members and administrators in the high attrition colleges regarding the items in Subscale 3. These items (primarily social factors) were perceived to be the least important factors in attrition of any of the four subscales for these colleges. 84 TABLE 4.20. Analysis of Variance for Subscale 3, Hypothesis 2 SOURCE S.S. D.F. M.S. F Between Groups .052 1 .052 .104 Within Groups 168.549 333 .506 Total 168.601 334 SIG. = .748 ETAZ = .000 Faculty members at low attrition colleges had a 2.42 mean response on Subscale 3, while administrators at the same colleges had a mean response of 2.63. The results of the analysis of variance between the two groups are presented in Table 4.21. The results support null hypo- thesis 3 for Subscale 3. 3. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. These data yielded no significant differences between faculty members and administrators in the low attrition colleges, however, administrators perceived the items in Subscale 3 to be of greater importance as factors in attrition than they perceived the items in Subscales 1 and 2 to be. 85 TABLE 4.21. Analysis of Variance for Subscale 3, Hypothesis 3 SOURCE S.S. D.F. M.S. F Between Groups 1.652 1 1.652 3.227 Within Groups 91.150 178 .512 Total 92.802 179 518. = .074 ETAZ = .018 Analysis of Subscale 4: Student Ability, Interest, and Preparation. Subscale 4 included the following items: low academic ability, inadequate academic preparation for college, lack of interest or motivation, and low academic ability of students admitted to this college. The mean response on Subscale 4 for faculty and administrators at high attrition colleges was 3.31; while for faculty and administrators at low attrition colleges the mean was 3.23. Table 4.22 presents the results of the analysis of variance between the two groups. The results support null hypothesis 1 for Subscale 4. 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. On each of the three previous subscales, signifi- cant differences existed between the high and low attrition institutions. On Subscale 4, the results indicated no 86 significant differences between these groups. However, the mean responses for faculty and administrators at both high and low attrition colleges were considerably higher on Subscale 4 than for any other subscale. Respondents from both high and low attrition colleges perceived items regarding student ability, interest and preparation to be of greater importance as factors in attrition than any other subscale grouping. TABLE 4.22. Analysis of Variance for Subscale 4, Hypothesis 1 SOURCE S.S. D.F. M.S. F Between Groups .862 l .862 1.553 Within Groups 292.434 527 .555 Total 293.296 528 318. = .213 ETAZ = .003 Faculty members from high attrition colleges had a mean response of 3.40 on Subscale 4, while administrators at the same colleges had a mean of 3.10. The results of the analysis of variance between the two groups are pre- sented in Table 4.23. Null hypothesis 2 for Subscale 4 was not accepted. 2. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. 87 The data indicate the existence of significant differences between faculty and administrators in the high attrition colleges regarding the items in Subscale 4. Faculty members attached significantly more importance to these items than did administrators in the same institutions. The mean responses for both groups, however, were higher on Subscale 4 than for any of the other subscales. TABLE 4.23. Analysis of Variance for Subscale 4, Hypothesis 2 SOURCE S.S. D.F. M.S. F Between Groups 6.422 1 6.422 11.989 Within Groups 181.597 339 .536 Total 188.019 340 818. = .001 ETA2 = .034 Table 4.24 presents the results of the analysis of variance for Subscale 4 for faculty and administrators at low attrition colleges. For faculty, the mean response was 3.28; for administrators, it was 3.10. The results support null hypothesis 3 for Subscale 4. 3. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. Although the data yielded no significant differ- ences between faculty members and administrators in the low 88 attrition colleges regarding the items in Subscale 4, both groups perceived these items to be of greater importance as factors in attrition than any other subscale grouping. TABLE 4.24. Analysis of Variance for Subscale 4, Hypothesis 3 SOURCE S.S. D.F. M.S. F Between Groups 1.151 1 1.151 2.073 Within Groups 103.264 186 .555 Total 104.415 187 $10. = .152 ETA2 = .011 An analysis of variance completed on each of the four subscales disclosed no interaction effect (type of institution by position held), but did reveal significant differences between the main effects in each subscale. These differences are reflected in the breakdowns of each subscale as reported above. Out of Class Faculty-Student Contact Twelve types of faculty-student out of class interaction were identified from previous research. Eight of these interactions were combined to form a scale for statistical analysis. The subjects were asked to report the number of contacts they had with students in each of the eight capacities during the two week period prior to 89 answering the questionnaire, according to the following choices: 1 = No Contacts 2 = One or Two Contacts 3 = Three or Four Contacts 4 = Five or Six Contacts 5 = Seven or More Contacts The types of interactions included: academic advisor, campus citizen, career advisor, out of class instructor, participant, personal counselor, referral agent, and social interactor. Table 4.25 presents the results of the analysis of variance between faculty in high attrition colleges and those in low attrition colleges concerning the amount of contact. The mean for faculty from high attrition colleges was 2.75, and for faculty from low attrition colleges, the mean was also 2.75. The results support null hypothesis 4. 4. There are no significant differences in the stated amount of out of class student contact that faculty have in high and low attrition colleges. The data indicate that no significant differences existed between faculty in high attrition colleges and those in low attrition colleges regarding the amount of out of class student contact each group had during the stated two week period of time. In fact, the mean for each group was 2.75, which indicated that the frequency of faculty-student interaction was identical for both groups. 90 TABLE 4.25. Analysis of Variance for Faculty-Student Contact SOURCE S.S. D.F. M.S. F Between Groups .000 1 .000 .000 Within Groups 189.109 360 .525 Total 189.109 361 518. = .995 ETA2 = .000 The three types of interactions reported most frequently by faculty from both high and low attrition colleges were: academic advisor (mean = 3.77), out of class instructor (mean = 3.12), and social interactor (mean = 2.88). Faculty were also asked to indicate whether or not they interacted with students in the four capacities not included in the contact scale used to test null hypo- thesis 4. Table 4.26 presents the results of faculty responses, in percentages, to the following questions: 1. Do you serve as a sponsor or advisor to a student club or organization? 2. Do you serve on a campus committee that has student representation? 3. Do you serve as a practicum/internship supervisor? 4. Do you supervise any students in an 91 independent study? As indicated in Table 4.26, 70.5 percent of the faculty from low attrition colleges and 62.6 percent of those from high attrition colleges supervised students in independent studies. Almost 70 percent of all faculty respondents served on campus committees that had student representation, while almost half served as practicum/ internship supervisors. The percentage of faculty from low attrition schools who served as club sponsors was 52.8 percent, while 38.1 percent of those in high attrition colleges served as club sponsors. There appeared to be little difference between low and high attrition college faculty regarding the type and amount of out of class stu- dent contact they state they had. Only regarding club sponsorship was there a considerable difference in the response percentages between these two groups. Impact of Faculty Activities in Reducing Studenthttrition Twelve faculty out of class activities, which may have some impact in reducing student attrition, were identified from previous research. The subjects were asked to report their perceptions of the amount of potential impact each of the activities may have in reducing student attrition in their institutions. The response choices included: Very Low Impact Low Impact Moderate Impact High Impact Very High Impact UlobWNI-J II II II II II TABLE 4.26. 92 Summary of Responses, Faculty Out of Class Activities (in Percentages) LOW ATTRITION HIGH ATTRITION TYPE OF CONTACT RESPONSE FACULTY FACULTY Club Sponsor Yes 52.8 38.1 No 47.2 61.9 Committee Member Yes 69.5 69.0 No 30.5 31.0 Practicum/Internship Yes 47.9 47.4 Supervisor No 52.1 52.6 Independent Study Yes 70.5 62.6 Supervisor No 29.5 37.4 The faculty activities were grouped into two subscales (Subscale 5 and Subscale 6). Null hypotheses 5, 6, and 7 were tested for each subscale. 5. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact faculty can have in reducing student attrition. There are no significant differences in the perceptions of faculty and those of adminis- trators in high attrition colleges concerning the impact faculty can have in reducing student attrition. There are no significant differences in the perceptions of faculty and those of adminis- trators in low attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. 93 Analysis of Subscale 5: Faculty Out of Class Impact-~Advice, Instruction, and Supervision. This sub- scale included the following items: academic advisor, career advisor, independent study supervisor, out of class instructor, and practicum supervisor. The mean response for faculty and administrators at high attrition colleges on Subscale 5 was 3.90. The mean response for faculty and administrators at low attrition colleges was 3.92. Table 4.27 presents the results of the analysis of variance between the two groups. The results support null hypothesis 5 for Subscale 5. 5. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact faculty can have in reducing student attrition. The results indicate that no significant differ- ences existed between the high and low attrition colleges regarding the items in Subscale 5. In fact, the mean responses for each group were almost identical. ‘The size of the means (3.90 and 3.92) indicate that the respondents in both high and low attrition colleges perceived that faculty can have considerable impact in reducing student attrition through performance of activities such as aca- demic advising and career advising. 94 TABLE 4.27. Analysis of Variance for Subscale 5, Hypothesis 5 SOURCE S.S. D.F. M.S. F Between Groups .037 1 .037 .088 Within Groups 215.536 516 .418 Total 215.573 517 318. = .767 ETAZ = .000 The mean response for faculty at high attrition colleges on Subscale 5 was 3.81, while the mean for adminis- trators at the same institutions was 4.11. Table 4.28 pre- sents the results of the analysis of variance between the two groups. Null hypothesis 6 for Subscale 5 was not accepted. 6. There are no significant differences in the perceptions of faculty and those of adminis- trators in high attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. These data reveal that significant differences existed between faculty and administrators in the high attrition colleges regarding the items in Subscale 5. Although the faculty members in these colleges perceived that they can have considerable impact in reducing student attrition (mean = 3.81), administrators perceived potential faculty impact to be even greater (mean = 4.11). 95 TABLE 4.28. Analysis of Variance for Subscale 5, Hypothesis 6 SOURCE S.S. D.F. ' M.S. F Between Groups 6.055 1 6.055 14.483 Within Groups 138.803 332 .418 Total 144.858 333 $18. = .000 ETA2 = .042 Faculty members from low attrition colleges had a mean response of 3.83 on Subscale 5, while administrators at the same colleges had a mean of 4.13. The results of the analysis of variance between the two groups are pre- sented in Table 4.29. Null hypothesis 7 for Subscale 5 was not accepted. 7. There are no significant differences in the perceptions of faculty and those of adminis- trators in low attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. These data indicate significant differences between the perceptions of faculty and those of administra- tors in low attrition colleges regarding the items in Sub- scale 5. These results were strikingly similar to those for the high attrition institutions, in that administrators (mean = 4.13) perceived that faculty members (mean = 3.83) can have considerably greater impact in reducing student attrition than did the faculty members themselves. 96 TABLE 4.29. Analysis of Variance for Subscale 5, Hypothesis 7 SOURCE S.S. D.F. ' M.S. F Between Groups 3.357 1 3.357 9.075 Within Groups 67.321 182 .370 Total 70.678 183 src. = .003 ETA2 = .048 Analysis of Subscale 6: Faculty Out of Class Impact--Persona1, Social, and Special Interest Interactions. This subscale included the following items: campus citizen, club sponsor, committee member, partici— pant, social interactor, and personal counselor. The mean response for faculty and administrators at high attrition colleges on Subscale 6 was 3.30. The mean response for faculty and administrators at low attrition colleges was 3.35. The results of the analysis of variance between the two groups are presented in Table 4.30. The results support null hypothesis 5 for Subscale 6. 5. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact faculty can have in reducing student attrition. The results indicate that significant differences did not exist between high attrition and low attrition 97 colleges regarding the items in Subscale 6. Although respondents in both low and high attrition colleges per- - ceived that faculty can have moderate to high impact in reducing student attrition, the mean responses were lower for the items in Subscale 6 than for those in Subscale 5. TABLE 4.30. Analysis of Variance for Subscale 6, Hypothesis 5 SOURCE S.S. D.F. M.S. F Between Groups .391 1 .391 .686 Within Groups 297.843 522 .571 Total 298.234 523 SIG. = .408 ETAZ = .001 The mean response for faculty at high attrition colleges on Subscale 6 was 3.19. The mean for administra- tors at high attrition colleges was 3.56. Table 4.31 pre- sents the results of the analysis of variance between the two groups. Null hypothesis 6 for Subscale 6 was not accepted. 6. There are no significant differences in the perceptions of faculty and those of adminis- trators in high attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. These data reveal that significant differences existed between faculty and administrators in the high 98 attrition colleges regarding the items in Subscale 6. The results were similar to those for Subscale 5 in that adminis- trators (mean = 3.56) perceived potential faculty impact to be greater than faculty (mean = 3.19) perceptions of their own potential impact. TABLE 4.31. Analysis of Variance for Subscale 6, Hypothesis 6 SOURCE S.S. D.F. M.S. F Between Groups 9.950 1 9.950 19.530 Within Groups 171.688 337 .510 Total 181.638 338 $18. = .000 ETA2 = .055 Faculty members from low attrition colleges had a mean response of 3.21 on Subscale 6, while administrators at the same colleges had a mean of 3.72. The results of the analysis of variance between the two groups are pre- sented in Table 4.32. Null hypothesis 7 for Subscale 6 was not accepted. 7. There are no significant differences in the perceptions of faculty and those of adminis- trators in low attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. These data indicate significant differences between the perceptions of faculty and those of administra- tors in low attrition colleges regarding the items in 99 Subscale 6. Again, faculty respondents (mean = 3.21) per- ceived their potential impact in reducing attrition to be less than administrator (mean = 3.72) perceptions of faculty impact. TABLE 4.32. Analysis of Variance for Subscale 6, Hypothesis 7 SOURCE S.S. D.F. M.S. F Between Groups 9.591 1 9.591 16.462 Within Groups 106.614 183 .583 Total 116.205 184 818. = .000 . ETA2 = .083 An analysis of variance completed on both sub- scales disclosed no interaction effect (type of institution by position held), but did reveal significant differences between the main effects in each subscale. These differen- ces are reflected in the breakdowns of each subscale as reported above. Summary A review of each of the seven null hypotheses tested in this study reveals the following summaries: I 1. Null hypothesis 1, which was used to compare the perceptions of faculty and administrators at high attrition colleges with those at low attrition colleges 100 concerning the factors associated with attrition, was not accepted for three subscales: Subscale 1, Counseling and Academic Services; Subscale 2, Student Problems or Goals; and Subscale 3, Nonacademic Institutional Services. Null hypothesis 1 was accepted for Subscale 4, Student Ability, Interest, and Preparation. 2. Null hypothesis 2, which was tested to compare the perceptions of faculty in high attrition colleges with those of administrators in high attrition colleges con- cerning the factors associated with attrition, was not accep- ted for one subscale: Subscale 4, Student Ability, Interest, and Preparation. Null hypothesis 2 was accepted for Sub- scale 1, Counseling and Academic Services; Subscale 2, Student Problems or Goals; and Subscale 3, Nonacademic Institutional Services. 3. Null hypothesis 3, which was used to compare the perceptions of faculty in low attrition colleges with those of administrators in low attrition colleges concerning the factors associated with attrition, was accepted for each of the four subscales. 4. Null hypothesis 4, which was tested to com- pare the amount of out of class student contact that faculty have in low attrition colleges with those in high attrition colleges, was accepted. 5. Null hypothesis 5, which was used to compare the perceptions of faculty and administrators at high attrition colleges with those at low attrition colleges 101 concerning the impact faculty may have in reducing attrition, was accepted for both subscales. 6. Null hypothesis 6, which was tested to com- pare the perceptions of faculty in high attrition colleges with those of administrators in high attrition colleges concerning the impact faculty may have in reducing attrition, was not accepted for either subscale: Subscale 5, Faculty Out of Class Impact--Advice, Instruction, and Supervision; and Subscale 6, Faculty Out of Class Impact--Personal, Social, and Special Interest Interactions. 7. Null hypothesis 7, which was used to compare the perceptions of faculty in low attrition colleges with those of administrators in low attrition colleges concerning the impact faculty may have in reducing attrition, was not accepted for either subscale: Subscale 5, Faculty Out of Class Impact--Advice, Instruction, and Supervision; and Subscale 6, Faculty Out of Class Impact--Personal, Social, and Special Interest Interactions. A review of the descriptive data in this study reveals the following highlights: 1. 92.2 percent of the faculty and administrators from the high attrition colleges (attrition rate greater than 60%) estimated the attrition rate in their institutions to be less than 60 percent. Although the respondents from high attrition colleges tended to underestimate the rate of attrition, 87.1 percent of them consider student attrition to be a problem in their institutions. In contrast, the 102 respondents from the low attrition colleges (attrition rate less than 50%) were considerably more accurate in their estimates of the rate of attrition, yet, 65.4 percent of them consider student attrition to be a problem in their institutions. 2. 