h. mm; .. . . t I .l, n ‘_ u-zu'JIV 'l'qr?‘ 7 but“ .‘ h---" ;'.v~~ .u. f .w...’., o w A. -u. V'.‘ v, .v o ». . I. ‘5‘“ 1-. _“ .v u- v, I“ 1' ‘7'“12 .~ rm ,, u ‘o-r n a 5»: v. a. y: u_; r»- .. ' flex '1}. , ' n n ,. 1.4 ._ or , '1', q . 9.1! o -‘-'."' . . . m . '35! VJ; ’1' . r. (a. 1004,! .u / . V .4. 2 {37‘ 'M‘ ‘m Ki .2" 'f,‘ u! 1’56“: ’ ’ vm.» .m *5 ._~ ! o w... I: TIP. 33:1! "“ ‘ “ v ‘0‘ ”fix? w v. ,1 w . 3'4”" .‘3‘ 4»:‘ Hum» .‘ .r‘ “a, ' ‘ x 4,. '7‘ s a. ' .., rw v... 7. .1“: 'v ‘J-u .5” I. 'r- r-y- new . ‘l>li‘| 1' ' ”Mist." ' w" . W! .1 J ~ n a": ,t. - P¢'~’ wv . L » u r 233" , . a ' «3’3" , 1’1",“ .1- . \'\ .93 .w , ‘ A... riodl'..- - I...” w», . .1 viii“ HIT»! . .I n1 “.1 ”‘2’ _ "If tr; .1 1: ,fi;“‘ u'" "' M .‘i 1.; 1‘1? x 1: .«r ufk . t‘ ”L .4 an‘ .. , "53’: .7 ”my. a. a... “.x ’ 'gvh 35'. .1 -‘ g n.3,, r .. - m6..~.: hwy". ’ "w, vn. , 1w '4; ,._,, .. _,,.. -.~-. w... '"v'. v ' t 0‘. I .. r“; .... _. p... ”3' 'v.y' .. - -'" v .1 .L m.) .- m..- -<‘-' E: ’ v—V . .... n.“ ....‘.. I": 2-2% . ow - . , J‘ V . '31-".- , 1‘ 3:; r“ ‘ak‘. L “‘h‘fiMm y 1;! 1‘“ gram» .,_ A .I :5» an r "V. .‘12 ...,.., w“ \- 4. (:1. N'k.‘ J; x ”-m. r ,7 «7., i 7 99*.” ,_ a”. €131 H v.6 3. 3 5;" flwifi‘v’; :13 ‘ 13-- n.3,? n3. (7 V“.- "We? ‘7 v --.-. ‘5' 9“ m . v. .- \ a". nu». xvi-hag“ .»-.E' ....,.u.. -7. 4, I w -I‘n “rocr . o - i f F: surf win-é; 1kg." m... 5.. try: » » :wa -- -; " v. m um... u-w-o... .. . w. 1 m .4 lllllllllllllllllllllll}Illllllllllllllllllllllllllllll 131293 This is to certify that the dissertation entitled Grant-Sponsored Academic Support Services: Do They Improve Community College Special Population Students' Academic Achievement and Persistence? presented by Patricia Ann Graves has been accepted towards fulfillment of the requirements for Doctor of Philosophy degreein Educational Administration ut\f /€C" //I ’ Majitf professor Dr. Eldon Nonnamaker Date April 23, 1992 MS U is an Affirmative Action/Equal! Opportunity Institution 0-12771 l LIBRARY 1 Michigan State 1 University PLACE IN RETURN BOX to remove this chockom from your record. TO AVOID FINES return on or before date duo. DATE DUE DATE DUE DATE DUE W l | L_J JEL_| IDl—i: '_J_u7 J MSU In An Affirmative Action/Equal Opportunity institution cMMmS-nt GRANT -SPONSORED ACADEMIC SUPPORT SERVICES: DO THEY IMPROVE COMMUNITY COLLEGE SPECIAL POPULATION STUDENT S' ACADEMIC ACHIEVEMENT AND PERSISTENCE? BY Patricia Ann Graves A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Educational Administration 1992 ABSTRACT GRANT-SPONSORED ACADEMIC SUPPORT SERVICES: DO THEY IMPROVE COMMUNITY COLLEGE SPECIAL POPULATION STUDENTS ACADEMIC ACHIEVEMENT AND PERSISTENCE? By Patricia Ann Graves The purpose OI this study was to determine if grant-sponsored academic support services were used by eligible community college special population stu- dents and if these services improved special population students' academic achieve- ment and persistence. A secondary focus of the study was to estimate the financial impact on a community college's budget as a result of providing academic support services to special population students. The research methodology was a quasiexperimental design. An experimen- tal group of 200 randomly sampled special population students enrolled at Delta College and a control group of 200 students with similar characteristics were sur- veyed to determine their usage of institutional and grant-sponsored academic sup- port services during the Winter 1991 semester. An institutional group of 500 stu- dents was randomly selected to permit statistical comparisons to the overall student body. but these students were not surveyed. institutional data was obtained to provide information on experimental. control, and institutional group students' aca- demic achievement and persistence. The three groups were compared to determine Patricia Ann Graves if differences existed between: experimental and control group students; experimen- tal and institutional group students; and control and institutional group students. This study reinforced previous findings that academic support services are underutilized by the students who report needing assistance to achieve their educa- tional goals. Experimental group students reported using grant-sponsored academic support services approximately 1 hour during the WInter 1991 semester. it appears that the experimental group students' academic achievement was significantly lower than control and institutional group students; experimental group students completed significantly fewer courses during the Winter 1991 semester than students in the control and institutional groups; and experimental group students persisted to enroll in Fall 1991 semester courses at the same rate as control and institutional group students. In fact, control group students whose characteristics were ”matched” with experimental group students and did not access supplemental academic support services "mirrored” the institutional group. 7 it was determined that there was a significant cost incurred by the community college for providing the supplemental academic support services to special popula- tion students. A variety of suggestions were offered on how community colleges can deter- mine if grant-sponsored academic support services benefit special population stu- dents' academic achievement and persistence and the financial impact on the institution's budget. Copyright by PATRICIA ANN GRAVES 1992 To my father, William D. Graves. Thanks dad, for your love, support, guidance and understanding. it's finally done! ACKNOWLEDGMENTS- There are several individuals who have made significant contributions to this study and should be publicly recognized. At Michigan State University, my advisor, Dr. Eldon Nonnamaker, has encouraged me since my initial enrollment in the pro- gram and has provided a considerable amount of assistance and support during the research study. My committee members, Dr. Marvin Grandstaff, Dr. Louis Hekhuis, and Dr. Fred Whims, have been very supportive and helpful by being accessible, Offering suggestions to strengthen the study, and providing encouragement. My co-workers and friends at Delta College have also been helpful in assist— ing me to complete the study. In particular, Dr. Gene Packwood, the Director of Research and Development, was very generous with his time and expertise in assisting me to design the questionnaires and conduct the statistical analysis. Jeannie Laverty, the Research Specialist, was helpful in assisting me to use the SPSS software to analyze the data. Lorraine Priebe generously assisted me with data entry and to locate all of the elusive data necessary to successfully complete the study. Two of Delta College's computer specialists, Keith DeLong and Rich Adams, developed computer programs to obtain the necessary institutional data. The staff of the Desktop Publishing Department, especially Sue Dunckel, were very patient and helpful in creating the final document. I would also like to thank my other friends and colleagues at Delta College who have been very supportive and understanding during my entire program, but particularly during the last six months as i focused on a singular topic. i appreciate their support, understanding and friendship. iv TABLE OF CONTENTS LIST OF TABLES .............................................................................................. viii Chapter Page I. lNTRODUCTION TO THE STUDY ...................................................... 1 Statement of the Problem ............................................................. 3 Purpose of the Study ............................................................. 4 Research Questions .............................................................. 5 Research Hypotheses of the Study .............................................. 5 Importance of the Study ........................................................ 7 Delimitations, Limitations, and Generalizability ............................ 8 Definition of Terms ........................................................................ 9 The Study Setting ......................................................................... 11 Organization of the Study ............................................................. 12 ll. REVIEW OF RELATED LITERATURE ................................................ 13 Historical Origins of the Community College ................................ 13 Demographics/Community College Students ............................... 18 Federal Legislation impacting Community Colleges .................... 20 State Legislation Impacting Community Colleges ........................ 23 Retraining the Unemployed ................................................... 24 Student Persistence in Postsecondary Education ........................ 26 Student Persistence in Community Colleges ........................ 31 Academic Support Services ......................................................... 38 Counseling/Academic Advising ............................................. 38 Tutoring ................................................................................. 49 Cost implications of Academic Support Services ......................... 53 Cost-Benefit Analysis ............................................................ 54 Cost-Effectiveness Analysis .................................................. 56 ill. METHODOLOGY ................................................................................ 61 Research Methodology of the Study ............................................ 61 Research Population ............................................................. 62 Hypotheses of the Study ....................................................... 66 V Page Distribution of Questionnaires ............................................... 67 Design of the Questionnaires ................................................ 69 Confidentiality of Responses ................................................. 70 Collection of the Data ............................................................ 71 Processing of the Data ........................................................... 73 Statistics ................................................................................ 73 Financial Impact Component - Data Collection ..................... 75 Summary ...................................................................................... 79 IV. RESULTS OF THE DATA ANALYSIS ................................................. 81 Introduction ................................................................................... 81 Results of the Study ...................................................................... 82 Use of Academic Support Services .............................................. 82 Institutional Resources .......................................................... 82 Grant-Sponsored Resources ................................................. 86 Academic Achievement ................................................................ 88 Winter 1991 Semester Grade Point Average ........................ 88 Academic Persistence .................................................................. 93 Completion of \Mnter 1991 Semester Courses ..................... 93 Enrollment in Fall 1991 Semester Courses ........................... 96 Cost Analysis ................................................................................ 99 Institutional Counseling Services ........................................... 100 Institutional Tutoring Services ............................................... 101 Grant-Sponsored Academic Support Services .................... 102 Total Academic Support Services ......................................... 103 Summary ...................................................................................... 106 V. SUMMARY, CONCLUSIONS, AND lMPLICATIQNS FOR PRACTICE AND FOR FURTHER RESEARCH .................................. 111 Summary ........................ 111 Conclusions .................................................................................. 1 17 Implications for Practice ............................................................... 122 Implications for Further Research ................................................ 124 APPENDICES .................. . .................................................................................. 1 27 A. QUESTIONNAIRE AND LETTER SENT TO EXPERIMENTAL GROUP STUDENTS ........................................................................... 128 B. QUESTIONNAIRE SENT TO CONTROL GROUP STUDENTS ........................................................................................ 132 vi Page C. FOLLOW-UP CORRESPONDENCE SENT TO EXPERIMENTAL AND CONTROL GROUP STUDENTS ................................................ 135 D. PERMISSION LETTER FROM UNIVERSITY COMMITTEE ON RESEARCH INVOLVING HUMAN SUBJECTS ............................ 138 E. RESPONSES TO OPEN-ENDED QUESTIONS ................................. 140 BIBLIOGRAPHY ............................................................................................... 145 vii LIST OF TABLES Table Page 1. Response Rate to Survey .................................................................... 69 2. Response Rate by Group .................................................................... 69 3. Institutional Counseling Services Accessed During the Winter 1991 Semester .................................................................................... 83 4. Students' Perceptions of the Benefits of Institutional Counseling Services Accessed Winter 1991 Semester .......................................... 83 5. institutional Tutorial Academic Support Services Accessed Winter 1991 Semester .................................................................................... 84 6. Students' Perceptions of the Benefits of Institutional Tutoring Services Accessed Winter 1991 Semester .......................................... 85 7. Total Institutional Academic Support Services Accessed Winter ' 1991 Semester .................................................................................... 86 8. Total Institutional and Grant-Sponsored Academic Support Services Accessed Winter 1991 Semester ........................................................ 87 9. Students' Perceptions of the Benefits of Grant-Sponsored Academic Support Services ................................................................................. 88 10. Winter 1991 Semester Mean Grade Point Average Experimental and Control Group Students ................................................................ 89 11. ASSET Scores for Experimental and Control Group Students Writing Skills Subtest ........................................................................... 90 12. ASSET Scores for Experimental and Control Group Students Reading Skills Subtest ......................................................................... 90 viii 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. ASSET Scores for Experimental and Control Group Students Numerical Skills Subtest ..................................................................... Winter 1991 Semester Mean Grade Point Average Experimental and Institutional Group Students ........................................................ Winter 1991 Semester Mean Grade Point Average Control and Institutional Group Students ............................................................... Winter 1991 Semester Course Completion Rate for Experimental and Control Group Students ............................................................... Winter 1991 Semester Course Completion Rate for Experimental and Institutional Group Students ........................................................ Winter 1991 Semester Course Completion Rate for Control and Institutional Group Students ............................................................... Fall 1991 Semester Enrollments Experimental and Control Group Students .............................................................................................. Fall 1991 Semester Enrollments Experimental and Institutional Group Students ................................................................................... Fall 1991 Semester Enrollments Control and lnstitutionaIGroup Students .............................................................................................. Estimated Cost of Providing Institutional Counseling Services .......... Estimated Cost of Providing Institutional Tutoring Services ............... Estimated Cost of Providing Grant-Sponsored Services to Special Population Students ............................................................................ Total Mean Student Cost of Academic Support Services Winter 1991 Semester ................................................................................... Academic Support Services Vlfinter 1991 Semester by Group and Funding Source .................................................................................. Page 91 92 93 94 95 96 97 98 99 101 102 103 104 105 CHAPTER I INTRODUCTION TO THE STUDY Community colleges began in the late 18003 as junior colleges whose pri- mary responsibility was to provide 18-22 year old students with low cost, quality educational opportunities within commuting distance of the student's residence. Students primarily enrolled in two-year academic programs and transferred to four- year institutions to complete their degree. Community colleges were further legitimized by the President's Commission on Higher Education (1948), which endorsed the institutions as a viable means of increasing the accessibility of higher education to the general population. Promoting the ”Open-door” philosophy, community colleges enrolled students regardless of educational background or credentials in a variety of academic transfer programs, occupational/vocational programs, general education courses, and community education activities. Prior to the 1970's, there was a relatively large labor market demand for low-skilled workers, primarily in the manufacturing sector. However, as the manu- facturing sector became more technologically advanced and downsized during the 19803, the demand for skilled workers increased. As a result, thousands of low- skilled and educationally underprepared individuals who relied on the manufacturing sector for high paying, lifetime employment became dislocated workers. At the 1 same time, many educationally underprepared individuals who were previously able to enter the manufacturing sector workforce directly upon high school graduation (or dropout) were presented with limited, low wage employment opportunities. As a result of the shifting workforce, many low-Skilled individuals enrolled at a community college to obtain the educational and technical skills necessary to initially enter or reenter the labor market. As the unemployment rate increased, so did community college enrollments. Both the federal and state governments authorized grant funding to provide financial aid to offset the cost of tuition, books, and fees for special population stu- dents to attend community colleges to Obtain training or retraining. In addition, legislatures also approved grant funding to community colleges to operate academic and student support services such as counseling/academic advising, assessment, tutoring, and developmental/remedial education to enable special population stu- dents to be academically successful and achieve their educational goals. For example, federal legislation has provided funding to assist special popula- tion students through the Carl D. Perkins Vocational Education Act. Carl D. Perkins grant funding has provided special population students enrolled in Michigan commu- nity colleges with $2,331,949 in Special Needs funding and $2,120,533 in Single Parent/Homemaker and Sex Equity grant funding to support tuition expenses and provide academic and student support services during the 1990-91 fiscal year (Michigan Department of Education). In addition, during the 1990-91 fiscal year, the State Of Michigan has provided $2 million to fund a state-wide Michigan Job Oppor- tunity Bank-Refrain Program and an additional $2 million to operate At-Risk Student Success programs at community colleges. Both of these programs were designed to support the enrollment and enhance the academic success and persistence of special population students (State of Michigan 1990-91 Fiscal Year Budget). Willem Community college administrators, legislators, and policy makers have real- ized that in contrast to recent high school graduates, many special population stu- dents require additional academic and student support services to overcome previ- ous educational deficiencies and successfully achieve their educational goals. Fed- eral and state legislatures have appropriated grant funds to provide supplemental counseling/academic advising, assessment, tutoring, developmental/remedial edu- cation, and other academic support services to improve special population students' academic achievement and persistence. Through grants, federal and state legislatures have provided funding to community colleges to provide supplemental academic support services to assist special population students to successfully achieve their academic objectives. How- ever, community colleges operating grant-sponsored academic support services for special population students have been required to submit limited data to the funding agencies indicating student use of these services and student outcomes. Limited data exists to indicate if these grants have any impact on students' academic perfor- mance. Therefore, it was important to determine if eligible special population stu- dents accessed grant-sponsored supplemental academic support services and if these services had a positive impact on special population students' academic achievement and persistence. Further, Since community college enrollments of special population students is expected to increase WW5 February 1988), an understanding of their enrollment on an institution's budget is important to future resource and financial planning, both by external agencies and college administrators. Wild! The purpose of this study was to determine if academic support services (primarily counseling and tutorial services) provided through grant funded programs (Single Parent/Homemaker Program, Special Needs Program-disadvantaged com- ponent, Michigan Job Opportunity Bank-Retrain Program, and At-Risk Student Success Program) were used by eligible community college special population students and if these services improved special population students' academic achievement and persistence. The research methodology used was a quasiexperimental design that compared three groups of students. The experimen- tal group consisted of 200 randomly sampled special population students enrolled at Delta College during the Winter 1991 semester and the control group was 200 ”matched pairs” students enrolled during the Winter 1991 semester with characteris- tics similar to experimental group students. An institutional group of 500 students enrolled during the Vlflnter 1991 semester was selected by random sample to permit statistical comparisons to the overall Delta College student body. Institutional group students' characteristics were not "matched" with either experimental or control group students. A secondary focus of the study was to estimate the financial impact on a community college's budget as a result of providing academic support services to special population students. W A number of research questions were identified to assist in the implementa- tion of this study: 1. Do special population students use the supplemental (grant-sponsored) academic support services available? 2. Do special population students use institutional academic support ser— vices at the same rate as non-special population students? 3. DO grant-sponsored academic support services provided to special popu- lation students supplement or replace institutionally funded academic support ser- vices? 4. When supplemental (grant-sponsored) academic support services are available, do special population students academically achieve at a rate comparable to non-special population students? 5. When supplemental (grant-sponsored) academic support services are available, do special population students persist at a rate comparable to non-special population students? 6. What portion of supplemental (grant-sponsored) academic support ser- vices Costs are borne by the community college? 7. When grant-sponsored academic support services are provided to special population students, what is the estimated impact on a community college's budget? W The research questions were formulated into null hypotheses to determine if there were differences in the academic achievement, persistence, and usage of academic support services by experimental group (special population) students and control group (non-special population) students. The researcher compared three groups to determine if differences existed between: experimental and control group students; experimental and institutional group students; and control and institutional group students. Hypothesis; The mean hours of institutional academic support services used during the Winter 1991 semester by control group students will not statistically significantly exceed (p > .05) the mean hours of use by experimental group students. Hypothesis}. The mean hours of academic support services used during the Winter 1991 semester by experimental group students will not statistically signifi- cantly exceed (p > .05) the mean hours used by control group students. Hypothesis}. The experimental group students'WInter1991 semester mean grade point average will not statistically significantly (p > .05) exceed the Winter 1991 semester grade point average of control group students. HypothesisA. Experimental group Students' proportion ofWInter1991 se- mester courses completed will not statistically significantly exceed (p > .05) the proportion of courses completed during the WInter 1991 semester by students in the control group. W. The Fall 1991 semester enrollment of experimental group students will not statistically significantly exceed (p > .05) the Fall 1991 semester enrollment of control group students. Hypothesisj. The mean grade point average during the Winter 1991 semes- ter for the experimental group will not statistically significantly (p > .05) exceed the institutional group students' mean grade point average. amnesia. The proportion of courses completed during the Vlfinter 1991 semester by the institutional group will statistically significantly exceed (p > .05) the proportion of courses completed by the experimental group. Hypnthesjsfi. The enrollment of experimental group students in Fall 1991 semester courses will not statistically significantly exceed (p > .05) the proportion of institutional group students. Hypothesisj. The institutional group students' Winter 1991 semester mean grade point average will not statistically significantly exceed (p > .05) the Winter 1991 semester mean grade point average of control group students. Hypothesisjfl. The proportion of courses completed by institutional group Students during the Winter 1991 semester will not statistically significantly exceed (p > .05) the proportion of courses completed during the Winter 1991 semester by control group students. Hymjhesisjl. Fall 1991 semester enrollment by institutional group students will not statistically significantly (p > .05) exceed the enrollment of students in the control group. W The numbers of special population students enrolling at community colleges who require supplemental academic support services are expected to increase rather than decrease (Amen‘canflemggrapbjgs February 1988). Concurrently, legislatures, policy makers, and community college administrators must be fiscally responsible as they attempt to effectively maximize and allocate limited resources. The results of this study can provide information to community college administrators to estimate the usage, costs, and associated benefits of providing supplemental academic support services to special population students through grants and institu- tional funding. These data may determine if additional funding is necessary to support these services and if so, may provide a rationale for community colleges to request additional grant funding and/or to reallocate institutional funding. W The study was delimited by being applied to a limited number of experimental and control group students enrolled at one community college. The study was also delimited by examining four grant programs at the institution as compared to the number of grant-sponsored programs available. A major limitation to this study is that it was not longitudinal; it only studied a limited number of students over a one year period. This timeframe did not allow for the identification of transfers, stopouts, or permanent leavers. A second major limitation to this study is that it was not condUcled during the Winter 1991 semester, but approximately one year later. This limitation was further compounded as modifications in legislation, administrative rules, and funding levels also made comparisons across time and grant programs difficult. The definition of academic support services can also differ between grant programs and across time. Finally, while institutional recordkeeping can be used to verify student survey responses, the research was conducted on self-reported data. Since limited conceptual models of student academic achievement and persistence have been specifically designed and tested at community colleges, transferability to other institutions should be carefully considered. However, the generalizability of the findings of this study to other community college settings is possible if the institution operates similar academic support services Sponsored by similar grant funded sources. Due to the national federal funding policies that pro- vide grants to assist community college special population students to be academi- cally successful, generalizations to other institutions is highly possible. W The following terms are defined in the context in which they were used in this study. W. A student's grade point average during the WInter 1991 semester. W. A student's successful completion of at least 60 percent of the courses in which they enrolled during the Winter 1991 semester; and completion of a stated program during the Winter 1991 or Spring/Summer 1991 semesters or subsequent enrollment in courses during the Fall 1991 semester. WW. Counseling and tutoring. ASSEI. A basic skills assessment instrument that IS nationally normed and identifies students' scores for writing, reading, numerical, and algebra skills. Wrap]. A state funded program designed to provide supplemental academic support services to students enrolled in develop- mental courses as designated on the institution's Academic Classification System Report. W. 1990-91 fiscal year; July 1, 1990-June 30, 1991. Cumulatmfiredjtflours. The number of credit hours completed prior to the Winter 1991 semester, Classified in ranges of credit hours 1-14, 15-29, 30-44, 45-59, 10 60-74, 75-89, 90-104, 105-119, 120-134, and 135 and above. EnmflmenLStams. Part-time enrollment is defined as enrolled in 0-11 credit hours and full-time is defined as enrolled in 12 or more credit hours during the Winter 1991 semester. Institufinnalfimup. A group of 500 randomly selected students attending Delta College during the Winter 1991 semester. The 500 students will not be in either the control or experimental group and their student characteristics will not be idenfifled. W. A state funded program designed to assist displaced workers to receive training and regain employment. TO receive services, students must be unemployed or have received a layoff notice. Eligible students must be enrolled in training that will be completed and allow job placement within one year of enrollment. Wren]. A federal program to support activities including tuition-free vocational education to assist single parents/homemakers or pregnant Single women to acquire the job skills and training necessary to provide adequate financial support for themselves and their dependents. Eligible students must be enrolled in a state-approved vocational education program or course, approved basic literacy instruction, or approved career guidance workshops, semi- nars, or courses. W A federally funded program that provides supplemental academic support services to students who are academically deficient and enrolled in state-approved vocational education programs or COUI' SOS. 11 WW. Students who federal and state grant agencies have determined require additional academic support services to be academically successful, and who enrolled at Delta College during the Winter 1991 semester and received supplemental academic support services funded through the At-Risk Stu- dent Success, Single Parent/Homemaker, Special Needs-disadvantaged compo- nent, and Michigan Job Opportunity Bank-Refrain programs during the 1990-91 academic year. ..- - . -.. u. .-. ; ..-.' .... - '.Academic support services provided during the 1990-91 academic year to eligible Delta Col- lege students enrolled in the At-Risk Student Success, Michigan Job Opportunity Bank-Retrain, Single Parent/Homemaker, and/or Special Needs-disadvantaged component programs. W03 The study was conducted at Delta College, University Center, Michigan. Delta College is a community college established in 1961 to provide postsecondary education Opportunities to Saginaw, Bay, and Midland County residents. The Col- lege is controlled by a nine-member Board of Trustees elected by the constituents of the three counties. During the 1990-91 academic year, Delta College enrolled approximately 12,500 students each semester in thirty-six transfer programs, fifty- three associate degree programs, and seventeen certificate programs. In the 1990-91 academic year, some form of financial aid (grants, loans. scholarships, and/or student employment) was received by approximately 42 percent of the students who enrolled on at least a part-time basis. Many special population, 12 students received financial assistance from grants, loans, employer tuition reim- bursement programs, and State and Federal retraining assistance programs. W This study is presented in five chapters. Chapter II contains a review of the published literature regarding student academic achievement, student persistence, related academic and student support services, the financial resources utilized by community colleges to provide these services, and relevant grant legislation and funding. Chapter III is a presentation of the research methods and an explanation of the procedures used to gather and analyze the data. Chapter IV contains the results of the data analysis and Chapter V includes a summary of the study, conclusions, and recommendations. Appendices that contain the questionnaires used in the study, copies of various communications with respondents, and other documents related tO the study follows the text material which is followed by an extensive bibliography. CHAPTER II REVIEW OF RELATED LITERATURE HEW: The open-door philosophy and mission of the community college in the United States evolved in response to several societal factors. The community col- lege movement began through several alternatives including private academies, university extension centers, private two-year colleges, and junior colleges, with junior colleges having the most significant impact on the development of the commu- nity college. While the ”first” community college was Moline Community College in 1947, community colleges have a history dating back to the 18005. There is evi- dence that two-year colleges were in existence before 1850, however, the Morrill Act of 1862 provided the impetus for the open-door philosophy of the community college (Plaucher, 1987). Key university presidents encouraged the diversion of academi- cally inferior students into an upper extension of the high school. If academically underprepared students demanding access to higher education enrolled at a two- year college, then universities could focus on research and advanced professional training. As a model, the University of Chicago (1892) was separated into two divisions; a lower division for freshmen and sophomores and an upper division for juniors and seniors. In 1900, the University of Chicago began awarding Associate degrees to students completing lower-division coursework. Associate degree 13 14 students who indicated academic promise and achievement could continue their education and enroll in upper division courses (Brint and Karabel, 1989). In the early 19003, the junior college's mission was to provide the first two years of college for the increasing numbers of high school graduates who wanted to attend college but were unable to access four-year universities due to their geo- graphically distant location, high cost, and required academic standards. By 1921, there were 207 junior colleges established in the United States (Plaucher, 1987) and their student enrollments grew in response to several societal factors. The factors that impacted on junior college growth included the need for trained workers for employment in industries skilled for new technologies, the need to educate youth beyond high school, the drive for social equality that was enhanced by opening more schools and encouraging everyone to attend, and the means for achieving upward mobility (Cohen and Brawer, 1984). The Great Depression further increased junior college enrollments as indi- viduals attended college as an alternative to unemployment. State legislatures authorized the establishment of new colleges, so that by 1940, ten percent of all United States college students were enrolled in public two-year institutions. During this time, the junior college expanded its mission and established its role as a pro- vider of transfer courses, community services courses, and vocational training (Plaucher, 1987). In 1948, the President's Commission on Higher Education validated the junior college's mission and role in higher education. The Commission's report suggested that the name ”community college” be applied to institutions designed to primarily serve local community needs. These institutions could have various forms of 15 organization and curricula of various lengths. Their dominate feature is their intimate relation to the community they serve. The Commission established five basic pur- poses and functions of these institutions: 1) survey the community and adapt pro- grams to the needs of its full-time students, 2) consider apprenticeship training to provide older students alternate periods of college attendance (classroom study) and work experience, 3) provide students with a total educational effort that integrates general and vocational education, 4) provide students with academic preparation to transfer to four-year institutions, and 5) administer a comprehensive adult education program (Bogue, 1950). Federal legislation played a significant role in the expansion of the community college system and increased student enrollments. Enrollments at community colleges surged during the post-World War II period as the GI. Bill of 1944 spon- sored World War II veterans and brought a new population of adults and returning students to community colleges. Community colleges were the institution of choice of a large number of veterans who wanted to resume their education at a location geographically close to their home and at a relatively low cost. The 1963 Vocational Education Act initiated increased vocational funding to two-year institutions. Under the 1963 Vocational Education Act, a majority of the funding was appropriated to technical institutes rather than community colleges. The Higher Education Acts of 1965 and 1968 provided increased federal funding to community colleges and Significantly expanded their vocational education programs. The Educational Amendment Act of 1972 further increased the federal financial commitment to vocational programs for disadvantaged and handicapped students. The increase of federal funding to community colleges in the early 19705 16 ”channeled” students into vocational education programs. Between 1970 and 1977, full-time community college vocational education student enrollments increased from 33 percent to over 50 percent and student enrollments in transfer programs signifi- cantly declined. Community colleges continued to revise their curricula from a transfer orientation to vocational in response to consumer preferences, changing market conditions, and social preferences in their immediate geographical area (Brint and Karabel, 1989). The greatest increase in community college growth and public acceptance occurred during the 1960s and early 19705. In 1960, there were 315 public two-year colleges that enrolled approximately 392,000 students, accounting for 11 percent of the total higher education enrollment. In 1979, 926 community colleges enrolled over 4 million students, accounting for 35 percent of all postsecondary enrollments, and enrollments have remained steady during the 19805 with approximately 4.8 million students enrolling at 1,240 institutions that account for approximately 50 percent of all first-time college enrollments in the United States. There are two key reasons that have accounted for the tremendous student enrollment at community colleges since 1960: 1) the post-World War II ”baby boom” generation came Of college age, and states had to expand the number of positions in higher education institutions. Community colleges could be built and opened quickly and were as- sumed to be a cost-effective way to expand postsecondary education, and 2) the open-door phIIDSOphy provided the opportunity to postsecondary education to indi- viduals who could not gain admission to more selective institutions. Federal policy encouraged equal educational opportunity that was supported by need-based finan- cial aid programs. Enrolling increased numbers of minority and disadvantaged 17 students in higher education institutions was a particular goal (Breneman and Nelson, 1981). By the 19705, the community college appeared to have a universal definition of its mission, at least one that was accepted by community college professionals. The community college is an open-door institution with cOmprehensive programs that include transfer education, developmental education, vocational education, continuing education, general education, and student support services. Vocational and transfer programs are the instructional priorities of community colleges. The purpose of vocationalism is to teach specific skills to students to Obtain employment and strong linkages exist between the local labor market, employers, and the com- munity college's vocational training programs (Richardson and Leslie, 1980). Due to the community college's Open-door policy, developmental education and student support services became necessary to assist special population stu- dents to achieve their educational goals. In 1960, 16 percent of community college students who were given a standardized test were classified as remedial and re- quired academic support services. Student support services such as academic advising, guidance, and counseling services were initiated after World War II based on recommendations by the Veteran's Administration (Bogue, 1950). As community colleges became the institutional opportunity for the less academically prepared or at—risk student, increased resources were allocated to developmental education and student support activities (Richardson and Leslie, 1980). 18 WW Demographers predict by the year 2000 that there will be a significant short- age of qualified people to fill the available jobs; and many of the individuals who should fill these jobs will be unmotivated, undereducated, underhoused; a perma- nent underclass. The future economic strength of our country is in danger if the at- risk population is not empowered to be a contributing force (Parnell, 1990). By 2000, nearly 75 percent of all jobs will require postsecondary education or technical training. However, literacy and educational inequities currently exists among various ethnic and socioeconomic groups: High School College Bachelor's Graduation Enrollment Deuces Anglo 83% 38% 23% Black 72% 29% 12% Hispanic 550/0 230/0 70/0 Native American 55% 17% 6% (Parnell, 1990, p. 110) Community colleges are enrolling increasing numbers of educationally underprepared students due to their open-door admissions policies. Colleges who enroll academically at-risk students need to provide student and academic support services to increase students' academic success, retention, and to assist students to achieve their educational goals. Programs and services to assist at-risk students should include: an assessment upon admission to assure basic skills in reading, math and written English expression, with developmental courses provided for students with identified deficiencies; a process to continually monitor Student 19 progress; academic advising to assure students have accurate, appropriate informa- tion on what is required to obtain a certificate or degree; counseling or academic advising to assist students to select the appropriate courses; self-paced enrichment and tutorial programs; and a core curriculum that is required of all degree students to assure that proficiencies are met (Parnell, 1990). The INQLKIQLQEZQQQ study indicates that by the turn of the century, the mean years of education required for employment will rise to 13.5, and only 27 percent of all jobs will be in low-skilled occupations. Similarly, the W W demonstrates that "occupations requiring a college education or postsecondary technical training head the list of the fastest-growing occupations over the next decade and are expected to account for 6.2 million of the 16 million jobs projected to be added from 1984 to 1995" (Gold, 1990, p. 6). The economy will need to fill more of its high-skilled jobs through adult retraining as the birth rate declines and the workforce grows older, more female, and more disadvan- taged (Gold, 1990). At-risk adults are the individuals with the most to gain by attending college. "They are persons who are not employed at or near their productive capacity be- cause of their personal circumstances or changes in the workplace" (Gold, 1990, p. 6). These adults include welfare recipients, displaced homemakers, dislocated workers, and people whose skills are rendered obsolete by advancing technology (Gold, 1990). Members of these special populations are enrolling in college in greater numbers. A recent MW indicated that 52 percent of the nation's undergraduate students are 22 years of age or older, and approximately 20 percent are 30 years of age or older. At community colleges, 20 these percentages are even higher, as the average age of students attending a community college is 29 years of age. Between 1972 and 1985, the number of 30 to 34 year old students enrolled in higher education institutions increased by 122 percent and the number of students 35 years of age and older increased by 112 percent. As a result, institutions are offering more evening courses, expanded administrative services, and expanded hours for academic support services. How- ever, nontraditional student enrollments, special programs, and developmental projects can only maintain enrollment levels for so long. The additional costs and stress on faculties and support personnel are often at a great cost to the institution (Melia and Goodman, 1988). r I L l l I l in mm I II The federal government Operates a variety of programs aimed at providing aid to college students, tax incentives for employer-paid training, and funds for conducting manpower training. Recently, the key legislation impacting community colleges has been the Carl D. Perkins Vocational Education Act. The purpose of this legislation is to make the United States more competitive in the world economy by fully developing the population's academic and occupational skills. This will be accomplished by concentrating resources to improve educational programs that lead to academic and occupational skill competencies that are necessary to work in a technologically advanced society. The Carl D. Perkins Vocational and Applied Technology Education Act Amendments of 1990 amended the previous legislation. The primary purpose of the amendment is to redirect federal assistance to improve vocational education and services for special population students. States must 21 assure that special population students will be given equal access to vocational education and receive supplementary and other services necessary to succeed in vocational education. States must give priority to programs that serve the highest concentrations of special population students (EederaLBegjster, October 11, 1991). The legislation focuses federal resources on the following community college programs. Careerfiujdanmndggunselmg programs provide special population stu- dents with career awareness, planning, and decision-making; placement skills; an understanding of local, state, and national occupational, educational, and labor market needs, trends, and opportunities; and assists individuals in making and implementing informed educational and occupational choices (StatacLMIchican WWW) SpegjaLNeeds program funding provides academically disadvantaged stu- dents with supplemental academic support services and counseling. In Michigan during the 1990-91 academic year, community colleges received $2,331,949 to Operate Special Needs programs. To receive supplementary services, students must be unable to succeed in regular vocational education courses without addi- tional assistance. The following indicators are evidence of a special population student's inability to succeed: 1) a grade point average based on high school or prior college work of less than a 2.0, or demonstrated inability to succeed based upon standardized testing procedures; and 2) less than a 2.0 grade in any federally reimbursable vocational course, or enrollment in a developmental course needed to prepare the student for course work in a vocational curriculum. Approved activities to assist special population students include: outreach activities to inform the public 22 of the availability of programs and services; assessment to determine the interests, , abilities, and needs of each special population student; peer tutorial or professional instructional assistance; curriculum modification to revise curriculum materials or instructional methods; the purchase of special instructiOnal equipment; and guid- ance, counseling, and career development activities (W015 W) W programs serve special population students who are displaced homemakers, homemakers, or single parents. Services available include tuition and fee reimbursement for vocational training, reimbursement for the appropriate tools, supplies, and materials necessary to participate in vocational training, counseling, tutoring, developmental Skill building, and job placement. In Michigan during the 1990-91 academic year, community colleges received $2,120,553 to operate the Single Parent/Homemaker Program. Students meeting the following eligibility criteria can receive program services: 1) aSingle Parent is an individual who is unmarried or legally separated from a spouse, and has a minor child or children for whom the parent has either custody or joint custody; 2) a Home- maker is an individual who Is an adult and has worked primarily without remunera- tion to care for the home and family, and for that reason has diminished marketable skills; 3) a Displaced Homemaker is a homemaker who because of divorce, separa- tion, or the death or disability of a spouse must prepare for paid employment. This individual must also be an adult who has worked primarily without remuneration to care for the home and family and for that reason has diminished marketable skills; has been dependent on public assistance or on the income of a relative but is no longer supported by such income; or is a parent whose youngest dependent child 23 will become ineligible to receive assistance under the Aid to Families with Depen- dent Children program within two years of the parent's application for assistance; or is unemployed or underemployed and is experiencing difficulty in obtaining any employment or suitable employment as appropriate; and 4) a Single Pregnant Woman is an individual who is unmarried or legally separated from a spouse and is Pregnant (StatacLMicmcanDeanlsfiuidmLMmaticnalfiducaflanfiall I l I In addition to federally sponsored activities, state-funded programs to assist special population students have been implemented. In Michigan, Governor Blanchard initiated the At-Risk Student Success Program in September 1989. A total of $2.15 million was appropriated in fiscal year 1990 (Community College Appropriation Act No. 175, PA. of 1989). Each community college received a mini- mum grant of $40,000 with the remaining funds allocated on a formula based on the number of unduplicated students served during the 1988-89 academic year through the Michigan Job Opportunity Bank-Refrain Program, the Tuition Incentive Program, and the Michigan Occupational Skills Training Program. Community colleges ac- cepting At-Risk Student Success funding were required to pretest all incoming Special population students' academic ability using a nationally standardized test and provide a minimum Of one counseling contact during the first half of each term. At- Risk Student Success Program funding could be used to provide assessment test- ing, counseling, mentoring, or related activities. During the 1991 fiscal year, student eligibility was revised to serve Special population students enrolled in developmental courses as defined by institutional Academic Classification System Reports. 24 WW Retraining workers for new jobs has become a major challenge for commu- nity colleges, especially in communities where residents face either immediate layoffs or long-term unemployment. Displaced workers must be able to operate new technologies, implement new manufacturing processes, and practice quality control strategies required by business and industry. For displaced workers to successfully reenter the job market, retraining to obtain these skills is often required (Wojcicki and Kauhnan,1990) In the past, local economies experienced periodic highs and lows, with corre- sponding changes in the unemployment rate. Workers grew accustomed to ”riding out” these recessions because they knew that in time, the economic situation would improve and their jobs would return. Workers felt secure that their Skills would still be adequate when the local economy returned to normal. Unfortunately, in today's industrial economy, the reality is that some worker's skills are no longer adequate to meet the needs of today's highly technical, automated manufacturing systems (Wojcicki and Kaufman, 1990). Prior to entering a training program to develop new occupational skills, dis- placed workers should participate in a series of transitional services. These services will assist the displaced worker to realistically examine the current and projected labor market, identify the skills required by employers, and select the appropriate vocational training program to acquire these skills. Transitional services should include an orientation, evaluation process, career path exploration, career choice qualifications, and retraining to complete the career change (Wojcicki and Kaufman, 1990). 