92.2 percent of all respondents indicated that faculty members can play a role in reducing student attrition without lowering academic standards. 3. 59.7 percent of the respondents from low attrition colleges have held their current positions for seven or more years, while only 39.4 percent of those from the high attrition colleges have held their current positions for seven or more years. 4. 67.1 percent of the faculty respondents from low attrition colleges were tenured at the time of this study, while only 44.8 percent of the faculty respondents from the high attrition colleges held tenure. Chapter V reports the summary, conclusions and recommendations of the study. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS The problem of student attrition in higher education has been a subject of investigation for over fifty years and has been studied in numerous ways. Recently, the subject of attrition has received increased attention due to projections for declining enrollments and shifts in postsecondary enrollment patterns. In many colleges, improving retention is being equated with the survival of the institution. As the pool of traditional college students becomes smaller, the need to create a low attrition or staying environment becomes greater. Much of the research to date has been concerned with the characteristics of the dropout, has been limited in scope, and has neglected to present solutions to the problem. Only recently have investigators attempted to develop a conceptual perspective of attrition and models of student-institutional congruence. Several investigators have speculated that faculty members can play a major role in student retention programs and a number of specific functions have been suggested. Performance of these functions would require an increase in the quality and/or quantity of faculty-student inter- actions. To date, however, no research studies have been 103 104 located in which faculty members have been asked what impact they feel they can have in reducing student attrition. The purpose of this study was to ascertain the perceptions of faculty and administrators Concerning the factors associated with attrition, the amount of out of class contact that faculty have with students, and the impact that faculty and administrators feel that faculty can have in reducing attrition. This chapter presents a summary of the develop- ment of the study, its conclusions, recommendations and implications. Summary of the Development of the Study In the first chapter, the investigator explored the nature of the problem of attrition and developed the parameters of this study. The chapter included an intro- duction of the problem, the purpose of the study, the questions for investigation, the need for the study, an abbreviated review of related literature to support the need for the study, the hypotheses to be tested, the def- initions of terms used in the study, the limitations and assumptions of the study, and a brief section on the methodology that was employed in this project. The hypotheses generated for this study include: 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. 2. There are no significant differences in the 105 perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in high attrition colleges. 3. There are no significant differences in the perceptions of faculty and those of adminis- trators concerning the factors which affect student attrition in low attrition colleges. 4. There are no significant differences in the stated amount of out of class student contact that faculty have in high and low attrition colleges. 5. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact faculty can have in reducing student attrition. 6. There are no significant differences in the perceptions of faculty and those of adminis- trators in high attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. 7. There are no significant differences in the perceptions of faculty and those of adminis- trators in low attrition colleges concerning the impact faculty can have in reducing stu- dent attrition. The related and relevant literature was reviewed and reported in Chapter Two. The review of the literature included the following areas: (1) why study attrition, (2) the problem of definitions, (3) criticisms of attrition research, (4) factors associated with attrition, (5) rates of attrition, and (6) faculty impact upon student persis- tence. The review demonstrated the paucity of supporting research to answer the questions identified by the inves- tigator in Chapter One. The research methodology and design of the study were presented in the third chapter. For the purposes of 106 this investigation, a questionnaire was developed which included general institutional perceptions, student—related factors in attrition, institutional factors in attrition, faculty activities outside the classroom, impact of faculty activities in reducing attrition, and demographic infor- mation. The student-related and institutional factors in attrition were factor analyzed into four subscales for statistical analysis: (1) counseling and academic services, (2) student problems or goals, (3) nonacademic institutional services, and (4) student ability, interest and preparation. The faculty activities in reducing attrition were factor analyzed into two subscales: (1) faculty impact--advice, instruction and supervision, and (2) faculty impact--personal, social and special interest interactions. The questionnaire was administered to all faculty and administrators in twelve small, private, four-year liberal arts colleges. Each of the colleges was a partici- pant in the Strengthening Institutions Through Improving Retention (SITIR) project at the time of the study. Eight of the colleges were considered to be high attrition schools and four were low attrition schools. From a possible 659 respondents, 544 usable questionnaires were returned (391 faculty and 153 administrators). In Chapter Four, the results of the study were presented. Selected portions of the data were analyzed using SPSS Crosstabulations and Condescriptives for 107 descriptive purposes (general institutional perceptions and demographic information). Each of the hypotheses was tested using one way, fixed-effects analysis of variance. Hypotheses 1, 2, 3, 5, 6, and 7 were tested for each sub- scale appropriate to the hypothesis. The findings of this study are presented below: Findings Hypotheses l, 2, and 3 were tested on each of the four subscales concerning the factors associated with attri- tion. Hypothesis 4 was tested on a single scale concerning faculty-student contact outside the classroom. Hypotheses 5, 6, and 7 were tested on each of two subscales concerning the impact of faculty activities in reducing student attri- tion. Within the framework of the limitations of this study (described in Chapter I), several findings were made. Null Hypothesis 1. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the factors which affect student attrition in their institutions. This hypothesis was not accepted for Subscale 1 (Counseling and Academic Services), Subscale 2 (Student Problems or Goals), and Subscale 3 (Nonacademic Institu- tional Services). Analysis of variance applied to each of these subscales yielded significant differences (at the .05 level) between high and low attrition institutions. ‘Respondents from high attrition colleges attached greater 108 importance to the items in Subscales l and 2 as factors in attrition than did respondents from low attrition colleges. On Subscale 3, respondents from low attrition colleges perceived these items to be of greater importance in attrition than did respondents from the high attrition schools. Hypothesis 1 was accepted for Subscale 4 (Student Ability, Interest, and Preparation). The data revealed the following findings: 1. Based upon the perceptions of faculty and administrators, the inadequacy of counseling and academic services is significantly more important as a factor in student attrition in high attrition colleges than in low attrition colleges. 2. Based upon the perceptions of faculty and administrators, the inadequacy of nonacademic institutional services (primarily social factors) is significantly more important as a cause of student attrition in low attrition colleges than in high attrition schools. 3. Student problems and goals (including items such as financial difficulties, personal problems and change in personal or career goals) are perceived by faculty and administrators to be significantly more impor- tant as factors in attrition at high attrition colleges than at low attrition colleges. 4. Although there are no significant differences between the perceptions of faculty and administrators from 109 high attrition colleges and those at low attrition colleges concerning the items in Subscale 4 (Student Ability, Inter— est, and PreparationL,these items clearly are the most important factors in student attrition at the colleges studied; in that, the mean responses were considerably higher for these items than for any others listed. Null Hypothesis 2. There are no significant dif- ferences in the perceptions of faculty and those of administrators concerning the factors which affect student attrition in high attrition col- leges. Analysis of variance applied to each of the sub- scales yielded significant differences (at the .05 level) between faculty and administrators at high attrition colleges on Subscale 4 (Student Ability, Interest, and Preparation). As a result, null hypothesis 2 was not accepted for Subscale 4. It was accepted, however, for Subscale 1 (Counseling and Academic Services), Subscale 2 (Student Problems or Goals), and Subscale 3 (Nonacademic Institutional Services). The data revealed the following findings: 5. Faculty members and administrators from high attrition colleges generally agree upon the importance (or lack of importance) of most of the factors which affect student attrition in their institutions. 