25 in Michigan, Governor Blanchard initiated the Michigan Job Opportunity Bank-Retrain Program on January 1, 1986 to assist unemployed Michigan residents to obtain the retraining necessary to reenter the labor market. During fiscal year 1986, the State provided $2 million to Michigan community colleges to serve a minimum of 1,000 dislocated workers. An additional $2 million was provided annu- ally through fiscal year 1991 to support the program's objectives. The goals of the Michigan Job Opportunity Bank-Retrain Program were to: 1) successfully utilize Michigan's community colleges to provide assessment, training and placement services to dislocated workers; 2) train at least 1,000 dislocated workers in occupa- tions with employment opportunities available in the local labor market; and 3) place at least 70 percent of participating dislocated workers into training-related jobs at a rate of at least $5.00 per hour. To be eligible to receive services, individuals must be: 1) a citizen or national of the United States and a Michigan resident; and 2) a dislocated worker who is: unemployed or underemployed due to a layoff or plant closing; not currently enrolled in other federal or state training programs for similar services; employed at least three years in the same industry or occupation before being displaced; and capable of successfully completing the training within one year or less as determined by the community college staff. Community colleges participating in the Michigan Job Opportunity Bank- Retrain program were required to provide eligibility determination services, assess- ment, counseling, vocational skills training, and job development and placement. Each of these legislated programs appropriate funding to assist special population students to successfully complete their educational objectives and obtain 26 the occupational Skills necessary to enter or reenter the labor market in the shortest timeframe possible. While community colleges are required to compile data and submit final reports to the funding agency indicating the numbers of students served, the activities implemented, and financial expenditures, limited information exists on the impact these grant-funded programs actually have on the special population students they were designed to serve. Additional research is required to determine if grant funds awarded to community colleges positively impact special population students' ability to achieve their educational objectives. n n n i Attrition and persistence are both reflections of the functioning of the higher education system. AS a national phenomenon, attrition has been a stable feature Of higher education and the rates of attrition have remained constant over the past 100 years. With the exception Of the period during and immediately following World War II, the rates of attrition have remained stable at about 45 percent-despite the marked growth and alteration in the character of the higher education system. It is unlikely that this rate will be significantly altered without massive changes in both the struc- ture and functioning of higher education in the United States. Vincent Tinto conducted several research studies on student persistence in four-year institutions. The model developed by Tinto (1975) is the most widely recognized and tested model of student retention. Tinto's model is a longitudinal, institution-specific retention model in which emphasis is placed on integration. Integration into the academic system of the college leads to new levels of student commitment. The interaction between the student's commitment to the institution 27 and his/her commitment to the goal of college completion determines whether or not the student persists. In the academic domain, goal commitment influences student's grade performance and intellectual development, which increases their likelihood of persistence. There are two forms of academic integration. The first form is mea- sured in terms of grade performance and intellectual development and in the sec- ond, grades are the single most important factor in predicting persistence in college (Ammons, 1971; Astin, 1972; Blanchfield, 1971; Coker, 1968; Greive, 1970; Jaffe and Adams, 1970; Kamens, 1971; and Mock and Yonge, 1969). In conducting a retention study, researchers need to identify the longitudinal process Of interactions by building into the model sets of individual characteristics and dispositions relevant to educational persistence. The model should identify student's background characteristics, expectations and motivational attributes (mea- sured by career and educational expectations and levels of motivation for academic achievement), educational goal commitment, and institutional commitment. Other things being equal, the higher the student's degree of integration into the college's academic system, the greater their commitment to the specific institution and to the goal of college completion. If unchanging external conditions are assumed (chang- ing labor/job market, repeal of the draft law, etc.), attrition is a result of the student's experiences in the college's academic system. A number of attrition studies at four-year residential institutions have validated portions of Tinto's model. The influence of academic integration on student retention has been supported in a number of studies including Nelson, Scott, and Bryan, 1984; and Pascarella and Terenzini, 1980. Pascarella and Terenzini's study as- sessed the dimensions of academic integration. The study sought to determine 28 whether a multidimensional measure of academic integration based on Tinto's model would significantly discriminate between freshman year persisters and voluntary withdrawals, with the student's pro-college characteristics held constant. The study's results support the predictive validity of the Tinto model, and have significant implica- tions for educational decision-makers concerned with maximizing institutional re- sources associated with attrition. A significant portion of student attrition can be prevented through timely and carefully planned institutional interventions. These interventions will be effective if students with a high probability of withdrawing can be accurately identified (Pascarella and Terenzini, 1980). Munro used path analysis to test Tinto's model using a sample drawn from the National Longitudinal Study of the High School Class of 1972. In this study, it was found that while pre-college characteristics predicted college integration, they did not directly affect persistence. Factors related to the student's integration into the college's academic environment were more important. This lends support to the findings of Terenzini and Pascarella (1978) in which academic integration variables accounted for nearly twice as much variation in persistence as did other variables. As the costs of higher education continue to rise, colleges and universities are increasingly under pressure to document the consequences or outcomes Of the educational programs and services they offer. Numerous studies have focused on the personal traits of students and their influence on educational outcomes. Fewer studies have attempted to assess the relationship between student's developmental progress and the various collegiate programs presumed to influence that growth and over which institutions have some control. A good deal is known about how students grow in the collegiate environment, but far less is known about the institutionally 29 induced forms of student development (T erenzini, Pascarella, and Lorang, 1982). A large number of student retention studies were conducted without control- ling for academic aptitude and other student background characteristics. The ab- sence of control creates interpretive problems for studies seeking to isolate institu- tional effects. In studies of institutional influences, failure to control for students' pre- college characteristics means that one cannot reject the hypothesis that students performed or grew as they did because of what they were like when they first ma- triculated. For college administrators, the question is not whether students Change or grow in various ways during their college enrollment but whether (and to what extent) the growth or change is attributable to the collegiate experience. The issue is not what the influences are, but rather what institutionally controllable influences have a positive effect on desirable educational outcomes (T erenzini, Pascarella, and Lorang, 1982). Terenzini, Pascarella, and Lorang (1982) conducted a study with two primary purpOses: 1) to determine whether students' perceptions of their growth in certain areas could be reliably related to their experiences at a higher education institution, and 2) to identify those features of the collegiate experience over which the institu- tion has some control that influence students' perceptions of growth in various areas. Because Tinto's model suggested that students' academic growth is related to a variety of background characteristics, and due to the interest in assessing the influ- ence of the collegiate experience on educational outcomes, this study controlled for the following pre-college characteristics that the research literature suggests are important correlates of academic development (Lenning, 1974; Feldman and Newcomb, 1969; Astin, 1977): gender, race, intended academic major, combined . 30 formal education of parents, university's rank as a college choice, parent's combined annual income, high school achievement, academic aptitude, and highest academic degree planned. The results of Terenzini, Pascarella, and Lorang's study (1982) determined that with the exception of gender, race, and combined SAT scores, the pre-college traits measured in this study appear to be unrelated to student's perceptions of their growth during the freshman year. The research findings suggest that even when differences in a battery of student pre-enrollment traits are statistically controlled, measures of student involvement with faculty, peers, campus social environment, and classroom experiences account for significant increases (from 3 to 25 percent) in the explained variance of a range of educational outcomes. These outcomes include freshman year academic achievement and perceived growth in four areas: perceived personal development, academic and intellectual processes and content, and preparation for the future. Pantages and Creedon (1987) have stressed the importance of identifying at- risk students so that intervention with counseling or other institutional programs can be undertaken before withdrawal decisions are made. The performance of five institutional integration scales in this study suggests that they may be useful in identifying freshman students with a high probability of withdrawing. The five scales are: 1) interactions with faculty, 2) students' perceptions of faculty concern for student development and teaching, 3) peer-group interactions, 4) academic and intellectual development, and 5) institutional and goal commitment. Different forms of institutional actions for student retention must be carefully timed to meet the changing situations and needs of students as they attempt to 31 progress along the path to college completion. Programs and evaluation strategies must be employed to determine which forms of institutional action are most effective in treating departure at different points in the student's career. Whatever forms of action institutions take on behalf of student retention, these actions should be con- centrated on the very early stages of the student's college career rather than on later stages alter serious problems have surfaced. Institutions must not ignore student needs beyond the first year, however, it is evident that the first year, indeed the first semester is critical to the student's eventual persistence until degree completion (Tinto, 1988). i m i I During the last two decades, the most common point of entry into higher education has been the community college. More than half of new college entrants begin their higher education in a two-year college (Tinto, 1987). While Tinto's re- searCh focused on four-year residential institutions, all college students have to make some form of intellectual transition to the academic demands of college, eventhough these demands will vary from institution to institution. Tinto determined that variances in the attrition rates in different types of institutions are due to the entering students. The attrition rates are higher at public institutions than private institutions and the attrition rates are higher at two-year colleges than at four-year institutions, even after students' characteristics have been taken into account. The higher rates of attrition at two-year colleges are primarily attributable to the lower level of motivation and academic ability of the entering students. However, student retention rates at two-year colleges are still lower than would be expected 32 from the characteristics of their students alone. Some authors have concluded that it may be the function of two-year colleges to screen-out students from transferring to a four-year institution (Astin, 1972; Bayer, 1973; and Berls, 1969). Other authors have suggested that two-year colleges may screen out lower economic status stu- dents from transferring to a four-year institution and reinforce inequality of opportu- nity within the higher education system (Karabel, 1972; Spady, 1970; and Tinto, 1971). Although a number of theoretical models of college student degree persis- tence have been developed since Vincent Tinto first proposed his theoretical model in 1975, a majority of the models are based on research conducted at four-year colleges and universities. The few studies that have focused on two-year colleges have generally tested the applicability of Tinto's model at two-year institutions. How- ever, differences in the nature of two-year and four-year colleges and their students as pointed out by Tinto (1987) and others, may make Tinto's model unsuitable for two-year institutions. In addition, Tinto's model fails to place sufficient emphasis on subgroup characteristics such as those associated with age, gender, or racial/ethnic categories (Tinto, 1987). Because nontraditional students comprise a large propor- tion of the two-year college population, this is a serious flaw in the Tinto model. Halpin's study examined the application of Tinto's model as operationalized by Pascarella and Terenzini (1980) to analyze student persistence at a community college during the first semester of enrollment. The research results indicated that Tinto's model did have utility for researchers and administrators concerned about retention. Varying levels of integration were significant predictors of persis- tence, withdrawal, and academic dismissal, with the effects of background and 33 environmental factors controlled. The greater influence of academic integration compared to other variables was particularly noteworthy. The Tinto model does predict persistence or exit outcomes. Perhaps the explanation for the predictive power of the model and the importance of academic integration for this population has to do with the fact that the academic systems of commuter and residential colleges are more similar than their respective social systems. Pascarella and Chapman (1983) compared withdrawal from three types Of institutions: four year residential, four-year commuter, and two-year commuter institutions. They determined that Tinto's model had predictive validity for all types of institutions. Other studies have reported the same general findings including Pascarella, Duby, Miller, and Pasker, 1981; Pascarella, Duby, and lverson, 1983; and Pascarella and Wolfe, 1985. In two-year and four-year commuter institutions, academic integration has greater indirect effects on attrition than do other variables (Pascarella and Chapman 1983a; Pascarella and Wolfe, 1985; and Tinto, 1987). In addition, the external environment has been found to be a factor influencing the retention decisions of students at commuter institutions (Chacon, Cohen, and Strover, 1983; Tinto, 1987). A study by Webb (1988) reviewed the Tinto model and its applicability to understanding freshman persistence at two-year colleges. Webb concluded that none of the existing models were adequate for two-year populations, if the intent of a retention program is to identify potential dropouts as early as possible. Webb identi- fied variables that can be used to predict retention using information that can be obtained prior to, or at the time of matriculation. In Webb's model, the following factors had primary effects on student persistence: 1) high school academic 34 achievement (ASSET test scores, high school graduation status), 2) external envi- ronment (need for financial aid, hours planned to work, day/evening status, need for employment), 3) goal commitment (educational goal, certainty of major, enrollment status), 4) expected student/college fit (reason for attending, college choice rank), 5) academic intent (vocational education status, two-year degree plans), and 6) academic integration (first semester grade point average, number of courses passed first semester). In addition, there were two secondary effects in the model: 1) background characteristics (race, gender, veteran status) and 2) academic self- confidence (need for academic assistance). Webb's model differs from Tinto's model due to the inclusion of external environment, academic self-confidence, and expected student/college fit. Bean and Metzner (1986) developed a separate model of persistence for nontraditional students. This model presents variables other than academic integra- tion as having indirect effects on psychological outcomes and intent to leave, thus having a limited effect on attrition. Bean and Metzner‘s model appears more valid for Studying attrition at two-year institutions than Tinto's model. Pascarella. Smart, and Ethington (1986) focused specifically on the long-term persistence of two-year college students. Eight hundred and twenty-five students who initially enrolled in 85 two-year colleges were tracked over a nine year period. The variable with the most consistent pattern of positive effects on persistence was academic integration. A study by Voorhees (1987) followed re-enrollment patterns of 369 students in a community college over two semesters. In this study, academic integration defined as grade point average, number of informal conversations with faculty, and 35 number of hours spent studying was determined not to have an independent effect on persistence. Neumann conducted an ethnographic study of attrition in an urban, commu- nity college in the Northeastern United States. Neumann selected a group Of stu- dents who were high-risk or by institutional standards, were deemed unlikely to complete their degree program. The study focused on whether the experiences of the students who did persist differed from similar students who withdrew. Neumann discovered that successful students consistently made the successful transition to college, often with the help of faculty or staff. The implications for researchers and administrators concerned with retention in community colleges is clear. While little can be done to influence "background characteristics” or ”environmental” characteristics of community college students, the creation of institutional mechanisms to maximize student/staff contact is likely to result in greater levels of integration and therefore persistence. Developmental academic advising systems, mentoring, and small group learning projects, and a generally accessible, involved faculty and staff may be a significant portion of the prescription for retention in open—door community colleges. Few community colleges have had the financial resources to develop a formal student tracking system and few have created a system to identify high risk students. As a result, a serious information gap exists given the increasing concern in higher education with student retention and attrition. The differences in student retention and attrition between two and four-year institutions lie in the characteristics of the students attending and the nature of the community college as an institution. Most research on student retention is based on four-year colleges and either implicitly or- 36 explicitly defines retention as on-time graduation from the institution of first matricula- tion. Therefore, much of this research has limited application in the community college. A college Should be concerned with that portion of student attrition that occurs because of institutional weaknesses or failures. Therefore, the goal becomes one of helping students define and achieve their educational and vocational objec- tives and prevent the immense waste of individual and societal resources when students fail to achieve their goals (Walleri, 1981 ). The major thrust of retention research is to be able to identify which types Of students are most likely to withdraw. However, with only 10 percent of the variance in withdrawals explained, there is simply no basis for meaningful predictions. The factors associated with student retention and attrition can be grouped into four classes: 1) student Characteristics, 2) institutional characteristics, 3) the interaction effects between student and institution, and 4) societal or external factors (Walleri, 1981). 1 Student characteristics include background traits and process attributes; those factors that influence student behavior while in college such as motivational, psychological, financial and employment patterns. Institutional characteristics identi- fied in the literature as affecting Student persistence have less application in the community college than other segments of higher education. Students attend a community college primarily because of low cost, convenient location, or for particu- lar programs. Other institutional characteristics are of secondary importance (Walleri, 1981). The concept of fit between student and college has limited application to the community college. The community college often presents the only viable option for 37 students seeking a higher education. For students not aspiring to a four-year de- gree, community colleges provide the only option for vocational training within a college environment. In the area of transfer and pre-professional education, commu- nity colleges provide smaller classes and more personal attention than is feasible for large university systems to offer. For students who need it, community colleges provide a period of transition. Because of the diversity of needs served and because of the practical considerations that draw students to the community college, it is difficult to adapt the concept of fit to the community college environment. The chal- lenge facing a community college is to develop an institution flexible enough to respond to the diverse needs of a very heterogeneous audience without allowing the academic experience to become diluted. Retention Strategies in the community college will by necessity be multi-faceted and tailored to the specific needs of identifi- able groups within the general student population (Walleri, 1981). At Lethbridge Community College, a Student retention model was established to increase the retention of high-risk students. Community college administrators and faculty determined that institutional retention activities Should focus upon the positive aspects of upgrading the quality of education and improving the experiences provided for individual students both inside and outside of the classroom. Institu- tions and their student services staffs should deveIOp a greater awareness of the needs and characteristics of dropout-prone students. Once they identify these students, student services personnel should institute intrusive strategies to offer immediate support to any student experiencing difficulty. In this study of student retention, Heath found that academic variables such as high school rank, first se- mester college grades, study habits, and financial aid were predictive in identifying 38 high risk students. A student alert system which monitored high risk students was used to quickly begin proactive interventions when these students began to experi- ence difficulties and as a result, student retention was increased (Heath, 1991). Limited retention studies have been conducted at community colleges. The research indicates that academic integration is a key factor to student persistence, particularly for at-risk students. However, many of these studies have transposed Tinto's four-year residential model and applied it to a community college. This research method fails to consider the differences in student characteristics and academic support services inherent in the community college environment. Addi- tional research to examine these factors is necessary. AW W09 For many years, student affairs officers have touted the importance of aca- demic advising for student development and retention. Enrollment management concepts and reports such as lnvclxememjnmaming (Study Group on the Condi- tions of Excellence in American Higher Education, 1984) supported services such as freshman advising because these activities provided benefits for both students and the institution by improving retention. Academic advising is a factor in student persistence and retention. Baldridge, Kemerer, and Green (1982) contend that advising, orientation, and counseling are important retention activities in colleges. Because the major causes of freshman attrition are uncertainty about academic choices and unpreparedness for academic success (Noel and Levitz, 1982; Titley and Titley, 1980; and Cross, 1971), effective 39 first-year advising is especially important. Forrest (1982) concluded that a well- developed orientation and advising system make a significant and positive contribu- tion to student satisfaction and therefore to retention (Young, Backer, and Rogers, 1989). In a study of freshman attrition, Young, Backer, and Rogers (1989) found that grade point averages of students participating in academic advising and counseling services were significantly higher than students who did not receive those services, indicating that the services had an impact on student's grade performance. Attrition was indicated at the end of the fall semester of the first year and upon students' failure to re-enroll in the fall of the second year. The overall attrition rate of 29 per- cent was the lowest at the university in at least 8 years. From 1979 to 1985, first- year attrition ranged from 31 percent to 36 percent. In this study, students participat- ing in academic advising and counseling services had markedly higher grades and lower attrition than did nonparticipants (Young, Backer, and Rogers, 1989). 7 Walleri (1981) found that the availability and quality of student services in- cluding orientation, counseling, academic advising, developmental education, and tutorial services had a positive impact on community college students' persistence. Improvements in these areas have often resulted in increased retention rates. The research suggests that students who need this assistance receive it as early as possible in their college experience. In particular, the college should make every effort to ensure that students identified as ”high-risk” receive adequate counseling and academic advising (Walleri, 1981). Reyes (1981) randomly assigned 124 freshmen to either control or experi- mental groups in her study “The Effects of Peer Counseling on the Academic 40 Performance of College Freshman in Predominately Minority Institutions”. The effectiveness of peer counseling from sophomores trained to peer-counsel fresh- men; and academic groups, study groups, and outreach programs from which the freshmen seek information was evaluated. The freshmen were evaluated in five areas: grade point average; career maturity (Crites Maturity and Inventory Attitude Scale); study orientation (Brown and Holtzman Survey of Study Habits and Atti- tudes); retention indexes; and reactions of community college members, faculty, administration, peer counselors, and students. The findings between the counseled and non-counseled groups demon- strated that there was no significant difference in their grade point average; there was no significant difference in career maturity; there were considerable Changes on the counseled student's teacher approval and educational acceptance; 15 percent more students in the non-counseled groups withdrew; and generally positive reac- tions about the peer counseling program were indicated. The results of Reyes' study indicated peer counseling made a difference, not directly in the form of grade point average, but indirectly in terms of increased retention. Improvement in academic advising ranks among the most frequently recom- mended and implemented interventions for increasing student retention. Several major reasons appear to underlie the appeal of academic advising as a means of promoting student retention. These reasons include: 1) academic advising offers the potential of linking students' goals with institutional resources on a personalized basis; 2) high-quality advising can help students clarify their educational goals and relate these goals to the curriculum and to future careers; 3) encourage academic success by assisting students with a selection of coursework that is compatible with i8 VI {1) n) 41 their interests, abilities, outside commitments, and career aspirations; 4) facilitate referral to other services and programs at the institution; and 5) establish a personal bond between a student and college personnel. For students, the linkage of their goals with institutional resources may subse- quently create a better appreciation of the benefits of a college education, greater involvement in the institution, increased learning, a more satisfying college experi- ence, and stronger motivation for continued enrollment. The outcomes of high- quality advising for students suggests that advising may have indirect effects on retention through other variables related to retention. For example, academic advis- ing may influence students' college grade point average or their perception of the value of their college education for future employment, factors that in turn affect retention. Many intervention programs have been implemented on the assumption that an improvement in advising quality will decrease student attrition. However, empirical investigations have neither thoroughly tested this assumption in the context of the student attrition process nor estimated the magnitude of the total effects on attrition from differences in the quality of advising (Metzner, 1989). In Metzner's study, a model of the student attrition process was used to examine the influence of the perceived quality of academic advising on freshman attrition. Neither good advising or poor advising had a significant, unique, direct effect on student attrition. However, the indirect effects of the advising variables on attrition occurred through the intervening variables of satisfaction, utility, grade point average, and intent to leave. Good advising was found to have a negative association with attrition based upon the following factors: student's better academic performance (grade point average), their belief that a college education had greater 42 value for future employment opportunities, more satisfaction with courses and the role of being a student, and less intent to leave the university. Advising has direct links to student concerns and implications for perfor- mance in the academic area and other areas of development. Forrest (1982) con- cluded that ”probably the single most important move an Institution can make to increase student persistence to graduation is to ensure that students receive the guidance they need at the beginning of the journey through college to graduation" (Hartley, 1987, p. 88). Callis and DePauw (1985) reviewed a number of studies that found that counseling improves the academic performance of students, while Bishop and Breneman (1986) reported that the vast majority of students who were considered to be retention risks chose to continue their enrollment after receiving counseling. Eighty percent of the students participating in counseling center services were still enrolled as full-time students for the academic year following their initial contact with the counseling center. Twenty percent did not re-enroll at the institution during the next academic year. The results of this survey provided additional evidence that students who are identified as retention risks tend to persist in their academic ca- reers after receiving counseling (Bishop and Walker, 1990). Western New Mexico University developed a retention program that signifi- cantly reduced attrition and increased enrollment. For the eleven years prior to implementing the retention program, student attrition had increased, reaching a peak of 65 percent for freshman classes and 35 percent for sophomore classes during the 1980-81 academic year. To reverse this trend, Western New Mexico University evaluated the reasons for the high attrition. The institutional evaluation revealed that (O *1 43 the students who dropped out were primarily freshmen and sophomores who had not received enough individual attention; needed developmental programs; were confused about degree and program requirements and ways to meet requirements; needed more vertical degree options; and were, most'importantly, disinclined to take advantage of many of these programs and services should they be offered. As a result of implementing advisement and counseling retention strategies, university enrollment increased during the 1982-83 academic year by 18 percent and for the 1983-84 year by 13 percent. Attrition for freshmen was reduced from 66 percent to 48 percent during the 1981-82 academic year and from 48 percent to 25 percent during the 1982-83 year. Freshmen attempted a greater number of credit hours, more attempted hours were completed and a higher mean grade point average was earned (Heath, 1991). In 1977, Jefferson Community College faced an 11 percent crap in enroll- ment. The institution conducted an extensive study of in-class attrition and found that approximately 40 percent of the Class enrollments resulted in either a failing grade or in a withdrawal from the course. Institutional personnel determined that the overall class attrition of 40 percent was too high (eventhough it matched the national average) and used this as a starting point to initiate an institutional retention pro- gram. The program Included: an on-line computer student records and tracking system; an advising system capable of following the progress of entering freshmen during their first semester and continued monitoring of special population students; and a coordinated assessment of students from their point of entry to their point of exit. AS a result of this retention program, the grade point average of experimental (program) students during the Fall 1983 semester was 2.45 and a Similar control 44 group of students received a grade point average of 1.62 (Miller, 1985). Landward and Hepworth (1984) conducted a study of at-risk college students and placed the students into three groups: 1) an experimental group consisting of high-risk freshmen who participated in an academic support program, 2) a control group I, consisting of high-risk students who were not included in the experimental group, and 3) a control group II, consisting of high-risk students were did not partici- pate in an academic support program. The students were equivalent on ACT test scores and predicted grade point averages and there were no significant differences in age or gender. The criterion measures to compare the three groups were mean grade point averages achieved for each of the three quarters and for the full aca- demic year. A second criterion measure was the proportion of students who re- mained in school after completing the first quarter and during the full academic year. The program consisted of academic advising and counseling both on an individual and group basis. Landward and Hepworth found that there was a significantly higher'academic performance level for students in the experimental group. The results: Gmun hi Mean SD Elaine Experimental 30 2.14 1 .13 Control I 59 1.14 1.02 6.61 Control ll 1 15 1.34 1.05 Total 204 1 .47 1 .16 (Landward and Hepworth, 1984, p. 123) The experimental group also outperformed the control groups on rate of continuance in school. Eighty-Six percent of the experimental group students continued their 45 enrollment after the first quarter as contrasted to only 54 percent and 52 percent. The results: Beginning Remaining Iotal Experimental 30 26 86 Control I 61 35 54 Control II 121 62 52 (Landward and Hepworth, 1984, p. 124) Data for the three groups for the second quarter reflected a reverse trend. The experimental group mean grade point average decreased whereas the control groups slightly increased their mean grade point averages: 2nd_QuaneLGEA 3Ld_QuaneL.GEA GLQun M Mean SD EMalue N Mean SD F_Value Experimental 26 1.29 1.03 18 1.78 1.25 Control I 30 1.69 .97 1.07 29 1.63 1.10 .39 Control ll 59 1.54 1.09 58 1.53 .95 Total 1 15 1.53 1.05 105 1.60 1.04 (Landward and Hepworth, 1984, p. 124) The marked drop in academic performance could have resulted from the withdrawal of the academic support and assistance provided during the first quarter. The increase in grade point averages were lowest for the first quarter. In large measure, the effects of the academic support services were transitory, and served Vic Pic 9, . N 46 only to delay the attrition characteristic of the high-risk students in the experimental group. The composite grade point average of the experimental group (1.78) was slightly higher than those of Control I (1.57) and Control II (1.47). By the end of the freshman year, the experimental group had a Slightly higher retention rate (60 per- cent) than Control I (48 percent) and Control ll (47 percent). Therefore, academic support services provided during the first quarter of enrollment did not improve high- risk students' grade point average or retention overall. Students who received services during the first quarter outperformed their peers in grade point average and retention during the semester academic support services were provided (Landward and Hepworth, 1984). A study was conducted by Hudesman, Avramides, Loveday, Wendell, and Griemsmann (1986) to determine the impact of counseling on the academic perfor- mance of college students enrolled in special programs. There have been several attempts to develop counseling strategies for improving the academic performance of at-risk students. One approach is to provide high-risk students with a structured model of counseling services. Treadwell (1977) found that using a structured group approach in counseling and advising situations was associated with higher grade point averages and better retention of students after one year. These findings are consistent with conclusions drawn by Moore (1976) and Godbold (1978), who con- tend that disadvantaged or high-risk students place greater value on tangible ser- vices, specific referrals, and direct answers than on more nondirective counseling procedures (Hudesman, Avramides, Loveday, Wendell, and Griemsmann, 1986). Educational institutions are becoming more concerned about how to retain students. This can be inferred from the rise in student support services such as 47 study skills workshops, orientation courses, and tutoring services (Lenning, Beal, and Sauer, 1980; Pantages and Creedon, 1978; and Pascarella, 1982). Students report the need for such support services (Weissberg, Berentsen, Crote, Cravery, and Heath, 1982) and marginal students who use such support services experience greater academic success than those who do not (Collins, 1982). These services however, are probably underused (Friedlander, 1980; Lewicki and Thompson, 1982) or are resisted if their use is mandated (Benedict, Apsler, and Morrison, 1977). Even when support services are available, students often prefer to seek help from informal sources. A greater reliance on informal sources would help to explain the underuse of more formal sources of aid (Knapp and Karabenick, 1988). In a study conducted by Knapp and Karabenick (1988), 612 general psychol- ogy students were surveyed. Of those surveyed, 39 students (61 percent) reported no need for academic assistance. The 573 individuals who expressed some degree of need for help with courses or general study skills: 48 SeumeeLHelo ELecuencxeLQentaetuEereenn 0 1 2 3 4 5 6+ Formal Instructors 47 12 14 8 4 6 9 Student Tutors 81 8 4 2 2 1 2 Instructional Center 82 6 3 2 2 1 3 Career Planning and ‘ Placement Center 92 4 2 1 0 0 1 Instructional support center for general study skills 91 4 2 1 0 O 2 Informal Friends 37 5 6 11 7 19 24 Other students in class 43 7 11 9 5 8 16 (Knapp and Karabenick, 1988, p. 225) It is evident that formal sources of assistance other than instructors were rarely used. Even though instructors were more widely employed than other formal sources, 47 percent of the students in need reported they never obtained help from their instructors. The data indicates that informal sources of help were used to a greater degree than their formal counterparts. For example, 63 percent of the stu- dents received help from more knowledgeable friends at least once and 57 percent obtained some level of assistance from other students in their Classes. These per- centages can be contrasted with the 80-90 percent of students reporting no contact with formal sources other than their instructors (Knapp and Karabenick, 1988, p.225) Of the 62 percent of the students (355) who saw at least one formal source, 86 percent (305) saw their instructor rather than using an academic support service. Thus, it becomes apparent that the more formal, institutionalized helping services , W” i 5331 res- LOL 49 were rarely employed as the sole means for obtaining assistance (Knapp and Karabenick, 1988, p.225). The results of Knapp and Karabenick's study points to a pattern of low use of institutional support services, a result consistent with those reported by Friedlander (1980) and Lewicki and Thompson (1982). Although instructors were employed as a source of help by a substantial number of students, other formal sources of help were never employed by more than 80 percent of the students. It is important to point out that this pattern of underuse is based on students who reported some degree of need for academic assistance. The services provided by the institution are the very ones that students often state they need. Other studies conducted by Aitken, 1982; Baumgart and Johnstone, 1977; Bean, 1980; Disque, 1983; Enos, 1981; Kowalski, 1977; Staman, 1980; and Steele, 1978 failed to discover an asso- ciation between student retention and the frequency or quality of their advising. Some studies have found a positive relationship between retention and student's indication of the frequency or quality of the advising. These studies include research conducted by Bergman, Kuk, and Stager, 1982; Endo and Harpel, 1979; Louis, Colten, and Demeke, 1984; Meyers, 1981; Pascarella and Terenzini, 1977; Smith, 1980; and Taylor, 1982. 1111911051 Studying a group of at-risk college freshman enrolled at a four-year university, Abrams and Jernigan (1984) sought to correlate academic success variables as measured by college grade point averages, number of credit hours completed, and persistence in the university, with the traditional predictors of academic achievement, 50 admission test scores, and high school grades. The researchers projected that students' use of tutorial support services and developmental training would affect their academic success. The four hypotheses tested: 1) the number of hours spent in the reading and study skills program and the number of tutor contacts would significantly correlate with college grade point averages and retention of at-risk students in the university; 2) a significant decrease in students' use of the academic support services would occur from fall to winter semester; 3) first semester college grades would significantly correlate with reading test scores; and 4) participation in a structured reading program would result in measurable and significant increases in reading vocabulary, comprehension, and speed. The university involved in Abrams and Jernigan's study provided academic advising assistance, academic support services, tutoring, and counseling. Support— ive services instruction took two forms: 1) students could elect to attend scheduled workshops, or 2) students could come to the support center for individual assis- tance. A total of 229 students participated in the study and academic support ser- vices. Ten students did not complete the fall semester, so the population was re- duced to 219 students. Of the 219 students, 29 withdrew alter the fall semester (9 of the 29 students received fall semester grade point averages of 2.0 or above/passed; and 20 of the 29 students received fall semester grade point averages below 2.0/ failed). Of the 219 students who completed the fall semester, 77 or 35 percent failed and of the 190 students who completed the winter semester 78 or 41 percent failed. Overall, 43 percent of the at-risk college freshmen failed to achieve a C average their first year. In comparison, 3 percent of the university's other students were put on academic probation. 51 For the variable of hours spent in a tutorial program, statistical correlations with fall grade point average, first year grade point average, number of fall semester credit hours, and the total number of credit hours were positive. The difference in the number of hours spent in the tutorial program was statistically Significant when the group was divided into those who passed the fall semester and those who failed it. The difference in hours approached significance when the group was divided subsequently into those who passed the winter semester and those who failed. To test the retention hypothesis, Abrams and Jernigan investigated the differ- ences between the returning winter students and the nonreturning students. The only significantly different variable besides their fall semester grade point averages was the number of hours they received tutorial assistance. The mean number of hours spent in the program for returnees was 14.77 and the mean for nonreturnees was 10.71 (Abrams and Jernigan, 1984). Traditional predictors of academic success are not accurate determinants of academic achievement for at-risk students. The number of hours spent obtaining supportive services and tutoring were the major variables contributing to the aca- demic success achieved by these at-risk students. Abrams and Jernigan deter- mined that 26 percent of the fall semester grade point average variance was ac- counted for by the number of hours spent in a tutorial program. A study of winter grade point averages revealed that the additional variable of tutoring was the second best predictor of winter semester grade point averages. The number of hours of tutoring during the fall semester remained the strongest indicator of winter grade point averages. Abrams and Jernigan's study (1984) demonstrates that at-risk entering 51:11 the mls u: 'nn II I“ let. ill W I 52 students' willingness to seek assistance from tutors is the most accurate predictor of their first-semester college grade point averages. The usual entrance criteria, ad- mission test scores and high school grades were not reliable predictors of academic success for this particular population. At-risk students who took advantage of tutor- ing enjoyed greater academic success. lnstltutionally, the university benefited by retaining 57 percent of these at-risk students through their freshman year. A Learning Center was established at Indiana University of Pennsylvania in 1984 to develop and implement learning enhancement programs for students. The tutorial program is one of the learning improvement services initiated at the institu- tion. The high rate of use during the first year (1,438 students; 5,136 contacts) of implementation indicated that student users believed the services could be beneficial and perhaps raise their course grades and ultimately their overall grade point aver- ages. To determine if students believed their interaction with the tutorial program positively affected their course performance, Harrar and Ender (1987) conducted a survey of students using the tutoring services. Of those contacted, 44 percent reported that they came to supplement Class presentations; 22 percent to get better grades; 15 percent to keep from failing; and 15 percent for other reasons. Students were asked what final grade they believed they would have received if they had not been tutored and what grade they actually received. The responses were: 42 percent thought they would have received a failing grade, but only 7 per- cent reported actually receiving that grade; 12 percent thought they would have received an above-average grade without tutoring; and 27 percent reported receiving a final grade of A or B. Overall, 68 percent of the students surveyed reported that their actual final grade increased at least one grade level over their perceived actual 53 grade without tutoring. Eighty-two percent of the students who believed they were earning a grade of F before tutoring indicated a minimum improvement of one letter grade. Students with a perceived grade of B before tutoring indicated an improve- ment of 40 percent. Students who reported an increase were asked an additional question to determine whether they believed that the Increase was due to tutoring. All (100 percent) replied that their increase was a result of the tutoring they received. Harrar and Ender (1987) concluded that regular users of the Learning Center's tutoring services have a positive view of the program and its effects. The data indicates that failing students have a significant opportunity to improve their course grade with tutoring (especially after 5 visits). A study conducted by House and Wohlt (1990) supports the findings of Abrams and Jernigan (1984) and Harrar and Ender (1987). House and Wohlt deter- mined that academically underprepared college students' participation in a tutoring program during their freshman year was related to their academic performance. Students who participated in the tutoring program earned more credit hours during their first year than did students who did not participate. In addition, participation in tutoring during both semesters of the freshman year was associated with signifi- cantly higher cumulative grade point averages, particularly for male students. These results support tutoring as an important component of programs designed to improve the academic performance and retention of academically underprepared students. II n mi Attrition rates have remained steady at 50-60 percent over four-year spans despite extensive changes in American higher education and administrative efforts 54 to reduce attrition. Counselors, advisors, and psychologists have created numerous interventions designed to reduce attrition. A combination of educational reform movements, declining enrollments in some jurisdictions and rising enrollments in others, exploration of ways to improve the education of a burgeoning population of at-risk students, and increasing pressure on educational budgets has created in- ducements to consider systematically both the costs and effects of alternative courses of action. Federal budgetary deficits have reduced the prospects for addi- tional federal funding to support these educational reforms. Although state and local governments have responded, both levels of government are hard-pressed to ex- pand educational funding in light of rising needs for other constituencies (Levin, 1988). For example, in Michigan, each community college is appropriated to receive $150 per student and $25 for each Pell Grant recipient for counseling, financial aid, admissions, and job placement assistance. The goal is to encourage community colleges to effectively serve at-risk students (State of Michigan Executive Budget, Fiscal Year 1986-87). W Cost benefit analysis is a systematic approach to allocating public sector resources. This technique is particularly useful for determining which programs to support and the appropriate size of those supported. An economic analysis of the costs and benefits of alternative programs of a university counseling center ”poses in particularly stark terms the problems of what to measure and how to measure it" (Weiss and Giddan, p. 261, 1986). Cost-benefit analysis refers to a set of techniques for deciding whether the 55 benefits of a program outweigh its costs. The application of cost-benefit analysis raises several practical issues: 1) the preferences of individuals should be most important versus administrators; 2) emphasis should be on economic efficiency because of the theoretical and practical difficulties of including problems of equity; and 3) measuring and valuing the benefits of projects in dollar terms is difficult. Cost-benefit analysis helps to identify those counseling center programs for which the value of the net benefits is greater than the value of the cost, thereby assisting in decisions regarding which programs are economically efficient. lnstltu- tions need cost-benefit analysis information as part of the decision-making process. To conduct a cost-benefit analysis, the appropriate costs and benefits must be identified and quantified. Then the net present value (i.e. the benefits minus the costs measured at their present value) is estimated through the application of formal decision rules. Proper measurement of the net benefits of a counseling center intervention to reduce attrition requires the following approach: 1) define the pro- gram; 2) define and estimate the net present value of the costs to society; 3) define and estimate the net present value of the benefits to society; and 4) initiate or con- tinue the program only if the net benefits are positive. In a cost—benefit analysis from a counseling program's point of view, a mon- etary value is placed upon all costs and benefits. The costs are those involved with the counseling center program, including staff and administration costs but excluding student fees or the social costs of attrition. Benefits are increased tuition, fees, and subsidies, but do not include individual benefits such as long-term salaries, or other societal and educational benefits. Cost-benefit analysis' appeal is outweighed by practical problems. Unit costs 56 such as per student or cost per session are useful to policy makers because they may indicate the current state of affairs in a counseling center, unusual costs such as a high cost per session, and a basis for gauging and comparing comparable problems or centers at other universities or unit costs. These problems outweigh the benefits and as a result, cost-effectiveness analysis is a more practical method for utilization within educational institutions. Qesfifleetlxeneseenelxels Cost-effectiveness analysis is a comparison of the costs of alternative pro- grams with the major benefit or outcome not measured in dollar terms. One or more decision alternatives are evaluated by comparing the cost of achieving one unvalued output unit and comparing the results across programs. A study of the cost-effec- tiveness of programs to reduce attrition would be based on five steps: 1) definition of the program(s) under consideration; 2) selection of the program(s) that are of interest to the appropriate decision-maker; 3) measurement of the productivity of the program(s); 4) measurement of the costs of the program(s); and 5) calculation of the costs per unit of output. The basic technique has been to derive results for educational effectiveness of each alternative by using standard evaluation procedures or studies (Rossi and Freeman, 1985) and to combine such information with cost data that are derived from the ingredients approach. The ingredients approach was developed to provide a systematic way for evaluators to estimate the costs of social interventions (Levin, 1975; 1983; 1988). The main strength of the ingredients approach is that it is based upon a straightforward approach to estimating costs that is comprehensive to 57 evaluators and policy makers while meeting rigorous standards of economics meth- odology. The ingredients approach to cost estimation entails three distinct phases: 1) identification of ingredients, 2) determination of the value or cost of the ingredients and the overall cost of an intervention, and 3) an analysis of the cost in an appropri- ate decision-oriented framework. The first step is to ascertain which ingredients are required for an intervention (Levin, 1983). Most educational interventions are labor-intensive, so an initial con- cern is to account for the number and Characteristics of personnel. Beyond this, it is necessary to identify the facilities, equipment, materials, and other resources which are required for the intervention. Identification of ingredients requires a level of detail that is adequate to ensure that all resources are included and are described ad- equately to place cost values on them. Once the ingredients have been identified and stipulated, it Is necessary to ascertain their costs (Levin, 1983). In doing this, all ingredients are assumed to have a cost to someone, even if the institution did not pay for them in a particular situation. Once each of the ingredients is costed, the costs can be added to obtain a total cost for the intervention. The next stage entails the use of these costs in an analytic framework. The two most important concerns for cost summary and analy- sis are the appropriate unit for expressing costs and who pays the costs. Cost- effectiveness ratios are usually based on the average effects and costs per student. Who pays the cost is a different issue. The overall cost-effectiveness ratio may be irrelevant to a decision-maker who pays only part of the cost for one inter- vention, but all of the cost for an alternative. Therefore, cost-effectiveness analysis was designed to assist decision-makers to choose among successful practices on - 58 the basis of which ones would maximize the impact of available resources. Counseling center programs to reduce attrition can supply more services and increase the number of students persisting in school if they are run efficiently. Effi- ciency in economic terms is defined as maximum output (such as additional student credit hours generated if students remain in school) for a given budget or for use of resources. Data on students, staffing, and other resources allow the measurement of average and marginal output by constructing a measure of the relationship be- tween the input of staff and equipment into the program and the output of added attendance at the college solely because of the counseling center intervention. The relationships among program resources then may be combined with increased attendance due to counseling. The production relationships may be used as the basis for measurement of program costs and benefits. Counseling centers are concerned with the identification, measurement, and usefulness of cost information in relation to programs to prevent attrition. Questions about a counseling center's efficient use of resources or the relation of costs to common outcomes of decision alternatives require that costs be correctly measured. Administrators must distinguish average from marginal costs and become more familiar with the concept of unit cost. Cost-effectiveness analysis in retention studies is a worthy goal. When based on careful evaluation techniques, it holds genuine promise for improving the efficiency and effectiveness of student retention programs. Cost-effectiveness can be used to determine which programs are worthy of pursuing, given a limited budget, to help design programs based on staff parameters and to target them to students who meet decision-makers' efficiency and equity criteria (Levin, 1988). 