6. In factors concerning the ability, interest and preparation of students, however, faculty members in high attrition colleges perceive these items (Subscale 4) to 110 be significantly more important as causes of attrition than do administrators in the same colleges. 7. In spite of the differences between faculty and administrators in high attrition colleges concerning the factors in Subscale 4, both groups perceive these fac- tors as more important than any others listed as causes of student attrition. Null Hypothesis 3. There are no significant differences in the perceptions of faculty and those of administrators concerning the factors which affect student attrition in low attrition colleges. Analysis of variance applied to each of the four subscales concerning factors associated with attrition yielded no significant differences on any of the subscales. As a result, null hypothesis 3 was accepted for each sub- scale. These data revealed the following finding: 8. Although, when means are compared, faculty and administrators from low attrition colleges differ somewhat concerning their perceptions of the causes of student attrition, these differences are not statistically signi- ficant. Null Hypothesis 4. There are no significant differences in the stated amount of out of class student contact that faculty have in high and low attrition colleges. This hypothesis was tested using a scale of eight items (types of faculty-student contact) with analysis of 111 variance. No significant differences were found, so null hypothesis 4 was accepted. The data revealed the following finding: 9. Faculty members from low attrition colleges do not interact more frequently with students outside the classroom than faculty members from high attrition colleges. Null Hypothesis 5. There are no significant differences in the perceptions of faculty and administrators at high attrition colleges and those at low attrition colleges concerning the impact faculty can have in reducing student attrition. Analysis of variance applied to both subscales disclosed no significant differences (at the .05 level) between respondents from high attrition colleges and those from low attrition colleges. As a result, null hypothesis 5 was accepted for Subscale 5 (Faculty Out of Class Impact- Advice, Instruction, and Supervision) and Subscale 6 (Faculty Out of Class Impact--Personal, Social, and Special Interest Interactions). Mean scores for all groups, how- ever, were highest on Subscale 5. These data revealed the following finding: 10. Although there are no significant differences in the perceptions of faculty and administrators from high attrition colleges and those from low attrition colleges concerning either Subscale 5 or 6, all respondent groups indicated that faculty impact in reducing attrition can be considerably greater from the activities in Subscale 5 than from those in Subscale 6. 112 Null Hypothesis 6. There are no significant differences in the perceptions of faculty and those of administrators in high attrition colleges concerning the impact faculty can have in reducing student attrition. This hypothesis was not accepted for Subscale 5 (Faculty Out of Class Impact--Advice, Instruction, and Supervision) or for Subscale 6 (Faculty Out of Class Impact-~Personal, Social, and Special Interest Interac- tions). Analysis of variance applied to each of these subscales yielded significant differences (at the .05 level) between faculty and administrators in high attri- tion colleges. Administrators perceived potential faculty impact in reducing student attrition to be significantly greater than faculty perceptions of their own impact on both subscales. In responses from both groups, potential faculty impact was rated higher for Subscale 5 than for Subscale 6. These data revealed the following findings: 11. Administrators in high attrition colleges perceive that faculty can have a high degree of impact (mean = 4.11) in reducing student attrition through per- formance of functions related to out of class advice, instruction and supervision. In addition, administrators' perceptions of potential faculty impact are significantly higher than the perceptions of faculty members themselves. 12. Perceptions of administrators from high attri- tion colleges are significantly higher than those of 113 faculty members from the same institutions concerning the impact faculty can have in reducing student attrition through the performance of out of class personal, social and special interest interactions with students. _Null Hypothesis 7. There are no significant differences in the perceptions of faculty and those of administrators in low attrition colleges concerning the impact faculty can have in reducing student attrition. This hypothesis was not accepted for Subscale 5 '(Faculty Out of Class Impact--Advice, Instruction, and Supervision) or for Subscale 6 (Faculty Out of Class Impact-- Personal, Social, and Special Interest Interactions). Analysis of variance applied to each of these subscales yielded significant differences (at the .05 level) between faculty members and administrators at low attrition colleges. Administrators perceived potential faculty impact in reducing student attrition to be significantly greater than faculty perceptions of their own impact on both subscales. In responses from both groups, potential faculty impact was rated higher for Subscale 5 than for Subscale 6. These data revealed the following findings: 13. Administrators from low attrition colleges perceive that faculty can have a high degree of impact (mean = 4.13) in reducing student attrition through per- formance of functions related to out of class advice, instruction and supervision. In addition, administrators' 114 perceptions of potential faculty impact are significantly higher than the perceptions of faculty members themselves. 14. Perceptions of administrators from low attri- tion colleges are significantly higher than those of faculty members from the same institutions concerning the impact faculty can have in reducing student attrition through the performance of out of class personal, social and special interest interactions with students. Conclusions and Discussion The seven hypotheses in this study were tested using a variety of scales and subscales. In all, nineteen statistical tests were performed, of which, eight yielded significant findings. Of the eight significant results, three involved comparisons of faculty and administrators from low attrition colleges with those in high attrition colleges; three involved comparisons between faculty and administrators in high attrition colleges; and two involved comparisons between faculty and administrators in low attrition colleges. Prior to any discussion of these findings, it should be noted that a measurement problem may have existed which could potentially cloud the results. Only a small proportion of the variance was accounted for in each statistically significant result. The amount of explained variance (eta squared) between the variables is found in 115 each ANOVA table in Chapter Four. These figures reveal that less than ten percent of the variance was accounted for in each analysis of variance performed. A small eta squared could indicate one or more of the following: (1) the measurement instrument was somewhat unreliable; (2) the large sample size influenced the amount of account- able variance; and/or (3) the results are a reflection of the individual differences between respondents. This problem will be discussed further in the next section of this chapter (Implications of the Study). As a result of the findings regarding the factors which affect student attrition in these colleges, it was concluded that: 1. Faculty and administration perceptions regard- ing the causes of student attrition are similar to those reported by students in previous research, except that faculty members and administrators attach greater impor- tance to the academic inadequacies (ability, interest and preparation) of students as factors in attrition. 2. Faculty members and administrators from high attrition colleges are concerned about the extent to which the (lack of) quality of academic services provided in their institutions may be contributing to the attrition problem on their campuses. 116 3. Faculty members and administrators from low attrition colleges are more concerned about the adequacy of the nonacademic institutional services (primarily social factors) as causes of attrition than they are regarding the adequacy of the academic services which are provided on their campuses. In considering the specific factors which affect student attrition in the colleges studied, the results of this study are generally supportive of previous research. Astin found the primary causes of attrition to be boredom with courses, financial problems, dissatisfaction with requirements or regulations, and change in student goals.l Cope and Hannah report that students leave primarily because of personal, unknown, financial and academic reasons.2 In this study, most student attrition was perceived to be attributable to lack of interest or motivation (boredom), inadequate academic preparation, financial problems, low academic ability, and lack of personal/emotional adjustment to college. Previous research has been concerned with student- reported reasons for leaving college, which may explain why dissatisfaction, personal and unknown reasons are so apparent in the literature.3 This study supports the 1Astin, pp. cit., 1975. 