59 Current trends such as a declining 18 to 24 year old population, expanding enrollment of nontraditional students, and rapidly rising costs of higher education tend to increase the likelihood of application of cost-effectiveness in counseling and student development studies of attrition. Educators and administrators need to know if retaining students by counseling is less costly and more effective than are recruit- ing, admitting, and enrolling them. The answer depends on the application of cost- effectiveness analysis to retention and admission programs. Key indicators are whether counseled students stay in college or return once they have temporarily ”stopped out”. In the Balance-Sheet Method (Russell, 1986) of presenting items for a cost- effectiveness analysis, the ingredients used for individual college counseling were actually identified by the center staff of the participating college. The cost of each of the ingredients used in individual counseling was estimated from the counseling center budget. Based upon staff goals, objectives, and performance evaluations, it was found that approximately 60 percent of counseling center resources were allo- cated to individual counseling. The costs of the ingredients do not include support costs for the remainder of the institution or other overhead costs because they would be unlikely to change because of this program. Additional costs of administration or maintenance would likely be small and not amenable to easy calculation from the institutional budget. Estimates of the costs of space and equipment are arbitrary, so they are also not included. The most costly counseling ingredient is the professional staff. Application of cost-effectiveness analysis to counseling center activities could be used with other relevant outcomes such as grade point average, enrollment, 60 educational benefits, and graduation rates, which are useful in a policy making context. Kiesler and Morton (1988) provide the following argument: ”For maximized policy relevance, outcome studies Should include cost information so...planners can evaluate treatment outcomes in terms of direct and indirect financial impact” (Kiesler and Morton, 1988, p. 1001). The literature provided numerous examples of student retention studies and the impact of academic support services, particularly counseling and academic advising. However, a majority of these studies focused on four-year institutions. Limited research was conducted on community college students, and none ad- dressed the impact of using grant-sponsored academic support services to enhance special population students' academic achievement and persistence. CHAPTER III METHODOLOGY BeseaLctLMetnedelecunneSJudv A sample of students attending one community college were surveyed and corresponding institutional data reviewed. Delta College was selected due to the accessibility of data, the number of grant-sponsored programs available for inclusion in the study, and the increasing enrollments of Special population students due to its location in a heavily industrialized manufacturing region. The research methodology was a quasiexperimental design. The experimen- tal group was 200 randomly sampled Special population students enrolled at Delta College during the Winter 1991 semester and the control group was 200 ”matched pairs" students enrolled during the Winter 1991 semester with characteristics similar to experimental group students. An institutional group of 500 students enrolled during the Winter 1991 semester was selected by random sample to permit statisti- cal comparison to the overall Delta College student body. Institutional group stu- dents' characteristics were not matched to either the experimental or control group students. 61 62 Researchmlllatlen Special population students enrolled during the 1990-91 academic year at Delta College were identified through lists generated by Single Parent/Homemaker Program, At-Risk Student Success Program, Michigan Job Opportunity Bank-Refrain Program, and Special Needs Program-disadvantaged component grant staff. The initial lists identified 156 students served through the At-RiSk Student Success Program, 165 students received assistance through the Single Parent/Homemaker Program, 313 students were served through the Special Needs Program-disadvan- taged component and 23 students enrolled in the Michigan Job Opportunity Bank- Retrain Program during the 1990-91 academic year, for a total of 653 grant-Spon- sored students. Since students were eligible to receive assistance through multiple grant programs, the lists were cross-referenced by student name and student num- ber. It was determined that a total of 85 students were served by multiple grant programs, and duplicate names were removed from the population (77 stUdents were served by both the At-Risk Student Success Program and the Special Needs Program-disadvantaged component; 4 Single Parent/Homemaker Program students were served by the At-Risk Student Success Program and Special Needs Program- disadvantaged component; 3 Single Parent/Homemaker Program students were served by the Special Needs Program-disadvantaged component; and 1 Single Parent/Homemaker Program student received services through the At-Risk Student Success Program). As a result of students receiving services from multiple grant programs, the initial research population was reduced to 568 unduplicated grant- sponsored students. 63 Enrollment in Wlnter 1991 semester courses was a criteria for students to be included in the study. The researcher accessed the College's PRIME computer system database by student name and student number and reviewed students' schedules and/or transcripts to verify enrollment in VWnter 1991 semester courses. Due to the criteria, the research population was narrowed from 568 unduplicated grant-sponsored students served during the 1990-91 academic year to 422 grant- sponsored students who were enrolled during the Winter 1991 semester. The research population included 248 students in the Special Needs Program-disadvan- taged component, 100 Single Parent/Homemaker Program students, 55 students served through the At-Risk Student Success Program, and 19 students who partici- pated in the Michigan Job Opportunity Bank-Retrain Program during the Winter 1991 semester. Once the students' Winter 1991 semester enrollment was confirmed, the 422 grant-sponsored students were listed by name and student number in a computer- ized database (Dc-File). At this point, the student was no longer associated with a particular grant program (Single Parent/Homemaker Program, At-Risk Student Success Program, Special Needs Program-disadvantaged component, or Michigan Job Opportunity Bank-Retrain Program) and became a ”generic" grant-sponsored student enrolled during the Winter 1991 semester. Delta College's PRIME computer system was used to identify a random sample of 200 special population students for inclusion in the experimental group. Once the random sample was completed, the characteristics of the experimental group students were identified including gender, ethnicity, age, ASSET scores for the writing, math and reading comprehension subtests, enrollment status during the 64 Winter 1991 semester, cumulative credit hours upon completion of the Fall 1990 semester, and program major. This Information was obtained through a combination of college records and computerized databases. Based on the characteristics of the 200 randomly sampled experimental group students, a ”matched pairs” control group of 200 non-special population students attending Delta College during the Winter 1991 semester were identified using the PRIME computer and placed in a database file (pc-File). It was planned that at least four characteristics of experimental group students would be matched with control group students. Special population students in the experimental group had counterparts in the control group with six similar characteristics including gen- der, ethnicity, age (Classified in age ranges 15-19, 20-24, 25-29, 30-34, 35-39, 40- 44, 45-49, 50-54, 55-59, 60-64, and 65-69), enrollment status (classified in ranges of credit hours 0-11 and 12 and above), cumulative credit hours completed at the end of the Fall 1990 semester (classified in ranges of credit hours 1-14, 15-29, 30-44, 45- 59.6074, 75-89, 90-104, 105-119, 120-134, and 135 and above) and program major. The computerized databases (pc-File) for both the experimental and control groups were placed in individual Statistical Package for the Social Sciences (SPSS) database files. Statistical Package for the Social Sciences (SPSS) was used to compute and verify that the means of the experimental and control groups were not statistically different at the .05 level of significance for the characteristics of gender, ethnicity, age, enrollment status, cumulative credit hours, and program major. To obtain a control group that was statistically significant at the .05 level, it was neces- sary to classify the characteristics of age, cumulative credit hours, and enrollment 65 status into ranges. ASSET scores for the writing, math and reading comprehension subtests were an unmatched variable. Delta College requires all full-time students and students enrolling after the Winter 1987 semester to have taken the ASSET test. Due to the part-time enrollment status of 33 percent of the experimental group students or their semester of initial enrollment occurring prior to the Winter 1987 semester, a large number of students (60 percent) in the experimental group had not taken the ASSET test and did not have ASSET scores on file. Therefore, this char- acteristic was not matched. Independence between the experimental and control groups was assured since experimental group Students were removed from the population prior to the selection of control group students. An institutional group of 500 students enrolled during the Winter 1991 semes- ter was randomly sampled using the PRIME computer database and placed in pc- File and Statistical Package for the Social Sciences (SPSS) databases. Unlike the experimental and control groups, there was no attempt made to match institutional group students' characteristics with those of experimental group students. Institu- tional group students permitted statistical comparisons to the overall Delta College student body. Independence between the experimental, control, and institutional groups was assured since experimental and control group students were removed from the population prior to conducting the random sample to obtain the 500 institutional group students. 66 WW! The research questions were formulated into null hypotheses to determine if there were differences in the academic achievement, persistence, and usage of academic support services by experimental group (special population) students and control group (non-special population) students. The researcher compared three groups to determine if differences existed between: experimental and control group students; experimental and institutional group students; and control and institutional group Students. Hypothesis; The mean hours of institutional academic support services used during the Winter 1991 semester by control group students will not statistically significantly exceed (p > .05) the mean hours of use by experi- mental group students. Hypothesis; The mean hours of academic support services used during the Winter 1991 semester by experimental group students will not statistically significantly exceed (p > .05) the mean hours used by control group students. Hypothesis}. The experimental group students' Wlnter 1991 semester mean grade point average will not statistically significantly (p > .05) exceed the Winter 1991 semester grade point average of control group students. Hypothesjsfi. Experimental group students' proportion of Winter 1991 se- mester courses completed will not statistically significantly exceed (p > .05) the proportion of courses completed during the Wlnter 1991 semester by students in the control group. Hypothesjsj. The Fall 1991 semester enrollment of experimental group students will not statistically significantly exceed (p > .05) the Fall 1991 semester enrollment of control group students. Hypothesjsfi. The mean grade point average during the \Mnter 1991 semes- ter for the experimental group will not statistically significantly (p > .05) exceed the institutional group students' mean grade point average. Hypothesjsl. The proportion of courses completed during the Wlnter 1991 semester by the institutional group will statistically significantly exceed (p > .05) the proportion of courses completed by the experimental group. 67 Hypothesjsfi. The enrollment of experimental group students in Fall 1991 semester courses will not statistically significantly exceed (p > .05) the pro- portion of institutional group students. Hypothesis-.3. The institutional group students' Winter 1991 semester mean grade point average will not statistically significantly exceed (p > .05) the Winter 1991 semester mean grade point average of control group students. Hypothesjm. The proportion of courses completed by institutional group students during the Winter 1991 semester will not statistically significantly exceed (p > .05) the proportion of courses completed during the Winter 1991 semester by control group students. Hypothesjsjl. Fall 1991 semester enrollment by institutional group students will not statistically significantly (p > .05) exceed the enrollment of students in the control group. W Experimental group students were surveyed to determine their usage of both grant-sponsored and institutional counseling and tutorial academic support services during the Winter 1991 semester. A questionnaire, cover letter, and a stamped self- addressed return envelope were sent to the 200 experimental group students on Nevember 15, 1991. The questionnaire requested information on: 1) the number of hours of grant-sponsored counseling accessed during the Vlfinter 1991 semester, 2) the number of hours of grant-sponsored tutoring accessed during the Wlnter 1991 semester, 3) the number of hours of institutional counseling accessed during the Winter 1991 semester, 4) the number of hours of institutional tutoring accessed during the Winter 1991 semester, and 5) whether the academic support services accessed were beneficial in assisting experimental group students to achieve their academic goals. Since grant-sponsored services were offered in a variety of physical locations on campus and at a major off-campus center, these locations 68 were identified on the survey to assist grant-sponsored Students to identify and differentiate between the services provided by grant and institutional funding. All special population students identified by random sample for the experimental group were surveyed. A copy of the questionnaire and cover letter sent to experimental group students is included in Appendix A. Concurrently, the 200 control group students were surveyed to determine their usage of institutional counseling and tutorial academic support services during the Vlfinter 1991 semester. A second questionnaire, cover letter, and stamped, self- addressed return envelope was distributed to the 200 control group students on November 15, 1991. This questionnaire requested information regarding: 1) the number of hours of institutional counseling accessed during the Winter 1991 semes- ter, 2) the number of hours of institutional tutoring accessed during the Wlnter 1991 semester, and 3) whether the academic support services accessed were beneficial in assisting control group students to achieve their academic goals. A copy of the questionnaire sent to control group students is included in Appendix B. Three weeks later (December 5, 1991), a follow-up letter was sent to remind both experimental and control group students of the previous request and to ask them to complete and return the questionnaire as soon as possible. A third reminder letter was mailed on December 18, 1991 along with another copy of the question- naire and a stamped, self-addressed return envelope. All survey responses were confidential and the surveys were not coded. If a respondent indicated his/her name on the return envelope, he/she did not receive follow-up mailings. Copies of the follow-up correspondence to experimental and control group students is included in Appendix C. 69 The cooperation and assistance from participants was similar to other community college studies surveying at-risk students, including those conducted by Delta College personnel. Five surveys were undeliverable due to students relocating without forwarding addresses. Undeliverable surveys were not included in the study. Tables 1 and 2 present the patterns of responses to the various mailings. The overall response rate seemed satisfactory for a mailed questionnaire, particularly when 40 percent of the students surveyed were no longer enrolled at the College. It should be noted however, that the final results contain proportionately more responses from control group Students than Special population students. Table 1.--Resp0nse Rate to Survey. Mailing % Returned Cumulative % First 58 surveys (15%) Second 23 surveys ( 6%) 21% Third 95 surveys (24%) . 45% Table 2.--Response Rate by Group. Type of Group % Returned Cumulative % Experimental 72 surveys 37% of group Control 104 surveys 53% of group Designeuhefiuestlennam The questionnaires were developed in draft form and reviewed by Dr. Gene Packwood, the Director of Research and Development at Delta College. 70 Dr. Packwood has extensive experience in the design and use of mailed question- naires. Based on Dr. Packwood's advice, the questionnaires were revised and modified for experimental and control group students' easier reading and under- standing. . A total of twenty Delta College students using the Teaching/Learning Center on October 24, 1991 were asked to review the revised questionnaires and comment on their readability, clarity, and ability to understand the information being requested. Ten students completed each questionnaire, and based on their advice, the ques- tionnaires were slightly revised to improve clarity and understanding. Due to the differences in the information being requested from the experimen- tal and control group students, two questionnaires were developed. They differed in one way: the control group questionnaire did not request information on students' utilization of grant-sponsored academic support services. Since grant-sponsored academic support services are limited in use to grant-eligible students, these ques- tions were not appropriate for control group students. W The plan for conducting this study was described in writing and submitted to the Committee on Research Involving Human Subjects at Michigan State University to be certain it conformed to legal and ethical standards. The Committee felt the Study was exempt from review and approved the plan as submitted. The letter of approval from the Committee Chairman, Dr. David E. Wright (November 8, 1991) is in Appendix D. The study was conducted as outlined in the plan. In all communications with students, they were assured of the confidentiality 71 of their response. The surveys were not coded and follow-up mailings to respon- dents were eliminated only if they provided their name and address on the return envelope. No respondents were identified by name and the results have been used in an aggregate form, untraceable to an individual student. On each questionnaire, both experimental and control group students were given the opportunity to provide open-ended comments. The comments are re- corded in Appendix E. The returned experimental and control group students' questionnaires were verified by Delta College data-entry staff. The data were entered into the study's computerized databases in the Statistical Package for the Social Sciences (SPSS) software and stored in a PRIME computer system. QellectleneLthenata Two computerized Statistical Package for the Social Sciences (SPSS) data- bases were established to collect and store information related to the study (experi- mental and control group). Institutional data regarding experimental and control group students' characteristics (gender, ethnicity, enrollment status, cumulative credit hours, ago and program major) were available and stored in Delta College's PRIME computerized databases (pc-File and SPSS). This information was ac- cessed by the experimental and control group student's student number, and the information was placed in the appropriate group's database. Additional information including the number of Winter 1991 credit hours enrolled, the number of Winter 1991 credit hours completed, the number of Winter 1991 credit hours withdrew, the student's Wlnter 1991 grade point average, the number of hours of institutional _ 72 counseling accessed in the Winter 1991 semester, the number of hours of institu- tional tutoring accessed during the Wlnter 1991 semester, the number of hours of grant-sponsored counseling accessed during the \Mnter 1991 semester (experimental group students only), the number of hours of grant-sponsored tutoring accessed during the Wlnter 1991 semester (experimental group students only), the number of credit hours enrolled during the Spring/Summer 1991 semester, the number of credit hours completed during the Spring/Summer 1991 semester, the number of credit hours the student withdrew during the Spring/Summer 1991 semes- ter, the student's Spring/Summer 1991 grade point average, the number of credit hours the student enrolled in during the Fall 1991 semester, and whether the student graduated during the Winter 1991 or Spring/Summer 1991 semesters was obtained through the questionnaires, grant staff records, institutional staff records, student transcripts, student schedules, and student admission applications and entered into the appropriate group's computerized database. There was a non-bias factor for institutional staff providing data on students. In the review of institutional records, students in the experimental and control groups were not identified to staff. A third computerized database (SPSS), was established for the 500 students who comprised the institutional group. By using the student's student number to access existing computerized databases, information was obtained on institutional group students that included: the number of Wlnter 1991 credit hours enrolled, the number of Wlnter 1991 credit hours completed, the number of Wlnter 1991 credit hours withdrawn, their Vlfinter 1991 semester grade point average, the number of Spring/Summer 1991 semester credit hours enrolled, the number of Spring/Summer 1991 credit hours completed, the number of Spring/Summer 1991 semester credit 73 hours withdrawn, their Spring/Summer 1991 semester grade point average, the number of Fall 1991 semester credit hours enrolled and whether the student graduated during the Wlnter 1991 or Spring/Summer 1991 semesters. W Once the flow of return questionnaires had stopped, the information from the questionnaires was verified by Delta College data-entry staff, entered into the appro- priate database (experimental or control group) and stored in the PRIME computer system in Statistical Package for the Social Sciences (SPSS) databases. All analy- sis of the data was completed on the PRIME computer system, using the Statistical Package for the Social Sciences (SPSS) software package that provides frequency tables and most of the standard statistical tests used in educational research. 