2Cope and Hannah, op. cit. 3Ibid. 117 previous findings that boredom and money are among the most important causes of attrition; yet, it is interesting to note that academic inadequacies emerge as important factors as well. It appears that Marks was correct in suggesting that students report more socially acceptable reasons for leaving, while others (faculty and administrators in this case) indicate that student abilities and weaknesses play a great role in attrition.4 Whereas all respondent groups reported that factors concerning student ability, interest and preparation are the most important factors in attrition, faculty members in both high and low attrition colleges cited these factors to be of greater importance than did administrators. Because faculty members interact with students daily within the academic setting, they appear to be somewhat more critical of students' abilities than administrators are. In comparing high and low attrition colleges, the results of this study indicate significant differences con- cerning the institutional factors which affect student attrition. In high attrition colleges, there is signifi- cantly more concern about the quality of the academic ser- vices (such as adequacy of curricular offerings and quality 'of academic advising) than in low attrition schools. In some of the high attrition colleges, courses and/or majors 4Marks, op. cit. 118 have recently been eliminated due to reduced enrollments or other financial considerations. The ramifications of these decisions may have had some impact upon the percep- tions of the respondents from these schools. Conversely, in the low attrition colleges, social factors (such as insufficient extracurricular activities) play a significantly greater role in attrition than they do in high attrition colleges. Since two of the four low attrition schools are single-sex colleges, the concern for social factors is readily apparent. Faculty and administrators from high attrition colleges also differed significantly from those in low attrition schools in their perceptions concerning student problems and goals as factors in attrition. It appears that items such as financial difficulties, conflict with full time job, and student originally planned to drop out or transfer, are of greater importance as causes of attri- tion in the high attrition colleges studied. If these colleges are admitting a large number of students who hold full time jobs or students who plan to drop out or transfer, then it seems that the attrition rate can be expected to be higher than at colleges where these factors are not as commonplace. As a result of the findings regarding the amount of faculty-student out of class contact which occurs in these colleges, it was concluded that: 119 1. Although previous research has indicated that the amount of out of class contact that faculty have with students may be an important component of efforts to reduce attrition, the findings of this study do not support this notion. In this study, there were no differences in the amount of faculty-student interaction that occurred in the low and high attrition colleges. Previous literature has indicated that the amount of out of class contact that faculty have with students may be an important component of efforts to reduce attrition. In this study, comparisons were made between faculty at low attrition colleges and those at high attrition colleges to determine if differences existed between these groups in the amount of contact they had with students outside the classroom. Wilson pp_ppp developed a set of out of class faculty role capacities for determining the amount of faculty-student contact that occurred in six diverse types of institutions. They found that faculty engaged in out of class instruction and academic advising far more frequently than any other capacities. Career advising and social interaction ranked third and fourth in frequency of contact.5 The findings of this study generally support those of Wilson pp pp. in that out of class instruction and 5Wilson pp pp., pp. cit. 120 academic advising occurred more often than other types of interactions, although the order was reversed. (Some of the colleges studied were starting a new term when the questionnaire arrived, which could account for the large number of academic advising contacts.) Social interaction and career advising, respectively, ranked third and fourth in frequency of contact in this study, again reversing the order of the findings of Wilson pp pl. Faculty in this study, however, engaged in nearly double the number of social and career-related student contacts than were reported by Wilson pp pp.6 These findings support those of Kamens concerning the willingness of faculty in small colleges to interact informally with students.7 In comparing the amount of contact that faculty state they had with students, no differences were found between high and low attrition colleges; in fact, the two groups were virtually identical in all role capacities studied. One the surface, these findings tend to refute the notion that the amount of faculty-student interaction which occurs on a campus may be related to the attrition rate. To substantiate these findings, however, single campus studies need to be undertaken, including surveys of student perceptions and experiences. 6Wilson, pp pp., pp. cit. 7Kamens, pp. cit. 121 As a result of the findings regarding the amount of impact that faculty and administrators perceive that faculty may have in reducing student attrition, it was con- cluded that: 1. Faculty members and administrators from both low and high attrition colleges are confident that faculty can have considerable impact in efforts to reduce student attrition. 2. Although faculty members believe that they can have positive impact in retention efforts, they do not perceive their potential impact to be as great as adminis- trators perceive it to be. It has been suggested by some investigators that the type and quality of faculty-student interaction is important in any discussion of faculty impact in reducing student attrition. In this study, faculty and adminis- trators were asked their perceptions of the potential impact that selected faculty-student interactions may have in reducing attrition on their campuses. First, it should be noted that 78% of all faculty members surveyed considered student attrition to be a problem in their institutions (compared with 84% of the administrators). Also, 90% or the faculty respondents indicated that faculty can play a role in reducing student attrition without lowering academic standards (compared with 99% of the administrators). 122 Twelve out of class role capacities were identi- fied in which faculty may have impact upon student persis- tence. These roles were separated into two groups (sub- scales). Comparisons of the responses of faculty and administrators from low attrition colleges with those from high attrition schools yielded no significant differences on either subscale. However, comparisons between faculty and administrators disclosed significant differences on both subscales in low and high attrition colleges. In each case, administrator ratings of potential faculty impact were higher than those of faculty. It appears that even though faculty members are concerned about attrition and feel that they can have impact in reducing attrition, administrators clearly expect faculty to have substantially greater impact than faculty expect of themselves. These findings suggest the need for greater on-campus communi- cation between faculty and administrators concerning mutual expectations in a retention program. Three role capacities were identified by all respondent groups as having the greatest potential impact in reducing student attrition: academic advising, out of class instruction, and career advising. These roles were also among the tOp four in faculty reports concerning the frequency of student contact, as reported earlier. Pascarella and Terenzini reported that these same three faculty activities contributed most to discrimination 123 between persisters and leavers in their research.8 These findings suggest that institutions should strongly empha- size faculty knowledge and skill development in interactions with students concerning academic advising, out of class instruction and career advising. It should be noted that the significant differences which existed between faculty members and administrators in this study are actually insignificant in their relation- ship to the problem of student attrition per se. The dis- parity in faculty and administration perceptions regarding attrition are apparently indicative of problems of communi- cation and expectations within institutions. Further research is needed to explore the nature of the differences which exist between faculty and administration regarding the impact that faculty may have in a retention program. Implications of the Study The results and conclusions of this study have implications for the administrators and faculty members in the colleges studied, as well as for future investiga- tions concerning this topic. With this information, campus personnel may be better able to generate improved campus communication and programs concerning the nature of faculty activities in reducing student attrition. Other investi- gators may be able to use this information for the 8Pascarella and Terenzini, pp. cit. 124 development of single campus studies and for building models for faculty development programs. The implications of this study include the following: 1. Previous investigators have suggested the need to compare faculty perceptions simultaneously in different institutions, since there is so little information avail- able about faculty attitudes and activities, especially concerning faculty involvement in retention efforts. In this study, an attempt was made to find patterns of per— ceptions, as suggested by previous research. However, in light of the small proportion of variance accounted for in this study, it may be that one is comparing apples with oranges when comparing different institutions. Perhaps, single campus studies, smaller sample sizes, or a modifi- cation of the survey instrument would yield stronger assoc— iations among the variables. As Lee Noel stated "... a retention agent on one campus may be an attrition agent on another."9 2. Research studies have generated a plethora of reasons why students leave college. This study has revealed that academic factors may not have received an adequate amount of attention. Faculty and administrator perceptions seem to indicate that lack of academic ability and/or 9Noel, pp. cit., p. 34. 