518.115.1125 The respondents involved in this study represented potentially all of the research universe of interest to the inquiry. As indicated in the paragraph on re- sponse rates, not all possible respondents completed and returned questionnaires, but all were contacted and encouraged to participate. In view of the 45 percent rate of return of the questionnaires, it was assumed for purposes of this research that respondents and non-respondents were similar. Usually in studies of this type, the data comes from a sample of a larger research population. Most statistics measure samples of a research population to determine the likelihood that other samples from the same population would produce comparable results. Normally, data are reported from the sample, and efforts are 74 made to generalize to the larger research population. In that process, null hypoth- eses are prepared and tested using one of several mathematical formulas. After computing a score with a standardized table, it is possible to conclude whether the initial hypotheses should be accepted or discounted. This process determines the statistical significance of the data (Glass and Hopkins, 1984). When data from the entire research population are being considered instead of from a sample, the process is much simpler. The data are summarized in an appropriate manner and the results compared with hypotheses. There is no need to perform further mathematical calculations to generalize about the trends of the responses because the data already reflect the total research population as re- ported. This process produces conceptual significance (Glass and Hopkins, 1984). In this study, the usage of academic support services, student academic achievement (grade point average), and student persistence (the proportion of Winter 1991 courses completed and subsequent enrollment in the Fall 1991 semes- ter) was reported and compared with hypotheses to determine if they were concep- tually significant. When specific hypotheses contained differences between seg- ments of the population or variables to be tested, the hypotheses were tested by computing t-test for independent means and Chi-Square statistics as appropriate, and using the .05 level of significance as the decision point. In those instances, statistical significance was determined and reported. The justification for using the Host for independent means and Chi-Square statistics is due to the type of data analyzed and the recognition that the students in the experimental, control, and institutional groups were not the entire research population. The statistics utilized allowed the researcher to measure differences in 75 the samples, evaluate the data, and permit comparisons to the research population. ElnanelalJmeaeuzcmecnentznatafiellectlen A secondary focus of this study was to estimate the financial impact on a community college's budget as a result of providing academic support services to special population students. While the literature indicates that there are several methods that can be used to estimate the financial impact to the college of providing counseling and tutorial academic support services to special population students, a simple cost analysis method was used in this study. To determine the institutional costs of providing counseling services to stu- dents, the 1990-91 fiscal year expenditures for counseling services were identified from the College's budget and divided by the number of academic credit hours for all institutional enrollments during the Fall 1990, Wlnter 1991, and Spring/Summer 1991 semesters to obtain a mean cost per academic credit hour of institutional counseling. The fOrmula was: - ° ° ° ' = Mean cost per academic Number of academic credit hours for institutional credit hour of institu- enrollments during the Fall 1990, Winter 1991, tional counseling and Spring/Summer1991 semesters services The calculation was: $1513.32 = $3.90 per academic 196,749 ' credit hour Information on the cost of operating the Teaching/Learning Center during the 1990-91 academic year was also obtained from the College budget. A similar 76 formula was derived to obtain a mean cost per academic credit hour of institutional tutorial academic support services: - Mean cost per academic Number of academic credit hours for institutional , credit hour of institu- enrollments during the Fall 1990, \Mnter 1991, tional tutorial services and Spring/Summer 1991 semesters The calculation was: 51.65.9115 = 3.839 per academic 196,749 credit hour Control group students enrolled in a total of 1,643 academic credit hours during the Winter 1991 semester. At the mean rate of $3.90 per academic credit hour, the calculation ($3.90 x 1,643 = $6,407.70) provided an estimate of the institu- tional cost of providing counseling services to control group students. A similar calculation was conducted to estimate the mean cost per academic credit hour of institutional tutoring. At the mean rate of $839 per academic credit hour, the calculation ($.839 x 1,643 = $1,378.48) provided an estimate of the institu- tional cost of providing tutorial services to control group students. Experimental group students can access institutional counseling in addition to grant-sponsored services. Experimental group students enrolled in a total of 1,669 academic credit hours during the Winter 1991 semester. At the mean rate of $3.90 per academic credit hour, the calculation ($3.90 x 1,669 = $6,509.10) provided an estimate of the institutional cost of providing counseling services to experimental group students. Likewise, a calculation was used to estimate the institutional cost of providing 77 tutorial services to experimental group students. The calculation ($.839 x 1,669 = $1,400.29) indicated the cost of providing institutionally-supported tutorial services to special population students. The costs associated with providing grant-sponsored counseling and tutorial academic support services available strictly to special population students was estimated based on the cost per academic credit hour of providing each service. Grant expenditures during the 1990-91 academic year for counseling were identified through final grant reports (Single Parent/Homemaker Program and Special Needs Program-disadvantaged component) and College financial records (Michigan Job Opportunity Bank-Refrain Program and At-Risk Student Success Program), includ- ing institutional matching funds designated for that purpose. Once these expendi- tures were identified, the total was divided by the number of academic credit hours generated by grant-sponsored students enrolled in the Fall 1990, Wlnter 1991, and Spring/Summer 1991 semesters (credit hours Obtained through student schedules stored in the College's PRIME computer system) to obtain a mean cost per aca- demic credit hour of grant-sponsored counseling funded by both the grants and the institution. The calculation is represented by: Number of academic credit hours for grant-sponsored students enrolled during the Fall 1990, Wlnter 1991, and Spring/Summer 1991 semesters The calculation was: $293151 9,574 Mean grant cost per academic credit hour of grant-sponsored coun- seling services $30.68 per academic credit hour 78 At the mean rate of $30.68 per academic credit hour, the total cost of provid- ing grant-sponsored counseling services to experimental group students ($30.68 x 1,669) was $51,205.00. This calculation indicates cost estimates of providing grant- sponsored counseling to experimental group students. A similar calculation was made to determine a mean cost per academic credit hour of grant-sponsored tutorial academic support services funded by both grants and institutional matching funds. Grant expenditures for tutorial services during the 1990-91 academic year were obtained through final grant reports (Single Parent/ Homemaker Program and SpeCial Needs Program-disadvantaged component) and College financial records (Michigan Job Opportunity Bank-Refrain Program and At- Risk Student Success Program). Once these expenditures were identified, the total was divided by the number of academic credit hours generated by grant-sponsored students enrolled in the Fall 1990, Wlnter 1991, and Spring/Summer 1991 semesters (credit hours obtained through student schedules stored in the College's PRIME computer system) to obtain a mean cost per academic credit hour 0f grant-spon- sored tutorial services funded by grants and the institution. The calculation was: - ' ' = Mean grant cost per Number of academic credit hours for grant- academic credit hour of sponsored students enrolled during the Fall grant-sponsored tuto- 1990, Winter 1991, and Spring/Summer rial services 1 991 SBITIBSIBI‘S The calculation was: 51.1.5255 = $12.14 per academic 9.574 credit hour 79 Therefore, the total cost of providing grant-sponsored tutoring to experimental group students was ($12.14 x 1,669) $20,261.66. These calculations provided cost estimates on the financial implications incurred by a community college for providing counseling and tutorial academic support services to Special population students. These implications will be further analyzed in Chapter IV. Summer Two questionnaires were developed and mailed to 200 experimental and 200 control group Students enrolled at Delta College during the Wlnter 1991 semester. The original and two follow-up mailings resulted in a final response rate of 37 per- cent for the experimental group and 53 percent for the control group. The question- naires were developed with expert advice and recommendations from a group of students. Each student participating in the survey was guaranteed confidentiality of his/her response and this was observed during the study. Descriptive statistics were used to analyze the data (t-test for independent means and Chi-Square) and where cross-tabulation was necessary, the Statistical Package for the Social Sciences (SPSS) software was used to compute the data and test the null hypotheses. The .05 level of significance was used to determine the statistical Significance of a rela- tionship. A secondary focus of the study was to determine an estimate of the financial impact on a community college's budget as a result of providing academic support services to Special population students. Based on information from institutional budgets and grant expenditure reports, financial data were used with institutional 80 enrollment data to estimate the cost of providing academic support services per academic credit hour. These calculations will provide data estimates on the cost implications incurred by a community college for providing counseling and tutorial academic support services to special population students. CHAPTER IV RESULTS OF THE DATA ANALYSIS This study provided the opportunity to test a number of hypotheses to deter- mine if eligible community college special population students used grant-sponsored academic support services and if these services improved their academic achieve- ment and persistence. A secondary focus of the study was to estimate the financial impact to a community college's budget as a result of providing academic support services to special population students. A total of 72 experimental group (grant- sponsored) and 104 control group students responded to the survey, for a total of 176 responses. However, one experimental group survey response was not in- cluded in the data analysis. In reviewing the respondent's estimates of institutional and grant-sponsored academic support services used during the Winter 1991 se- mester, the usage estimates appeared exorbitant. In checking institutional staff records in the Teaching/Learning Center (respondent indicated name on return envelope), it appeared that this respondent overestimated her usage by a consider- able amount (approximately 900 hours) and the response to that question was invalid. Since one response appeared invalid, the entire questionnaire was suspect and was not included in the data analysis. Therefore, the data analysis was 81 82 conducted on 175 respondents; 71 experimental and 104 control group students. In the pages that follow, data are presented that pertain to each of the null hypotheses stated in Chapter I. Besuitseflhefitudx AS indicated in Chapter I, the researcher developed null hypotheses to com- pare three groups to determine if differences existed between: experimental and control group students; experimental and institutional group students; and control and institutional group students. WW lnstltutlenaljeseurces In an attempt to determine if experimental or control group students used a greater amount of institutional academic support services, data were collected to examine three research hypotheses that were incorporated into one null hypothesis. First, it was hypothesized that experimental group students would access less insti- tutional counseling services. The data in Table 3 indicates that by their estimate, there was no statistically significant difference in the amount of institutional counsel- ing services used by experimental group and control group students during the Winter 1991 semester. Experimental group students reported an average of 1.24 hours of institutional counseling during the Wlnter 1991 semester and by their esti- mate, control group students reported an average of 1 hour. Null Hypothesis: The mean hours of institutional academic support services used during the Winter 1991 semester by control group students will not statistically significantly exceed (p > .05) the mean hours of use by experimental group students. 83 Alternative Hypothesis: The mean hours of institutional academic support services used during the Wlnter 1991 semester by control group students will statistically Significantly exceed the mean hours of use by experimental group students. Statistical Test of Significance: t-test for independent means; 1-tailed probability. Decision: p .238 > .05; accept the null hypothesis. Reject the alternative hypoth- esis. Table 3.--lnstitutional Counseling Services Accessed During the Wlnter 1991 Se- mester. Group N Y SD SE 1 DF l-Tail Prob. Experimental 71 1.2394 2.226 0.264 0.71 173 .238 Control 104 1.000 2.145 0.210 While it was not hypothesized, experimental and control group students were asked to indicate if the Counseling Center services they accessed during the Winter 1991 semester were beneficial in assisting them to meet their academic objectives. The responses of the 37 experimental group and 58 control group students are indicated in Table 4. Table 4.--Students' Perceptions of the Benefits of Institutional Counseling Services Accessed Winter 1991 Semester. Rating Experimental Control Group (n=37) Group (n=58) Very Beneficial 41 % 41 % Beneficial/Somewhat Beneficial 54 % 52 % Of Little Benefit 0 % 5 % No Benefit 5 % 2 % 84 Additional comments from experimental and control group respondents are included in the open-ended comments included in Appendix E. The second academic support service available to assist students was tutor- ing. It was hypothesized that experimental group students would use less institu- tional tutoring assistance from the Teaching/Learning Center. The data in Table 5 indicates that by their estimate, there was no significant difference in the amount of institutional tutoring used by experimental group students and control group students during the Winter 1991 semester. Experimental group students used the College's Teaching/Learning Center an average of 3.55 hours during the Winter 1991 semes- ter and by their estimate, control group students used an average of 3.43 hours of tutorial assistance from the College's Teaching/Learning Center during the Winter 1991 semester. Null Hypothesis: The mean hours of institutional academic support services used during the Winter 1991 semester by control group students will not statistically significantly exceed (p > .05) the mean hours of use by experimental group students. Alternative Hypothesis: The mean hours of institutional academic support services used during the Wlnter 1991 semester by control group students will statistically significantly exceed the mean hours used by experimental group students. Statistical Test of Significance: t-test for independent means; I-tailed probability. Decision: p .463 > .05; accept the null hypothesis. Reject the alternative hypoth- esis. Table 5.-- Institutional Tutorial Academic Support Services Accessed Wlnter 1991 . Semester. Group N TI" SD SE t DF 1-Tail Prob. Experimental 71 3.5493 6.742 0.800 0.09 1 73 .463 Control 1 04 3.4327 8.948 0.877 85 While it was not hypothesized, experimental and control group students were asked to indicate if the services obtained from the Teaching/Learning Center during the Wlnter 1991 semester were beneficial in assisting them to achieve their aca- demic objectives. Thirty-one experimental group students and 32 control group students responded to the question. The results are in Table 6. Table 6.--Students' Perceptions of the Benefits of Institutional Tutoring Services Accessed Wlnter 1991 Semester. Rating Experimental Control Group (n=31) Group (n=32) Very Beneficial 68 % 63 % Beneficial/Somewhat Beneficial 32 % 31 % Of Little Benefit 0 % 6 % No Benefit 0 % 0 % Additional comments from experimental and control group respondents are included in the open-ended comments included in Appendix E. As a summary, Table 7 indicates the third component of the-research hypoth- esis, the combined hours of institutional counseling and tutorial academic support services accessed during the Vlfinter 1991 semester by experimental and control group students. It was hypothesized that experimental group students would access less total institutional academic support services than control group students. Based on their estimate, there was no statistical difference in the amount of institutional academic support services used by experimental group students as compared to control group students during the Winter 1991 semester. Experimental group stu- dents reported an average of 4.79 hours of institutional academic support services during the Winter 1991 semester and control group students reported an average of 86 4.43 hours of institutional academic support services in the same timeframe. Null Hypothesis: The mean hours of institutional academic support services used during the Winter 1991 semester by control group students will not statistically Significantly exceed (p > .05) the mean hours of use by experimental group students. Alternative Hypothesis: The mean hours of institutional academic support services used during the Wlnter 1991 semester by control group students will statistically significantly exceed the mean hours used by experimental group students. Statistical Test of Significance: t—test for independent means; 1-tailed probability. Decision: p .387 > .05; accept the null hypothesis. Reject the alternative hypoth- esis. Table 7.--Total Institutional Academic Support Services Accessed Winter 1991 Semester. Group N x SD SE t DF 1-Tail Prob. Experimental 71 4.7887 7.111 0.844 0.826 173 .387 Control 104 4.4327 9.759 0.957 r - T It was hypothesized that experimental group students would use a greater amount of total academic support services during the Wlnter 1991 semester due to their eligibility to receive supplemental counseling and tutoring services through grant—sponsored programs in addition to institutional counseling and tutoring. The data in Table 8 indicates that by their estimate, experimental group students used an average of 5.77 total hours of institutional and grant-sponsored academic support services during the Winter 1991 semester and control group students used an average of 4.43 total hours. 87 Null Hypothesis: The mean hours of academic support services used during the Wlnter 1991 semester by experimental group students will not statistically signifi- cantly exceed (p > .05) the mean hours used by control group students. Alternative Hypothesis: The mean hours of academic support services used during the Wlnter 1991 semester by experimental group students will statistically signifi- cantly exceed the mean hours used by control group students. Statistical Test of Significance: t-test for independent means; 1-talled probability. Decision: p .164 > .05; accept the null hypothesis. Reject the alternative hypothesis. Table 8.--Total Institutional and Grant-Sponsored Academic Support Services Ac- cessed Winter 1991 Semester. Group N x SD SE t DF 1-Tail Prob. Experimental 71 5.7746 7.353 0.873 0.98 173 .164 Control 104 4.4327 9.759 0.957 While it was not a hypothesis, experimental group students were asked if the supplemental counseling and tutorial services provided through grant funding during the Winter 1991 semester were beneficial in assisting them to achieve their aca- demic objectives. The perceptions of the 27 experimental group students responding to the counseling question and 5 students responding to the tutoring question are indicated in Table 9. 88 Table 9.-Students' Perceptions of the Benefits of Grant-Sponsored Academic Sup- port Services. Rating Grant-Sponsored Grant-Sponsored Counseling (n=27) Tutoring (n=5) Very Beneficial 52 % 20 % Beneficial/Somewhat Beneficial 37 % ‘ 80 % Of Little Benefit 4 % 0 % No Benefit 7 % 0 % Additional comments from experimental group respondents are included in the open-ended comments in Appendix E. W We It was hypothesized that due to the availability and use of supplemental academic support services that experimental group students' \Mnter 1991 semester mean grade point average would be higher than control group students. The data in Table 10 indicates that control group Students who did not access supplemental academic support services during the Wlnter 1991 semester earned a significantly higher mean grade point average during that semester (2.72) than did experimental group students who accessed supplemental grant-sponsored academic support services (2.23). Null Hypothesis: The experimental group students' Wlnter 1991 semester mean grade point average will not statistically significantly (p > .05) exceed the Winter 1991 semester grade point average of control group students. Alternative Hypothesis: The experimental group students'Wlnter1991 semester mean grade point average will statistically significantly exceed the Wlnter 1991 semester grade point average of control group students. 89 Statistical Test of Significance: t-test for independent means; 1-tailed probability. Decision: control group mean exceeded experimental group mean; accept the null hypothesis. Reject the alternative hypothesis. Table 10.-Winter 1991 Semester Mean Grade Point Average Experimental and Control Group Students. Group N ? SD SE t DF I-Tail Prob. Experimental 200 2.2298 1.054 0.075 -4.79 398 <.001" Control 200 2.7197 0.993 0.070 " Control group mean exceeded experimental group mean. One of the unmatched characteristics between experimental and control group students was an assessment of their basic skills. While incoming, full-time Delta College students were required as of the Fall 1987 semester to have taken a basic skills assessment test (ASSET), 60 percent of the Students served by the grant programs included in this study did not have basic skills assessment scores on file. Due to this limited percentage of basic skills assessments available, basic skills were not one of the variables used for matching students in the experimental and control groups. A significant difference in basic skill levels of control group students could be an explanation why control group students achieved significantly higher grade point averages during the Winter 1991 semester than experimental group students. Therefore, the assessment scores available for both experimental and control group students were obtained and a t-test for independent means statistical analysis con- ducted. The statistical results of the ASSET scores are in Tables 11-13. 90 Table 11.--ASSET Scores for Experimental and Control Group Students Writing Skills Subtest. Group N x SD SE 1 DF 2-Tail Prob. Experimental 80 39.03 6.026 0.674 _ -2.56 152 .011 Control 74 41.76 7.123 0.828 ‘ Students obtaining a scaled score of 35-39 in the Writing Skills subtest would be advised to enroll in an Applied English Skills (ENG 101), Business English (OSE 150), or a Writing Methods/College Composition l (ENG 111A) course. Students scoring in the 40-41 range are in the decision zone. They can enroll in either the courses recommended for students in the 35-39 range or they can enroll in College Composition l (ENG 111) or Business Communications l (OSE 151). The advise- ment in the decision zone is based on other factors, including student's performance in recent courses and life experiences. According to Delta College's Dean of Enroll- ment Services, students in the decision zone in the Writing Skills subtest tend to underestimate their skills and enroll in lower level English courses. Table 12.--ASSET Scores for Experimental and Control Group Students Reading Skills Subtest. Group N x SD SE t DF 2-Tail Prob. Experimental 80 41.08 6.425 0.747 -1.89 152 7 .061 Control 74 39.24 5.682 0.635 " Students obtaining a scaled score on the Reading Skills subtest in the 38-40 range are advised to enroll in the Effective Reading ll (ENG 109) course. A scaled score of 41-53 is an indication that students have the skills appropriate for college- Ievel courses. 91 Table 13.--ASSET Scores for Experimental and Control Group Students Numerical Skills Subtest. Group N x SD SE 1 DP 2-Tail Prob. Experimental 78 39.46 6.322 0.716 -1.22 152 .224 Control 76 40.68 6.093 0.699 *" Students were not required to take each subtest. Therefore, the N varies in Tables 11-13. ““ Students obtaining a scaled score of 36-39 are in the decision zone and are advised to enroll in courses based on recent course performance and life experi- ences. According to Delta College's Dean of Enrollment Services, students in the decision zone in the Numerical Skills subtest tend to overestimate their skills and enroll in higher level math courses. Students scoring in the 40-55 range would be advised to enroll in Algebra I (MTH 107) or Business Mathematics (GB 110). As Tables 11-13 indicate, experimental and control group students' scores on the Reading Skills and Math Skills subtests did not significantly differ. However, control group students scored significantly higher than experimental group students in the Writing Skills subtest. I It was hypothesized that due to the availability and use of additional academic support services during the Winter 1991 semester that experimental group students would perform academically equivalent to a random sample of 500 students in an institutional group. The data in Table 14 indicates that the mean Winter 1991 se- mester grade point average of institutional group students (2.68) is higher than experimental group students (2.23) who used supplemental academic support services sponsored by grant funds. Null Hypothesis: The mean grade point average during the Vlfinter 1991 semester for the experimental group will not statistically significantly (p > .05) exceed the institutional group students' mean grade point average. 92 Alternative Hypothesis: The mean grade point average during the Wlnter 1991 semester for the experimental group will be equal to or exceed the institutional group students' mean grade point average. Statistical Test of Significance: t-test for independent means; 1-tailed probability. Decision: the institutional group mean exceeded the experimental group mean; accept the null hypothesis. Reject the alternative hypothesis. Table 14.-Wlnter1991 Semester Mean Grade Point Average Experimental and Institutional Group Students. Group N x SD SE t DF 1-Tail Prob. Experimental 200 2.2298 1.054 0.075 -4.99 698 (.001' Institutional 500 2.6767 1 .077 0.048 ’ Institutional group mean exceeded experimental group mean. It was hypothesized that due to their characteristics being "matched" with Special population students in the experimental group who used supplemental aca- demic support services, that the Winter 1991 semester mean grade point average earned by control group students would be lower than the mean grade point average earned by the 500 students in the institutional group. The data in Table 15 indicates that there was no statistical difference in the mean Wlnter 1991 semester grade point average of control group students (2.72) as compared to the mean grade point average of institutional group students (2.68). Null Hypothesis: The institutional group students'WInter1991 semester mean grade point average will not statistically significantly exceed (p > .05) the \Mnter 1991 semester mean grade point average of control group students. Alternative Hypothesis: The institutional group students'WInter1991 semester mean grade point average will statistically significantly exceed the Wlnter 1991 semester mean grade point average of control group students. 93 Statistical Test of Significance: t-test for independent means; 1-tailed probability. Decision: p .313 > .05; accept the null hypothesis. Reject the alternative hypothesis. Table 15. «Wlnter 1991 Semester Mean Grade Point Average Control and Institu- tional Group Students. Group N Y SD SE t DF 1-Tall Prob. Control 200 2.7197 0.993 0.070 0.49 698 .31 3 Institutional 500 2 .6767 1 .077 0.048 Aeademleflersletence Wm It was hypothesized that due to receiving supplemental academic support services during the Winter 1991 semester that experimental group students would withdraw from less courses and complete a higher proportion of their courses during the Winter 1991 semester than their counterparts in the control group. The data in Table 16 indicates that control group students who did not receive supplemental academic support services withdrew from a lower proportion of courses and per- sisted to complete a higher proportion of the courses they enrolled in during the Winter 1991 semester as compared to experimental group students. Null Hypothesis: Experimental group students' proportion of Winter 1991 semester courses completed will not statistically significantly exceed (p > .05) the proportion of courses completed during the Wlnter 1991 semester by students in the control group. Alternative Hypothesis: The proportion of Wlnter 1991 semester courses completed by experimental group students will statistically Significantly exceed the proportion of courses completed by control group students. 94 Statistical Test of Significance: Chi-Square. Decision: control group students exceeded the experimental group; accept the null hypothesis. Reject the alternative hypothesis. Table 16. -Wlnter 1991 Semester Course Completion Rate for Experimental and Control Group Students. Group All Some None Row Total Experimental 157 28 15 200 50.0 Control 1 81 12 7 200 50.0 Column 338 40 22 400 Total 84.5 10.0 5.5 100.0 Chi-Square D.F. Significance 11.01323 2 0.0041 " ‘ Control group proportion exceeded experimental group. It was hypothesized that the experimental group students' completion rate of Wlnter 1991 semester courses would equal the completion rate of institutional group students. The data in Table 17 indicates that students in the institutional group completed a significantly higher proportion of the courses they enrolled in during the Winter 1991 semester than did experimental group students who accessed supple- mental academic support services. Null Hypothesis: The proportion of courses completed during the \Mnter 1991 semester by the institutional group will statistically significantly exceed (p > .05) the proportion of courses completed by the experimental group. 