125 preparation are far more important than personal problems and other common student-reported reasons for dropping out. Colleges should strive to learn more about the real causes of attrition through research studies and improved exit- interview procedures. 3. Colleges should approach program reductions carefully. As enrollments decrease and the budget becomes tighter, programs are often reduced or eliminated. Respon- dents from.some high attrition colleges in this study indi- cate the possibility of a relationship between attrition and the curtailment of programs, but an uncertainty exists as to the order and magnitude of the cause and effect. 4. The concern for social factors as causes of attrition in the single sex colleges has implications for student affairs and other divisions of collegiate institu- tions. Traditionally, single sex colleges have greater holding power than coeducational schools, but there are indications that single sexness may be a factor in attri- tion. If so, greater efforts in coeducational social pro- gramming may be needed. 5. The faculty and administrators of the colleges in this study need to be aware of the goals and expectations of their incoming students. Other factors being equal, if a college recruits a large number of students who hold full time jobs, or students who expect to drop out or transfer, then the college can expect a higher than average dropout 126 rate. Within this framework, there are serious implica- tions for admissions officers regarding the number of students they need to recruit in order to maintain a student body of reasonable size. 6. The finding that there is no difference in the amount of faculty-student contact in high and low attrition colleges creates interesting implications con- cerning either the quality of that contact or the notion of the importance of faculty impact in student persistence. These results need to be compared with the perceptions of students in order to be more clearly understood. 7. Investigators have suggested that faculty members can play a role in reducing student attrition, yet no studies were located in which faculty were asked their perceptions of this notion. The finding that 90% of the faculty respondents feel that they can play a role in retention efforts without lowering academic standards is an interesting one. It appears that the faculty in these colleges are willing to help and feel that they can be of help. Presidents and chief academic officers should be encouraged by these findings, and should seek ways to initi- ate or increase faculty involvement in campus retention programs. 8. The differences between the perceptions of faculty and those of administrators concerning potential faculty impact in reducing attrition need to be studied 127 carefully. Do administrators overrate the abilities and impact of the faculty? Do faculty underrate these charac- teristics and associated outcomes? Why does this differ- ence exist, and what can be done to accurately determine characteristics and impacts? 9. The areas of greatest potential impact in reducing attrition found through this study are consistent 10 As the institutions of this with a previous study. study increase their efforts in faculty development programs, perhaps the areas of academic advising, out of class instruction and career advising should receive con- siderably more emphasis than in the past. Speculations A review of the results and conclusions of this study by the investigator revealed a few surprises, as well as several patterns of responses which could be accounted for through an understanding of specific insti- tutional situations. First, it should be noted that a surprisingly large number of these colleges may not survive the next few years, in spite of some very long standing traditions. Whereas improving retention has been equated with institutional survival in several research studies, improved retention is merely one factor which may help 10Pascarella and Terenzini, pp. cit. 128 some of these colleges continue to function. Reducing student attrition should be looked upon as a by-product of improved programs and services, not as a goal in itself. Yet, within some of these institutions, the methodology employed in retaining students is much more reactive than proactive. It is the belief of this investigator that the very real fear for institutional survival may have influ- enced the responses and perceptions of many of the respon- dents. In examining the factors which influence attrition in these colleges, there was a tendency among respondents to emphasize the lack of academic quality of the students as a primary reason for attrition. It may be that as the pool of applicants decreases, some of these colleges are presently admitting students who would have been rejected in the past. Several of these colleges have established basic skills, or remedial programs, which are nonexistent only a few years ago. In general, however, most of the students who attend these colleges are average to above average students. Since the curricular offerings are often narrow and courses and majors have been eliminated in several of these institutions, it was surprising to find that ques- tionnnaire items concerning the academic services of the colleges were not frequently reported as important factors in attrition. It would be interesting to compare the 129 perceptions of students who withdrew with those of faculty and administrators to determine whether or not students share similar perceptions regarding the adequacy of the academic services offered in these colleges. The great amount of out of class faculty-student contact reported in this study was of no surprise, in that many faculty and students do interact quite frequently in these small colleges. It would be interesting to determine, however, what percentage of the students (both dropouts and persisters) actually interact frequently with faculty, and what benefits are accrued from these interactions. In addition, as other investigators have suggested, different types of faculty may exhibit differing patterns and fre- quency of interaction. Therefore, it may be that only a small percentage of the faculty account for most of the interactions with students. One surprising result in this study was the con- sistency in which administrators perceived potential faculty impact in reducing attrition to be greater than the per- ceptions of faculty themselves. Again, within the frame- work of survival, administrators may be somewhat over- zealous in their perceptions of what faculty can actually do. As student affairs staffs and other administrative and support personnel become victims of budgetary con- straints, administrators (who remain) may be looking at faculty as institutional saviors. If such a speculation 130 could be based in fact, there would exist a great need for increased and more open communication between the faculty and the administration concerning expectations. Recommendations for Further Research The questions for investigation suggested in Chapter One of this study were answered. However, during the conduct of research, investigators often discover more questions than they could answer. The following sugges- tions for further research are based upon the review of the literature, the results of this study, and the conclu- sions drawn from this research: 1. In this study, perceptions were elicited from faculty and administrators concerning selected aspects of student attrition. The perceptions of students who dropped out or transferred from these colleges should be ascer- tained and compared with those of the faculty and adminis- tration in order to develop a more complete picture of the environment of these institutions. Do students leave these colleges for the same reasons that faculty and admin- istrators perceive? How much interaction with the faculty did the leavers have? What was the quality of this inter- action? What could have been done differently that would have helped these students to stay? 2. Research is needed concerning those students who persist and graduate from these colleges. How do their 131 perceptions of the campus environment compare with those of faculty and administrators? How much contact do these students state they have with faculty? What differences exist between the perceptions of persisters and those of dropouts regarding the institutional environment? 3. Due to the complexity of the attrition phenom- enon, along with other factors, there may have been some degree of a measurement problem in this study. The instru- ment used in this research needs to be tested further, perhaps within the framework of single campus studies. Single campus studies would allow for a smaller sample size, and include interviews with respondents to validate the findings, and other methodological considerations to insure that the investigator does not run the risk of comparing different sets of non-comparable phenomena which tend to exist in multi-institutional studies. 4. This study was undertaken in twelve small, private colleges, several of which are currently seeking ways to survive. Over the last few years, some larger colleges (even some state supported institutions) have closed their doors or merged with other institutions. Further research concerning perceptions of the causes and cures of attrition is needed in differing sizes and types of institutions in order for colleges to prepare themselves to meet the challenges of survival during the years ahead. 132 5. Research is needed on the local campus level regarding the goals, expectations and needs of matriculat- ing students. Colleges need to be able to-identify dropout- prone students, and need to develop the appropriate pro- grams and services to maximize the potential for success of those students. Additionally, colleges need to be able tO‘identify those students who intend to transfer, in order to help them make the transition from college to college as smooth as possible. Through the early identification of the likely dropouts and transfers, colleges will be better able to plan their enrollment maintenance strategies. 