95 Alternative Hypothesis: The proportion of courses completed by the experimental group will statistically significantly exceed or be equal to the proportion of courses completed during the Wlnter 1991 semester by the institutional group. Statistical Test: Chi-Square. Decision: institutional group students exceeded the experimental group; accept the null hypothesis. Reject the alternative hypothesis. Table 17.-Vlnnter 1991 Semester Course Completion Rate for Experimental and Institutional Group Students. Group All Some None Row Total Experimental 1 57 28 1 5 200 28.6 Institutional 432 37 31 500 71 .4 Column 589 65 46 700 Total 84.1 9.3 6.6 100.0 Chi-Square D.F. Significance 8.12852 2 0.0172‘ " Institutional group proportion exceeded experimental group. It was hypothesized that institutional group students would complete a higher proportion of the courses they enrolled in during the Vlflnter 1991 semester as com- pared to control group students. The data in Table 18 indicates that there is no statistically significant difference in the proportion of courses completed during the Wlnter 1991 semester by institutional and control group students. Null Hypothesis: The proportion of courses completed by institutional group stu- dents during the Vlfinter 1991 semester will not statistically significantly exceed (p > .05) the proportion of courses completed during the Winter 1991 semester by control group students. 96 Alternative Hypothesis: The proportion of courses completed by institutional group students during the anter 1991 semester will statistically significantly exceed the proportion of courses completed during the WInter 1991 semester by control group students. Statistical Test of Significance: Chi-Square. Decision: p .2735 > .05; accept the null hypothesis. Reject the alternative hypoth- esis. Table 18.--\Mnter1991 Semester Course Completion Rate for Control and Institu- tional Group Students. Group All Some None Row Total Control 181 12 7 200 28.6 Institutional 432 37 31 500 71 .4 Column 613 49 38 700 Total 87.6 7.0 5.4 100.0 Chi-Square D.F. Significance 2.59265 2 0.2735 WWW It was hypothesized that a greater proportion of experimental group students would return to college and enroll in courses during the Fall 1991 semester than their counterparts in the control group. As the data in Table 19 indicates, there was no statistically significant difference in the enrollment of experimental group students and control group students in Fall 1991 semester courses. 97 Null Hypothesis: The Fall 1991 semester enrollment of experimental group students will not statistically significantly exceed (p > .05) the Fall 1991 semester enrollment of control group students. Alternative Hypothesis: The Fall 1991 semester enrollment of experimental group students will statistically significantly exceed the Fall 1991 semester enrollment of control group students. Statistical Test of Significance: Chi-Square. Decision: p .2625 > .05; accept the null hypothesis. Reject the alternative hypoth- esis. Table 19.--Fall 1991 Semester Enrollments Experimental and Control Group Students. Group Enrolled Did Not Enroll Row Totals Experimental 1 13 87 200 50.0 Control 125 75 200 50.0 Column 238 162 400 Total 59.5 40.5 100.0 Chi-Square D.F. Significance 1 .25532 1 0.2625 It was hypothesized that experimental group students would return to college and enroll in Fall 1991 semester courses at the same rate as institutional group students. Null Hypothesis: The enrollment of experimental group students in Fall 1991 semes- ter courses will not statistically significantly exceed (p > .05) the proportion of institu- tional group students. Alternative Hypothesis: The enrollment of experimental group students in Fall 1991 98 semester courses will be statistically Significantly equal to or exceed the proportion of institutional group students. Statistical Test of Significance: Chi-Square. Decision: p .3439 > .05; accept the null hypothesis. Reject the alternative hypothesis. Table 20.--Fall 1991 Semester Enrollments Experimental and Institutional Group Students. Group Enrolled Did Not Enroll Row Totals Experimental 1 13 87 200 28.6 Institutional 261 239 500 71 .4 Column 374 326 700 Total 53.4 46.6 100.0 Chi-Square D.F. Significance 0.89578 1 0.3439 It was hypothesized that institutional group students would return to college and enroll in Fall 1991 semester courses at a higher rate than control group stu- dents. As the data in Table 21 indicates, control group students returned to Delta College and enrolled in Fall .1991 semester courses at a higher proportion than students in the institutional group. Null Hypothesis: Fall 1991 semester enrollment by institutional group students will not statistically significantly (p > .05) exceed the enrollment of students in the control group. 99 Alternative Hypothesis: Fall 1991 semester enrollment by institutional group stu- dents will statistically significantly exceed students in the control group. Statistical Test of Significance: Chi-Square. Decision: control group students proportion greater than institutional group; accept the null hypothesis. Reject the alternative hypothesis. , Table 21 .--Fall 1991 Semester Enrollments Control and Institutional Group Students. Group Enrolled Did Not Enroll Row Totals Control 125 75 200 28.6 Institutional 261 239 500 71.4 Column 386 314 700 Total 55.1 44.9 100.0 Chi-Square D.F. Significance 5.71778 1 0.0168* * Control group students enrolled in the Fall 1991 semester at a greater proportion than institutional group students. QQSIADelxsle The primary purpose of this study was to determine if supplemental counsel- ing and tutorial academic support services provided through grant-sponsored pro- grams (Single Parent/Homemaker Program, Special Needs Program-disadvantaged component, Michigan Job Opportunity Bank-Retrain Program, and the At-Risk Student Success Program) were used by eligible community college students and if 100 these services improved special population students' academic achievement and persistence. A secondary focus of the study was to estimate the financial impact on a community college's budget as a result of providing academic support services to special population students. lnstltutlenalseunsellncfindces The institutional cost of providing counseling services to students was identi- fied by obtaining the 1990-91 fiscal year expenditures for Counseling Center ser- vices from the College's budget and dividing that amount by the number of academic credit hours for all institutional enrollments during the Fall 1990, Winter 1991, and Spring/Summer 1991 semesters to obtain a mean cost per academic credit hour of institutional counseling. The calculation determined the cost of providing institutional counseling at the mean rate of $3.90 per academic credit hour. As indicated in Table 22, experimental group students enrolled in a total of 1 ,669 academic credit hours during the \Mnter 1991 semester. At the mean rate of $3.90 per academic credit hour, the estimated cost of providing institutional counseling to experimental group students during the Winter 1991 semester was $6,509.10, a mean cost to the institution of $32.55 per student. 101 Table 22.--Estimated Cost of Providing Institutional Counseling Services. Group Academic Credit Estimated Cost of Mean Cost Hours Enrolled Providing Counseling Per Student During Winter During Winter 1991 Vlfinter 1991 1991 Semester Semester Semester Experimental 1,669 $6,509.10 ‘ $32.55 Control 1 .643 $6,407.70 $32.04 A similar calculation was conducted to determine the cost of providing institu- tional counseling services to control group students. Control group students enrolled in a total of 1,643 academic credit hours during the Winter 1991 semester. At the mean rate of $3.90 per academic credit hour, the total cost of providing institutional counseling to control group students during the Winter 1991 semester was $6,407.70, a mean cost to the institution of $32.04 per student. n i l The institutional cost of providing tutoring services to students was identified by obtaining the 1990-91 fiscal year expenditures for the Teaching/Learning Center from the College's budget and dividing that amount by the number of academic credit hours for all institutional enrollments during the Fall 1990, Winter 1991, and Spring/Summer 1991 semesters to obtain a mean cost per academic Credit hour of institutional tutoring. The calculation determined the cost of providing institutional tutoring at the mean rate of $839 per academic credit hour. As indicated in Table 23, at the rate of $839 per academic credit hour, the total cost of providing institu- tional tutoring to experimental group students during the Winter 1991 semester was $1 ,400.29, a mean cost to the institution of $7.00 per student. 102 Table 23.--Estimated Cost of Providing Institutional Tutoring Services. Group Academic Credit Estimated Cost of Mean Cost Hours Enrolled Providing Tutoring Per Student During Winter During Winter 1991 Wlnter 1991 1991 Semester Semester Semester Experimental 1,669 $1,400.29 $7.00 Control 1 .643 $1,378.48 $6.89 A similar calculation was conducted to determine the cost of providing institu- tional tutorial services to control group students. At the rate of $839 per academic credit hour, the total cost of providing institutional tutorial services to control group students during the Winter 1991 semester was $1 ,378.48. a mean cost of $6.89 per student. - r m In addition to accessing institutional counseling and tutoring services. experi- mental group students also accessed additional counseling and tutorial services sponsored through grants. The grant cost of providing counseling services to special population stu- dents was determined through the 1990-91 final grant reports documenting counsel- ing expenditures for the Single Parent/Homemaker Program. the Special Needs Program-disadvantaged component. the Michigan Job Opportunity Bank-Refrain Program, and the At-Risk Student Success Program. The total counseling expenditures for these grants were divided by the number of academic credit hours that grant-sponsored students enrolled in during the Fall 1990. Wlnter 1991. and Spring/Summer 1991 semesters to obtain a mean cost of grant-sponsored 103 counseling. The calculation determined the cost of providing grant-sponsored coun- seling at the mean rate of $30.68 per academic credit hour. At the mean rate of $30.68 per academic credit hour, the total cost of providing grant-sponsored coun- seling services to experimental group students during the Wlnter 1991 semester was $51 .205.00. a mean cost of $256.03 per student. Table 24.--Estimated Cost of Providing Grant-Sponsored Services to Special Popu- lation Students. Service Academic Credit Estimated Cost of Mean Cost Hours Enrolled Providing Academic Per Student During Winter Support Service During anter 1991 1991 Semester Wlnter1991 Semester Semester Counseling 1 ,669 $51 .205.00 $256.03 Tutoring 1 ,669 $20,261.66 $101 .31 Grant-sponsored tutoring services were calculated at the mean rate of $12.14 per academic credit hour (determined by the tutorial expenditures indicated on grant final reports for the 1990-91 academic year and divided by the number of academic credit hours grant-sponsored students enrolled in during the Fall 1990, Wlnter 1991, and Spring/Summer1991 semesters). The total cost of providing tutoring to experi- mental group students during the Wlnter 1991 semester was $20,261.66, a mean cost of $101.31 per student. War. As Table 25 indicates. the total mean cost of providing academic support services to experimental group students during the WInter 1991 semester was 104 $396.89 and to control group students, $38.93. Therefore, it cost an additional $357.96 to provide academic support services to each experimental group Student. Table 25.--Total Mean Student Cost of Academic Support Services Winter 1991 Semester. - Group Institutional Institutional Grant- Grant- Total Counseling Tutoring Sponsored Sponsored Counseling Tutoring Experimental $32.55 $7.00 $256.03 $101 .31 $396.89 Control $32.04 $6.89 0 0 $ 38.93 in addition to the issue of the cost of providing academic support services to special population students is the question of the funding source providing the ser- vice. As indicated, grant-sponsored students use both institutional and grant funded academic support services. The College was required to provide matching funds to obtain the grants. In reviewing the final 1990-91 fiscal year grant financial reports. the College contributed $64,750 in matching funds to support grant-sponsored counseling and tutorial academic support services. The institutional matching funds for counseling and tutorial academic support services were contributed to the projects as a total matching dollar amount. Certain amounts of matching funds are not designated by line-item specifically for counseling or tutoring. Therefore, to determine the amount of institutional matching funds for grant-sponsored academic support services, a calculation was conducted to determine the total institutional cost for supplemental academic support services by mean academic credit hour. The calculation was: 105 Mean institutional match Number of academic credit hours for grant- per academic credit sponsored students enrolled during the Fall 1990. hour Winter 1991, and Spring/Summer 1991 semesters The calculation was: 555.150 $6.76 per academic 9,574 credit hour Table 26 describes the estimated cost of providing academic support services to experimental and control group students during the Wlnter 1991 semester by funding SOUTCB. Table 26.--Academic Support Services Winter 1991 Semester By Group and Funding Source. Group Institutional Funding Institutional Funding Grant Funded Academic Support Grant Match Mean Academic Support Services Mean Per Student Services Mean Per Student Per Student Experimental $39.55 $56.41 $300.93 _ Control $38.93 0 0 As indicated, experimental group students required a Slightly higher financial commitment from the College to use institutional academic support services during the Winter 1991 semester than students in the control group. Due to their being no statistically significant difference In the institutional academic support services re- portedly used by experimental and control group students. college resources were requiredat the rate of an additional $.51 to serve each experimental group student accessing institutional counseling and tutoring during the “MM 1991 semester than their control group peers. However, to operate supplemental academic support 106 services through grant-funded programs. the College contributed an average of $56.41 for each special population student. Therefore. while it cost the College a total of $38.93 to provide academic support services to control group students. the College used an average of $95.96 in institutional resources to serve each special population student in the experimental group. Summarx The purpose of Chapter IV was twofold: to determine if eligible community college special population students used grant-sponsored academic support ser- vices and if these services improved their academic achievement and persistence. A secondary focus of the study was to estimate the financial impact to a community college's budget as a result of providing academic support services to special popu- lation students. In reviewing students' usage of academic support services, the following research hypotheses were established. related data analyzed, and statistical conclusions obtained: 1. The research hypothesis predicted that experimental group students would use less institutional. counseling services during the Winter 1991 semester than control group students due to their ability to access supplemental counseling through grant-funded resources. In fact, the research indicated that grant-sponsored students accessed institutional counseling services (average 1.24 hours per student) at a similar rate as students in the control group (average 1 hour per student). 107 2. The research hypothesis predicted that due to their ability to access supplemental grant-funded tutorial assistance, that experimental group students would access less institutional tutoring during the Wlnter 1991 semester than control group students. Similar to institutional counseling services. the research showed that experimental group students accessed a comparable amount of institutional tutoring at the Teaching/Learning Center (average 3.55 hours) as students in the control group (average 3.43 hours). 3. The research hypothesis predicted that due to their ability to access addi- tional grant-sponsored academic support services. that experimental group students would access a greater amount of combined (institutional and grant-sponsored) academic support services during the Winter 1991 semester as compared to control group students. There was no statistically significant difference in the total academic support services accessed. During the Wlnter 1991 semester. experimental group students used an average of 5.77 hours of academic support services and control group students used an average of 4.43 hours. 4. The research hypothesis predicted that as a result of obtaining additional assistance from grant-sponsored counseling and tutorial services, that-experimental group students would achieve a higher mean grade point average during the Winter 1991 semester than control group students. In fact, control group students who did not use the supplemental academic support services had a significantly higher mean grade point average (2.72) than experimental group students (2.23). 108 5. The research hypothesis estimated that experimental group students who received supplemental academic support services during the Vlfinter 1991 semester would achieve a mean grade point average equal to a group of 500 students repre- senting the institution as a whole. The institutional group mean grade point average (2.68) significantly exceeded the experimental group mean grade point average (2.23). 6. The research hypothesis was that the institutional group mean grade point average for the Winter 1991 semester would significantly exceed the mean grade point average of the control group. There was no statistically significant difference in the mean grade point average for the Winter 1991 semester earned by control group students (2.72) as compared to students in the institutional group (2.68). 7. The research hypothesis predicted that due to receiving supplemental academic support services that experimental group students onld complete a higher proportion of the classes they enrolled in during the Winter 1991 semester than control group students. However, control group students completed a higher proportion of courses during the Winter 1991 semester than experimental group students. 8. The research hypothesis estimated that experimental group students would persist to complete the same proportion of courses attempted during the Winter 1991 semester as students in the institutional group. Institutional group students completed a higher proportion of their Wlnter 1991 semester courses than. 109 did students in the experimental group. 9. The research hypothesis predicted that institutional group students would complete a higher proportion of the courses they enrolled in during the Wlnter 1991 semester as compared to students in the control group. It appears that institutional and control group students completed approximately the same proportion of courses during the Winter 1991 semester. 10. The research hypothesis was that experimental group students would persist and enroll in Fall 1991 semester courses at a higher rate than students in the control group. While control group students persisted and enrolled in Fall 1991 semester courses at a slightly higher rate than experimental group students. there was no Statistically significant difference. 11. The research hypothesis was that experimental groUp students would persist and enroll in Fall 1991 semester courses at the same proportion as students in the institutional group. Experimental group students enrolled in Fall 1991 semes- ter courses at approximately the same rate as institutional group students. 12. The research hypothesis was that institutional group students would persist and enroll in Fall 1991 semester courses at a higher proportion than control group students. Control group students returned and enrolled in Fall 1991 semester courses at a higher proportion than institutional group students. 110 There were no research hypotheses identified for the secondary focus of the study. estimating the financial impact to a community college's budget as a result of providing academic support services to special population students. Conclusions regarding the financial impact to a community college's budget and whether grant- sponsored academic support services impact on special population students' aca- demic achievement and persistence will be presented in Chapter V along with the implications for practice and for further research. CHAPTER V SUMMARY, CONCLUSIONS, AND IMPLICATIONS FOR PRACTICE AND FOR FURTHER RESEARCH Summanr The purpose of this study was to determine if eligible community college special population students used grant-sponsored academic support services and if these services improved their academic achievement and persistence. A secondary focus of the study was to estimate the financial impact on a community college's budget as a result of providing academic support services to special population students. The research methodology was a quasiexperimental design. The experimen- tal group was 200 randomly sampled special population students enrolled at Delta College during the Winter 1991 semester. The control group consisted of 200 stu- dents enrolled during the Winter 1991 semester with similar Characteristics (gender. ethnicity. age. enrollment status, cumulative credit hours, and program major) to experimental group students. An institutional group of 500 students enrolled during the Winter 1991 semester was randomly selected to, permit statistical comparisons to the overall Delta College student body. There was no attempt made to "match” the characteristics of institutional group students to either experimental or control group 111 112 students. Questionnaires were sent to experimental group students to determine their usage of institutional and grant-sponsored academic support services (counseling and tutoring). Simultaneously, questionnaires were distributed to control group students to determine their usage of institutional academic support services. Institu- tional group students were not surveyed. A total of 71 experimental group re- sponses (36 percent) and 104 control group responses (53 percent) were analyzed and included in the study. Concurrently, institutional records were examined to obtain experimental group. control group, and institutional group students' Winter 1991 semester grade point averages, the proportion of Winter 1991 semester courses completed, and their subsequent enrollment in Fall 1991 semester courses. The researcher com- pared the three groups to determine if differences existed between experimental and control group students. experimental and institutional group students. and control and institutional group students. To obtain information to conduct the cost analysis portion of the research, institutional budgets for the 1990-91 fiscal year were examined to determine the expenditures used to operate the Counseling Center and Teaching/Learning Center. Final grant reports for the 1990-91 fiscal year for the Single Parent/Homemaker Program, Special Needs Program-disadvantaged component, At-Risk Student Success Program. and Michigan Job Opportunity Bank-Refrain Program were exam- ined to identify grant expenditures. The institutional matching funds designated for 113 counseling and tutorial academic support services were also identified. The institu- tional resources, matching funds committed by the College and grant funds ex- pended to provide the academic support services were divided by the appropriate number of academic credit hours to obtain a mean cost per academic credit hour by funding source. The intent of this study was to determine if grant-sponsored academic sup- port services improved special population students' academic achievement and persistence and to identify the financial impact on a community college's budget for providing supplemental grant-sponsored services. In the study, a series of questions were posed including: 1. Do special population students use the supplemental (grant-sponsored) academic support services available? 2. Do special population students use institutional academic support services at the same rate as non-special population students? 3. Do grant-sponsored academic support services provided to Special popu- lation students supplement or replace institutionally funded academic support services? 4. When supplemental academic support services are available, do special population students academically achieve at a rate comparable to non-special 114 population students? 5. When supplemental academic support services are available. do special population students persist at a rate comparable to non-special population students? 6. What portion of supplemental academic support services costs are borne by the community college? 7. When grant-sponsored academic support services are provided to Special population students, what is the estimated impact on a community college's budget? The data gathered in this research study permits tentative answers to these quesfions. 1. Community colleges provide a variety of academic support services in an attempt to enhance the skills and address the deficiencies of an increasingly diverse student population. Federal and state legislatures have provided grant funding to provide supplemental academic support services to assist special pOpulation stu- dents to successfully achieve their educational goals. The data from this study indicates that the special population students enrolled during the Winter 1991 se- mester Significantly underutilized the grant-sponsored counseling and tutorial ser- vices available. a finding that confirms previous research. Special population Stu- dents reported using these services an average of 1 hour during the Wlnter 1991 SBITIBSIBI’. 115 2. It was anticipated that special population students would rely on grant services and access less institutional academic support services. The data from this study indicates that special population students used approximately the same amount of institutional counseling and tutorial services as control group students. 3. The literature indicates that community college students who received additional academic advising and tutoring had a higher level of academic achieve- ment and persistence. Therefore, it was anticipated that special population students who received supplemental academic support services would achieve a higher grade point average during the Winter 1991 semester than their peers in the control group. In fact, control group students who did not receive supplemental academic support services significantly outperformed the special population students academi- cally. 4. Assessment scores for basic skills in writing. reading, and numerical skills were obtained for experimental and control group students. Experimental and control group students had the same (no statistical difference) scores on the Read- ing Skills and Math Skills subtests. However, control group students scored signifi- cantly higher than experimental group students in the Writing Skills subtest. 5. The literature indicates that students who receive academic support ser- vices will See an improvement in their academic achievement and persistence during the semester that academic support services are provided. This study did not sup- port previous findings. Experimental group students did not persist as well as control 116 group Students during the \Mnter 1991 semester. Control group students persisted to successfully complete a higher proportion of their Wlnter 1991 semester courses. 