6. Because faculty-student interaction appears to be a potentially important factor in reducing student attrition, there is a need to investigate methods of faculty orientation and training which can improve the quality of these interactions. 7. Assuming that emphasis on faculty involvement in reducing attrition will increase, research is needed regarding institutional programs which prepare faculty for instructional responsibilities. What types of programs for faculty training currently exist, and what are the areas of emphasis of these programs? Should these training pro- grams include a component which will heighten faculty aware- ness and sensitivity regarding the problem of student attri- tion? 133 Concluding Statement This study was undertaken to determine the per- ceptions of faculty and administrators in twelve small, private, liberal arts colleges concerning selected aspects of student attrition in these institutions. Perceptions of faculty were compared with those of administrators, and perceptions from those in high attrition colleges were compared with those in low attrition colleges. It was found that differences did exist between these groups regarding the importance of factors believed to affect attrition and the potential impact faculty may have in reducing attrition. The findings and conclusions of this study may be helpful to faculty members and administrators as they search for the causes and cures of the problem of student attrition in their institutions. APPENDICES 134 Exhibit 1 Exhibit 2 APPENDIX A Instructions to Student Retention Officers Regarding Administration of Faculty/ Administration Perceptions Questionnaire Target Population for Faculty/Administration Perceptions Questionnaire 135 INSTRUCTIONS TO STUDENT RETENTION OFFICERS REGARDING ADMINISTRATION OF FACULTY/ADMINISTRATION PERCEPTIONS QUESTIONNAIRE November 16, 1979 TO: Student Retention Officers FROM: Dick Mullendore, SITIR Consultant SUBJECT: Instructions for Faculty/Administration Perceptions Questionnaire As a part of the campus research component of the SITIR project for 1979-80, each institution is being asked to participate in a study of Faculty and Administration Perceptions Concerning Faculty Involvement in Reducing Stu- dent Attrition. The purpose of the study is to provide information for each college concerning some of the atti- tudes and perceptions which exist on each campus. In addition, comparative data will be generated in order to see what differences exist from campus to campus. A report of the results will be distributed to each institution. The questionnaire takes no longer than 10-15 minutes to complete, and anonymity is guaranteed for each respondent. The questionnaire should be administered to the groups identified on the enclosed sheet. It is suggested that you follow these steps in order to receive the greatest response rate: I. For Faculty Respondents A) Administer the questionnaire at the next regular faculty meeting, and collect completed questionnaires pp that time. (If there is no faculty meeting scheduled in time to meet the return deadline-December 15- then follow the procedure outlined for administrators.) B) Send a copy of the questionnaire to each faculty member who is absent from the faculty meeting at which you administer 136 137 the instrument. C) Send a follow-up note after one week to all faculty to whom you sent the question- naire asking that those who have not returned it do so immediately. II. For Administration Respondents A) Hand deliver or send questionnaires to administrators with a note requesting that they return completed questionnaires B) Send a follow-up note after one week to all administrators who received copies of the instrument asking that those who have not returned it to do so immediately. III. Return of Questionnaires to SITIR Office A) Once you are satisfied that you have received as many questionnaires back as possible, place them in the envelope provided, and mail them pp time pp reach the SITIR office py December pp. (Remember, this is Christmas rush time.) If you have any questions or problems, please contact Dr. Barnett or me. I can be reached at: 11228 Evans Trail #102 Beltsville, MD. 20705 (301) 937-2394 TARGET POPULATION FOR FACULTY/ADMINISTRATION PERCEPTIONS QUESTIONNAIRE GROUP I: All Full Time Faculty Members GROUP II: Full Time Administrators Who Perform the Following Functions on Your Campus: 4) 5) 6) 7) 8) 9) 10) 11) 12) 13) 14) 15) 16) President of the college Chief academic officer Assistant academic officer (if employed full time in administrative capacity) Chief business officer Chief student affairs officer Admissions office director Financial aids office director Career counseling office director Student activities director Institutional advancement director (Development office) Public relations office director Alumni office director Library director Residence halls director Student retention officer Others as appropriate for your campus (Use your discretion) In some cases (perhaps many), one person performs more than one of the above functions, such as Director of Admissions and Financial Aids. It is not necessary for someone pp each office to respond, only those persons who are in charge of the functions listed. On your campus, you may have less than 15 administrators as possible respon- dents. 138 APPENDIX B Exhibit 1 Copy of Instrument Used for the Study "Faculty and Administration Perceptions Concerning Faculty Involvement in Reducing Student Attrition" 139 FACULTY AND ADMINISTRATION PERCEPTIONS CONCERNING FACULTY INVOLVEMENT IN REDUCING STUDENT ATTRITION This questionnaire is designed to elicit the perceptions of faculty members and administrators concerning some aspects of student attrition in their institutions. For the purpose of this study, the following term has been defined: ATTRITION: Any loss of students from a given college as a result of dropping out, stopping out. transferring, academic failure. or disciplinary suspension or expulsion. PART I: GENERAL INSTITUTIONAL PERCEPT IONS Please respond to the following questions or statements according to your opinion or perception. Use the following scale for questions 1 throum 3: 5 . STRONGLY AGREE 4 I AGREE 3 - NEITHER AGREE NOR DISAGREE 2 - DISAGREE ‘I - STRONGLY DISAGREE CIRCLE YOUR RESPONSE 5 4 3 2 1 I) I consider student attrition to be a problem at this institution. 5 4 3 2 1 2) The primary responsibility for reducing student attrition should rest with administrators. 5 4 3 2 1 3) Faculty members can play a role in reducing student attrition without lowering academic standards. 4) I estimate that the percentage of students who leave this college prior to completion of a degree is: n _ LESS THAN 209: 3) _ 41.809: 2) _ 21409: 4) _ 81809: 5) _ MORE THAN 80% 5) This institution has a formal program (or reducing smdent attrition. 1) YES ’ 2) NO 3) DO NOT KNOW IF YES, PLEASE ANSWER 5. AND Eb, OTHERWISE, GO ON TO PART II. 5a) The program (or reducing student attrition at this institution is effective. 1) __ YES 2) _ NO 3) DO NOT KNOW 5b) Faculty members are involved in the program for reducing attrition. i) __ YES 2) NO 3) _ DO NOT KNOW 140 8) 87 3| )3 I? 3. mmmmmmmmmmumm 141 PART II: STUDENT-RELATED FACTORS IN ATTRITION Students report many reasons for Ieavmg college prior to completing a degree. Some are listed below. In your opinion, of what importance is each of these factors in causing student attrition at your institution? 5 - OF VERY HIGH IMPORTANCE 2 ' OF LOW IMPORTANCE 4 - OF HIGH IMPORTANCE 3 OF MODERATE lMPORTANCE I - OF VERY LOW IMPORTANCE CIRCLE YOUR RESPONSE 3 2 1) Low academic ability. 2) Inadequate academic preparation for college. 3) Lack of interest or motivation. 4) Lack of academic challenge. 5) Financial difficulties. 8) Dissatisfaction with academic requirements or regulations. 7) Dissatisfaction with social requirements or regulations. .‘d-fl—Id-Odd 8) Lack of personal/emotional adiustment to college. 9) Change in personal or career goals. 10) Conflict with full time job. It) Personal problems. 12) Illness or accident. 13) Lack of peer group identification. mmmmmmmmmmmmmm bbb‘bbé‘bfibbb NUUHHUUUUQUNU ”IOMMMNNMNMNNN I4)‘ Student originally planned to drop out or transfer. 15) Other student-related factors (Please specify). PART III: INSTITUTIONAL FACTORS IN ATTRITION There are many institutional factors which may affect student attrition. Some of them are listed below. In your opinion. of what importance is each of these factors in causing student attrition at your institution? 3 - OF MODERATE IMPORTANCE I I s - or venv HIGH IMPORTANCE 2 - OF LOW IMPORTANCE l 4 - or HIGH IMPORTANCE I - OF VERY LOW IMPORTANCE I CIRCLE YOUR RESPONSE I) Curricular offerings too narrow. (a) N 2) Courses or majors eliminated. 3) Inadequate academic advising. 4) Lack of flexibility in planning personalized academic programs. 5) Low instructional quality. a) ' Low academic ability of students admitted to this coilege. 7) Inadequate personal counseling. 8) Social policies too restrictive. 9) Social policies too liberal. 10) Lack of extracurricular activities. II) Inadequate studen' financial aid program. 12) Poorly maintained physical facilities. ‘bbb..b§“.& WUUUUUMQQQUU NNNNNMktNNNNN dad—eddd—ed—a—sdd 13) Inadequate career counseling. 14) Other institutional factors (Please specify), CO 43 4‘ CC 47 142 PART IV: FACULTY ACTIVITIES OUTSIDE THE CLASSROOM (FACULTY ON LY RESPOND TO THIS SECTION - ADMINISTRATORS GO TO PART V) Faculty members have a variety of contacts with students outside the classroom. Please try to estimate the number of times in the past two weeks that you have interacted with students in each of the following capacities. Count only conversations or activities which were substantive in time or content. assistants.“ W... mayhem CIRCLE YOUR RESPONSE 5 4 3 2 I I) ACADEMIC ADVISOR: To give a student information or advice about his/her academic program. 5 4 3 2 I 2) CAMPUS CITIZEN: To dimss a campus issue or problem with a student. 5 4 3 2 1 3) CAREER ADVISOR: To help a student consider matters related to his/her future career. 5 4 3 2 I 4) OUT OF CLASS INSTRUCTOR: To discuss intellectual or academic matters with a student. 5 4 3 2 I 5) PARTICIPANT: To attend or participate in school events. 5 4 3 2 I 6) PERSONAL COUNSELOR: To help a student resolve a personal problem. 5 4 3 2 I 7) REFERRAL AGENT: To direct a student to anather person or office for information or assistance. 5 4 3 2 I 8) SOCIAL INTE RACT OR: To socialize informally with a student. 9) Do you serve as a sponsor or advisor to a student club or organization? 1) YES 2) NO 10) Do you serve on a campus committee that has student representation? 1) YES 2) NO 11) Do you serve as a practicum/intemship supervisor? 1) YES 2) NO 12) Do you supervise any students in an independent study? I) YES 2) NO 13) Please list any other out of class student-related activities in which you are involved. PART V: IMPACT OF FACULTY ACTIVITIES IN REDUCING ATTRITION (FACULTY AND ADMINISTRATORS RESPOND TO THIS SECTION) Listed below are several activities which are performed by some faculty members. Can performance of these activities by faculty be helpful in reducing student attrition? Please-rate these activities according to your perception of their potential impact in reducing student arm'rion at your institution. 5 I VERY HIGH IMPACT 2 I LOW IMPACT 4 I HIGH IMPACT 3 MODERATE 'MPACT I I VERY LOW IMPACT - --——--—J CIRCLE YOUR RESPONSE 5 4 3 2 I I) ACADEMIC AOVISOR: To give students information or advice abOut their academic programs. 5 4 3 2 I 2) CAMPUS CITIZEN: To discuss campus issues or problems with students. 5 4 3 2 I 3) CAREER ADVISOR: To help students consider matters related to their future careers. O7 .7 143 PART V: IMPACT OF FACULTY ACTIVITIES IN REDUCING ATTRITION (CONTINUED) UIUIUIUI ‘ r555. 5 - VERY HIGH IMPACT 4 I HIGH IMPACT 3 - MODERATE IMPACT 2 - LOW IMPACT I I VERY LOW IMPACT 2 I 4) CLUB SPONSOR: To serve as an advisor or sponsor of a student club or organization. 2 I 5) COMMITTEE MEMBER: To serve on a campus committee that has student representation. 3 2 I 6) INDEPENDENT STUDY SUPERVISOR: To supervise students during an independent study experience. 3 2 I 7) OUT OF CLASS INSTRUCTOR: To discuss intellectual or academic matters with students 3 2 I 8) PARTICIPANT: To attend or participate in school events. 3 2 I 9) PERSONAL COUNSELOR: To help students resolve personal problems. 3 2 I I0) PRACTICUM SUPERVISOR: To supervise students during a practicum or internship experience. 3 2 I II) REFERRAL AGENT: To direct students to another person or office for information or assistance. 3 2 I I2) SOCIAL INTERACT OR: To socialize informally with students. PART VI: DEWGRAPHIC INFORMATION I) 2) 3) 4) 5) B) What is your position? (Your current. primary responsibility) I) __ HUMANITIES FACULTY 4) __ PROFESSIONAL-APPLIED FIELDS FACULTY 2) SOCIAL SCIENCES FACULTY 5) _ ADMINISTRATOR 3) _ NATURAL SCIENCES FACULTY 6) __ OTHER (PLEASE SPECIFY) Number of years you have held current. primary position at this institution: I) _ LESS THAN ONE YEAR 3) _ 3-6 YEARS 2) _ 1.2 YEARS 4) __ 7.12 YEARS 5) __ MORE THAN 12 YEARS Highest degree you have earned: I) BACHELOR'S 3) DOCTORATE 2) MASTER‘S 4) __ OTHER (PLEASE SPECIFY) FACULTY: Are you tenured? (If your institution does not grant tenure, please leave this item blank.) I) __ YES 2) NO FACULTY: Number of students for whom you are serving as academic advisor: I) NONE 3) I I-20 2) __ I-IO 4) __ 21-30 5) _ MORE THAN so FACULTY AND ADMINISTRATORS: Your total number of years of full time postsecondary teaching experience. I) __ LESS THAN ONE YEAR 3) __ 36 YEARS 2) _ 1-2 YEARS 4) __ 7.12 YEARS 5) _ MORE THAN I2 YEARS BIBLIOGRAPHY 144' BIBLIOGRAPHY Astin, A.W. Preventing Students from Dropping70ut. San Francisco: Jossey-Bass, 1975. Astin, A.W. Four Critical Years: Effects of College on Beliefs, Attitudes, and Knowledge. San Francisco: Jossey-Bass, 1977. Burton, R. and Johnson, P. "Identifying Potential Dropouts with Class Lists." Improving College and University Teaching 17 (Summer 1969): 178-179. Chickering, A.W. ”Student-Faculty Relationships: Bedrock for College Governance." Revised version of comments addressed to the 15th Annual Institute on College and University Administration, University of Michigan, Ann Arbor, June 1969. Cope, R.G. "Why Students Stay, Why They Leave." In L. Noel (Ed.), New Directions for Student Services: Reducing the Dropout Rate, No. 3. San Francisco: Jossey-Bass, 1978. Cope, R.G. and Hannah, W. Revolving College Doors: The Causes and Consequences of Dropping Out, Stopping Out, and_Transferring. New York: John Wiley and Sons, 1975. Eckland, B.K. "College Dropouts Who Came Back." Harvard Educational Review 34 (No. 3, 1964): 402-420. Feldman, K.A. and Newcomb, T.M. The Impact of College on Students: An Analysis of Four Decades of Research, Vol. 1. San Francisco: Jossey-Bass, 1969. Gamson, Z. "Utilitarian and Normative Orientations toward Education." Sociology of Education 39 (1966): 46-73. Gekoski, N. and Schwartz, S. "Student Mortality and Related Factors." Journal of Educational Research 54 (January 1961): 192-194. 4‘— Hays, W.L. Statistics for the Social Sciences. New York: Holt, Rinehart and Winston, Inc., 1973. 145 146 Hoyt, D.P. "A Retrospective and Prospective Examination of Retention-Attrition Research." In L. Noel (Ed.), New Directions for Student Services: Reducing the Dropout Rate, No. 3. San Francisco: Jossey-Bass, 1978. Hunter, J.E. and Gerbing, D.W. "Unidimensional Measurement and Confirmatory Factor Analysis." Occasional Paper No. 20, Institute for Research on Teaching, East Lansing, Michigan, May, 1979. Husband, R.L. "Significant Others: A New Look at Attrition." Paper presented at Seventh Annual Meeting on Future Solutions to Today's Problems, sponsored by the Association for Innovation in Higher Education, Philadelphia, February, 1976. Iffert, R.E. Retention and Withdrawal of College Students. U.S. Office of Education Bulletin 1958, No. 1. Washington, D.C.: U.S. Government Printing Office, 1957. Kamens, D.H. "The College Charter and College Size: Effects on Occupational Choice and College Attrition." Sociology of Education 44 (Summer 1971): 270-296. Katz, J. and Sanford, N. "Curriculum and Personality." In N. Sanford (Ed.), College and Character. New York: John Wiley and Sons, 1964. Knoell, D.M. "A Critical Review of Research on the College Dropout." In L.A. Pervin, L.E. Reik, and W. Dalrymple (Eds.), The College Dropout and the Utilization of Talent. Princeton, N.J.: Princeton University Press, 1966. Marks, E. "Student Perceptions of College Persistence, and Their Intellective, Personality and Performance Correlates." Journal of Educational Psychology 58 (No. 4, 1967): 210-221. Morehead, C.G. and Johnson, J.C. "Some Effects of a Faculty Advising Program." Personnel and Guidance Journal 43 (October 1964): 139-144. Neddy, J.C. "A Review of Research Concerning College Stu- dent Attrition." Mimeographed for the Retention Improvement Task Force, State University of New York, Buffalo, 1977. 147 Nie, N.H., Bent, D.H. and Hull, C.H. Statistical Package for the Social Sciences. New York: McGraw-Hill Book Co., 1970. Noel, L. "College Student Retention-~A Campus-Wide Responsibility." Journal of the National Association of College Admissions Counselors 21 (July 1976): 33-36. Orlich, D.C. Designing Sensible Surveys. Pleasantville, New York: Redgrave Publishing Co., 1978. Panos, R.J. and Astin, A.W. "Attrition Among College Students." American Educational Research Journal 5 (January 1968): 57-72. Pantages, T.J. and Creedon, C.F. "Studies of College Attrition: 1950—1975." Review of Educational Research 48 (Winter 1978): 49-101. Pascarella, E.T. "Student-Institutional Congruence: A Student Perspective." College Student Journal 8 (1974): 78-87. Pascarella, E.T. and Terenzini, P.T. "Patterns of Student- Faculty Informal Interaction beyond the Classroom and Voluntary Freshman Attrition." Journal of Higher Education 48 (September/October 1977): 540—552. Pervin, L.A., Reik, L.E. and Dalrymple, W. (Eds.), "The Dropout in Conflict with Society." The College Dropout and the Utilization of Talent. Princeton, N.J.: Princeton University Press, 1966. Rossman, J.E. "An Experimental Study of Faculty Advising." Personnel and Guidance Journal 46 (October 1967): 160-164. Rossman, J.E. "Released Time for Faculty Advising: The Impact on Freshmen." Personnel and Guidance Journal 47 (December 1968): 358-363. Shulman, C.H. "Recent Trends in Student Retention." ERIC Higher Education Currents, American Association for Higher Education, (May 1976): 1-4. Spady, W.G. "Dropouts from Higher Education: Toward an Empirical Model." Interchange 2 (No. 3, 1971): 38-62. 148 Stark, J.S. "The Relation of Disparity in Student and Faculty Educational Attitudes to Early Student Transfer from College." Research in Higher Education 3 (1975): 329-344. Summerskill, J. "Dropouts from College." In N. Sanford (Ed.), The American College. New York: John Wiley and Sons, 1962. Tinto, V. "Dropout from Higher Education: A Theoretical Synthesis of Recent Research." Review of Educational Research 45 (Winter 1975): 89-125. Vreeland, R.S. and Bidwell, C.E. "Classifying University Departments: An Approach to the Analysis of their Effects upon Undergraduates' Values and Attitudes." Sociology of Education 39 (1966): 237-254. Webb, S.C. "Faculty Contributions to Dropouts." In 0. Milton (Ed.), Proceedings: A Conference on Student Retention in Tennessee Colleges and Universities, University of Tennessee, Knoxville, March, 1966: 1-14. Wilson, R.C., Gaff, J.G., Dienst, E.R., Wood, L. and Bavry, J.L. College Professors and Their Impact on Students. New York: John Wiley and Sons, 1975. "I(IIIIIIIIIIIIIIIIIII“