6. It was anticipated that experimental group students would persist and enroll in the following (Fall 1991) semester courses at a higher rate than control group students. However. while a slightly higher proportion of control group students returned and enrolled in Fall 1991 semester courses than experimental group stu- dents, there was no statistically significant difference. 7. An institutional group of 500 students was randomly sampled to permit comparisons to the overall student body. It was anticipated that due to receiving supplemental academic support services that experimental group students would improve their academic achievement and persist at a level equivalent to institutional group students. As the results indicate. institutional group students significantly outperformed experimental group students during the Wlnter 1991 semester in grade point average and proportion of courses successfully completed. However. experi- mental group students enrolled in Fall 1991 semester courses at approximately the same rate (no statistically significant difference) as institutional students. 8. Eventhough their characteristics were ”matched” in six areas with experi- mental group students, control group students' academic achievement and persis- tence "mirrored” the performance of the institutional group students. In fact. there was no statistical difference in the mean grade point average achieved by control group students (2.72) during the Wlnter 1991 semester as compared to the mean 117 grade point average earned by institutional group students (2.68). Control group and institutional group students completed approximately the same proportion of Winter 1991 semester courses. and control group students enrolled in Fall 1991 semester courses at a statistically significantly higher rate thaninstitutional group students. 9. It was inferred that experimental group students would use less institu- tional academic support services. therefore reducing the institutional cost of provid- ing counseling and tutoring to special population students. However. there was no statistically significant difference in the use of institutional counseling and tutoring by experimental and control group students. 10. To obtain some of the grant funding. institutional matching funds were required. It appears from the data that institutional matching funds used to provide counseling and tutoring were required at almost twice the amount that it cost the institution to provide these same services through institutional resources. QQneluelene It appears that special population students did not use the supplemental academic support services available and as a result. grant-sponsored counseling and tutorial services were not maximized. Based on self-reported usage, grant resources were significantly underutilized. This does not mean the need for the services was not there. only that students reported they did not access the services. Special population students may not have been aware of the grant-sponsored 118 resources available and relied on the Counseling Center and Teaching/Learning Center staff for assistance. Special population students are able to determine on an individual basis. or refer themselves to the academic support services they prefer. Special population students ”voted with their feet” and obtained assistance from institutional staff and resources rather than those provided through grant funding. As a result. institutional resources which are already near or at capacity were utilized, while grant-sponsored resources were underutilized. While the writing skills of control group students were significantly higher than experimental group students, due to experimental and control group students being "matched" on six additional characteristics and with no statistical difference in two other ASSET subtests. this factor alone is not likely by itself to explain the higher grade point averages earned by control group students during the Wlnter 1991 semester. It appears that the supplemental academic support services accessed did not improve special population students' academic achievement. However. it should be noted that a pre-test was not conducted. Therefore. the researcher is unable to draw specific conclusions on how well a special population student would have academically achieved if supplemental academic support services had not been provided. Some special population students' VWnter 1991 semester grade point averages may have been even lower without the supplemental academic support services. One rationale for not enrolling in the following semester (Fall 1991) courses is if the student graduated upon completion of either the Vlfinter 1991 or 119 Spring/Summer 1991 semesters. The researcher reviewed student transcripts to determine the graduation status of both experimental and control group students. A total of 15 (7.5 percent) experimental group students and 13 (6.5 percent) control group students graduated between the Winter 1991 and Fall 1991 semesters. This differential would not explain the slightly lower proportion of experimental group students enrolling in Fall 1991 semester courses. A second rationale that was considered is that community college students often do not seek an Associates degree or certificate. but enroll in specific courses as life-long learners to upgrade specific skills. As a benchmark, student transcripts were reviewed to determine if experimental and control group students had previously earned a degree or certifi- cate at the College. Four (2 percent) experimental group students and 10 (5 per- cent) control group students had earned degrees prior to the Winter 1991 semester. In considering the data from the study and in reviewing additional institutional data, it appears that the supplemental academic support services accessed did not assist Special population students' persistence. and degree completion was not a determin- ing factor. Experimental group students' slightly higher (but not significant) Fall 1991 semester enrollment as compared to institutional group students can be partially explained by examining graduation records. Institutional group students graduated at a higher rate (9.8 percent) between the Winter 1991 and Fall 1991 semesters than did experimental group students (7.5 percent) and had completed an Associate degree or certificate prior to the Winter 1991 semester at a higher rate (5.2 percent) than experimental group students (2 percent). Therefore. a higher proportion of 120 institutional group students may have enrolled to upgrade specific skills and did not plan on attending Fall 1991 semester courses. It appears that the grant-sponsored services accessed did not assist special population students to improve their aca- demic achievement and persistence to the institutional average. The differential in enrollment in Fall 1991 semester courses between control group and institutional group students can be partially explained by reviewing gradu- ation records. Between the Winter1991 and Fall 1991 semesters. 9.8 percent of institutional students graduated as compared to 6.5 percent of control group stu- dents. Therefore. it was expected that control group students would enroll in Fall 1991 semester courses at a higher proportion than institutional group students. Based on this data, it appears that control group students who did not access supplemental academic support services equaled the average institutional student. With limited financial resources available, community colleges need. to pro- vide academic support services to special population students while minimizing the financial impact to the institution as much as possible. Since experimental group students used institutional counseling and tutoring at approximately the same rate as control group students. it appears that no cost savings was incurred by the College. As increasing numbers of special population students enroll at community colleges. it will be critical to evaluate the resources necessary to serve these students and assign the resources to the appropriate departmental budget. Community colleges may be assuming significantly higher costs for enrolling and serving increasing numbers of special population students than necessary. 121 Due to the relatively low utilization of supplemental academic support services by special population students. the per student cost of providing academic support services to experimental group students greatly exceeds the institutional cost of providing these services. Based on the academic achievement and persis- tence of experimental group students, it appears that significant grant funds are being expended to obtain minimal benefit in improving special population students' grade point average and persistence as compared to a comparable control group of students. It should be noted that two factors may have caused the grant-sponsored academic support services to reflect a higher cost per student than the institutional resources. F irst. a tutoring lab was established for special population students and a considerable amount of computer hardware and software was purchased. This was a one-time expenditure that would have inflated the cost of providing tutorial services to special population students during the 1990-91 academic year. Second, the Michigan Job Opportunity Bank-Retrain Program was being phased out by the funding agency. so student enrollments were limited. Therefore, the average cost of providing academic support services to special population students may have been slightly higher than normal. This study resulted in some adjustments in perceptions concerning the use. benefit, and financial implications associated with providing grant-sponsored aca- demic support services to special population students. Based on the data. it ap- pears that future legislation should encompass a variety of factors, including a quan- titative and qualitative evaluation of student outcomes. Special population students enroll in community colleges for a variety of reasons, one of which is their open-entry 122 admission policy. As community colleges enroll special population students, an assessment should be conducted to identify any deficiencies that the student may possess that will be a barrier to achieving their educational goals. Community colleges have a responsibility if they accept special population students to provide them with the academic support services to assist them to academically achieve and persist to complete their educational objectives. Community colleges that operate academic support services and the federal and state funding agencies that finan- cially support these programs have a responsibility to evaluate these services both in terms of student outcome measures and the financial impact of providing these services to assure that maximum student benefit (in this study defined as academic achievement and persistence) is obtained at the lowest cost possible. ImollcejlensJeLEractice 1. The numbers of special population students enrolling in postsecondary education institutions, particularly community colleges, is expected to significantly increase. Community colleges enrolling special population students should provide the academic support services to assist those students to achieve their educational goals and objectives. Community college administrators, faculty, and staff should quantitatively and qualitatively evaluate the academic support services currently available in their institution to ensure that they are positively impacting upon. and meeting the needs of their students. 2. Federal and State funding agencies currently request limited data on the special population students receiving grant-sponsored services. Granting agencies 123 should expand their current reporting requirements (headcount of students served. courses/programs enrolled. services provided, etc.) to include a quantitative evalua- tion of student performance outcomes. Community colleges should be accountable for the grant funding expended and assure that the services provided assist special population students to achieve their academic goals. 3. Several granting agencies currently provide funding to community colleges based on a formula (such as number of students enrolled, number of financial aid recipients, etc.) that does not incorporate any quantitative evaluation of the grant- sponsored services provided at that particular institution or the corresponding stu- dent outcomes. Granting agencies should place quantitative and qualitative perfor- mance measures as a criteria and focus on student outcomes in their application process or funding formula to reinforce the importance of improving special popula- tion students' academic achievement and persistence as a major program goal. 4. Federal funding agency staff should review the eligibility criteria of their specific programs to assure that grant-sponsored academic support services are provided to the special population students who require those services to be aca- demically successful. 5. Academic support services should be evaluated on an on-going basis. Community college staff should review student utilization of the academic support services available to determine if students are utilizing existing institutional and grant-sponsored resources. To identify a trend. this research should be conducted 124 or information gathered over several academic semesters. Community college staff should determine the rationale of why students access specific resources, students' perceptions of the quality of assistance provided, and based on the evaluation. improve the academic support services. If necessary. existing resources should be reallocated to meet the needs of the special population students enrolled at that institution. 6. If grant-sponsored supplemental academic support services are underutilized at an institution, and grant-sponsored students access institutional resources to a greater extent, community college administrators should examine the internal allocation of existing resources and redirect resources to assure that special population students are adequately served and that personnel are sufficient to meet their needs. 7. As additional academic support service initiatives are developed and initiated at the college, they should be in response to an identified need or deficiency that can be quantitatively documented. An evaluation should be conducted to deter- mine if these resources are used by the students they were intended to serve, and if these academic support services positively impact on students' academic achieve- ment and persistence. lmeltcatteneteLEuLtneLBeseaLcn As with most research studies. the data collected in this study have raised additional questions that other researchers may want to pursue. 125 1. This research was conducted on students who participated in four grant- sponsored programs (Single Parent/Homemaker Program, Special Needs Program- disadvantaged component. At-Risk Student Success Program, and Michigan Job Opportunity Bank-Retrain Program). Once the students were randomly selected and placed in the experimental group. they became ”generic", untraceable to a specific grant program. This was in no way an evaluation of any one grant funded project, but rather a composite of four programs. Based on the results of this study. grant programs should be reviewed individually to evaluate if the academic support ser- vices provided by a specific program positively impact on the academic achievement and persistence of the special population students served by that program. 2. This research was conducted on special population students enrolled during the Winter 1991 semester and followed their progress through their enroll- ment in Fall 1991 semester courses. Additional research should be conducted to study special population students' academic achievement and persistence over a lengthier timespan to allow for the inclusion of long-term academic Skill development and to permit the inclusion of community college students who temporarily stop-out or transfer to another postsecondary institution. 3. This research was conducted at a single community college. The study should be replicated at other community colleges (and perhaps four-year institutions) to determine if grant-sponsored supplemental academic support services positively impact on special population students' academic achievement and persistence in other postsecondary environments. 126 There are many other factors that could have impacted on the results of this study. The literature has outlined in considerable detail how special population students' academic achievement and persistence is improved by accessing supple- mental academic support services. This study is limited in that it was conducted at a single institution. and the results should not be generalized to other community colleges without further research. In addition, part of the study was conducted on self-reported data. While the survey return rate of 45 percent was acceptable and typical for this type of research. the utilization of academic support services of re- spondents Should not be inferred to reflect the usage of non-respondents. Additional research is recommended to evaluate the impact of grant-sponsored academic support services on special population students' academic achievement and persis- tence. and to determine the financial impact on community colleges' budgets as a result of providing these services. APPENDICES 127 APPENDIX A QUESTIONNAIRE AND LETTER SENT TO EXPERIMENTAL GROUP STUDENTS 128 129 / ’/ 47]., ) 2,31, q. o: 3. c . /” _ . .323:t ”o 112% (E . 1°, c. 9. DELTA COLLEGE be one [fie Oppor/um'l] November 1991 Dear Delta College Student: Delta College offers a variety of counseling and tutorial services to assist students to successfully achieve their academic goals. To determine if these services are being used by students and meet their academic needs. we are asking a sample of students enrolled during the Winter 1991 semester to participate in a survey. Results will be used to evaluate and as appropriate, improve services for current and future students. The survey information will also be included in a dissertation being completed at Michigan State University. A requirement of Michigan State University is that your participation is voluntary. You indicate your voluntary agreement to participate by completing and returning this questionnaire. All individual survey results will be confidential. Please take a few minutes to complete the survey and return it in the enclosed envelope. Thank you for your cooperation. If you have any questions regarding the survey, please contact me at 686-9218. Sincerely, Pat Graves Development Associate Enclosures UNIVERSITY CENTER 0 MICHIGAN 48710 0 PHONE (517)686-9000 0 FAX (517) 686-8736 130 Delta College's Counseling Center staff (located in the K-Wing) provide academic advising, career and personal counseling, and course scheduling services to students. During the Winter 1991 semester (January 3 - April 24, 1991), did you use any Counseling Center services? ___ 1. Yes ___ 2. No A1. If yes, please provide your best estimate of the number of hours you met with Counseling Center staff during the Winter 1991 semester: hours A2. If yes, were the services you received from Counseling Center staff beneficial in assisting you to meet your academic objectives? Very beneficial . Somewhat beneficial . Beneficial . Of little benefit . No benefit mwa-b Other office staff provide academic advising, career and personal counseling, and course scheduling services to students. During the Winter 1991 semester, did you use any of the following services? If yes, please provide your best estfllate of the number of hours you met with office staff during the Winter 1991 semester: 1. Student Development Center (located in B-116)...................... ' ' ' I I I—l 2. Single Parent/Sex Equity Office (located in Admissions) ............ :___J___J :____: 3. MJOB—Retrain Office (located in the lower West Concourse).... ...... :___j___: :____: 4. Student Development Center (located at the Ricker Annex)..... ...... :___:___: L___J 5. Other (please specify) -- {___I___I I____I B1. If yes, were the services you received beneficial in assisting you to meet your academic objectives? Very beneficial Somewhat beneficial . Beneficial . Of little benefit . No benefit (”50319-5 I I Delta College’s Teaching/Learning Center (located in the lower East Concourse) provides students with tutoring, study skills workshops, and test taking services. During the Winter 1991 semester, did you use the Teaching/Learning Center? ___ 1. Yes ___ 2. No Ci. If yes, please provide your best estfleate of the total number of hours you used the Teaching/Learning Center during the Winter 1991 semester: hours 02. If yes, were the services you received from the Teaching/Learning Center beneficial in assisting you to meet your academic objectives? ___ 1. Very beneficial 2. Somewhat beneficial 3. Beneficial 4. Of little benefit 5. No benefit 131 D. Tutoring is also provided to students at the Student Development Center (located in B-116) and at the Student Development Center located at the Ricker Annex. Did you use either of these tutoring services during the Winter 1991 semester? ___ 1. Yes ___ 2. No Di. If yes, please provide your best estfllate of the total number of hours you used the Student Development Center for tutoring during the Winter 1991 semester: hours 02. If yes, were the tutoring services you received from the Student Development Center beneficial in assisting you to meet your academic objectives? Very beneficial Somewhat beneficial Beneficial Of little benefit No benefit (”boom—e Other comments: THANK YOU FOR YOUR PARTICIPATION! Please return this questionnaire in the enclosed envelope to: Delta College Office of Research and Development University Center, WI 48710 10/91 rd: q002 APPENDIX B QUESTIONNAIRE SENT TO CONTROL GROUP STUDENTS 132 133 Delta College’s Counseling Center staff (located in the K-Wing) provide academic advising, career and personal counseling, and course scheduling services to students. During the Winter 1991 semester (January 3 - April 24, 1991), did you use any Counseling Center services? ___ 1. Yes ___ 2. No A1. If yes, please provide your best estimate of the number of hours you met with Counseling Center staff during the Winter 1991 semester: hours A2. If yes, were the services you received from Counseling Center staff beneficial in assisting you to meet your academic objectives? Very beneficial Somewhat beneficial Beneficial Of little benefit No benefit (flhOON—b Delta College’s Teaching/Learning Center (located in the lower East Concourse) provides students with tutoring, study skills workshops, and test taking services. During the Winter 1991 semester, did you use the Teaching/Learning Center? ___ 1. Yes ___ 2. No Bi. If yes, please provide your best estimate of the total number of hours you used the Teaching/Learning Center during the Winter 1991 semester: hours 82. If yes, were the services you received from the Teaching/Learning Center beneficial in assisting you to meet your academic objectives? . Very beneficial . Somewhat beneficial . Beneficial . Of little benefit . No benefit macaw-s 134 Other comments: THANK YOU FOR YOUR PARTICIPATION! Please return this questionnaire in the enclosed envelope to: Delta College Office of Research and Development University Center, MI 48710 10/91 rd: qOO2 APPENDIX C FOLLOW-UP CORRESPONDENCE SENT TO EXPERIMENTAL AND CONTROL GROUP STUDENTS 135 136 '0°o'o" / 0 (42:32:13., [DELJ7\13()LLEK3E awzaneAKE(2qpanhuu@y December 2, 1991 Dear Delta College Student: Approximately three weeks ago, we contacted you regarding a survey Delta College is conducting to determine the use and effectiveness of counseling and tutorial services. As you may recall, the results of the survey will be used to evaluate and as appropriate, improve services for current and future students. If you have already completed and returned a questionnaire, we would like to thank you for your cooperation. However, if you have not yet completed a questionnaire, we still need your help. Please take a few minutes to complete and return your questionnaire as soon as possible. All individual survey results will be confidential. We appreciate your assistance in our effort to provide the best possible services for our students. If you have any questions, please contact me at 686-9218. Sincerely, Pat Graves Development Associate UNIVERSITY CENTER 0 MICHIGAN 48710 O PHONE (517)686-9000 0 FAX (517) 686-8736 137 // @‘Sta . M M ”N O DELTA COLLEGE (We am 1.42 Upper/anti} December 18, 1991 Dear Delta College Student: Approximately a month ago we contacted you regarding a survey we are conducting to determine the use and effectiveness of counseling and tutorial services. As you may recall, the results of the survey will be used to evaluate and as appropriate, improve services for current and future students. If you have already completed and returned a questionnaire, please disregard the one we have enclosed. However, if you have not yet completed a questionnaire, we still need you help. Please take a few minutes to complete and return the enclosed questionnaire in the stamped return envelope as soon as possible. All individual survey results will be confidential. We appreciate your assistance in our effort to provide the best possible services for our students. If you have any questions, please contact me at 686-9218. Sincerely, Pat Graves Development Associate Enclosures UNIVERSITY CENTER 0 MICHIGAN 48710 O PHONE (517) 686-9000 0 FAX (517)686-8736 APPENDIX D PERMISSION LETTER FROM UNIVERSITY COMMITTEE ON RESEARCH INVOLVING HUMAN SUBJECTS 138 139 MICHIGAN STATE UNIVERSITY OFFICE Of VICE PRESIDENT POR RESEARCH EAST LANSING 0 MICHIGAN 0 48824-1046 AND DEAN OF THE GRADUATE SCHOOL November 8, 1991 Patricia Graves 5971 Weiss Apt. 0-8 Saginaw, MI 48603 RE: GRANT-SPONSORED ACADEMIC SUPPORT SERVICES: DO THEY IMPROVE COMMUNITY COLLEGE SPECIAL POPULATION STUDENTS' ACADEMIC ACHIEVEMENT AND PERSISTENCE7, IRB #91-477 Dear Ms. Graves: The above project is exempt from full UCRIHS review. II have reviewed the proposed research protocol and find that the rights and welfare of human subjects appear to be protected. You have approval to conduct the research. You are reminded that UCRIHS approval is valid for one calendar year. If you plan to continue this project beyond one year, please make provisions for obtaining appropriate UCRIHS approval one month prior to October 23, 1992. Any changes in procedures involving human subjects must be reviewed by the UCRIHS prior to initiation of the change. UCRIHS must also be notified promptly of any problems (unexpected side effects, complaints, etc.) involving human subjects during the course of the work. Thank.you for bringing this project to our attention” If we can be of any future help, please do not hesitate to let us know. ely, David E. Wright, Ph.D; Chair, UCRIHS DEW/deo cc: Dr. Eldon Nonnamaker MSU it an Affimatiw Action/Equal Opportunity Institution APPENDIX E RESPONSES TO OPEN-ENDED QUESTIONS 140 141 RESPONSES OF EXPERIMENTAL GROUP STUDENTS TO QUESTION OF WOULD YOU LIKE TO PROVIDE ADDITIONAL COMMENTS They should have more night hours and help provide more student financial services for students seeking financial aid help. I asked for a tutor, but was refused one. It was nice to know that there were student aids to help me with my problems in certain classes I had. I am not on campus that much. I have not used the tutoring services because I am doing well in all of my classes. Haven't used it that much because of the classes I've taken, but I know I'll need it more later on. I realize there is help available, but I've never used any of it except for the counsel- ing center in the K-wing and that was only for a half an hour. I've currently used the same center once again for about the same time and it was very beneficial. I find that a great deal of my time is spent in the Library or in the Career Planning Center. Single Parent grant was very helpful in helping me pay for things that my Pell grant didn't pay for. Some of the counselors are unable to give you a good idea of how you should spend your time on classes geared toward your degree. What might be most beneficial to your individual needs. But they try. There is a "tremendous need” for BIO 131-203 tutors and also Math 120, Nursing/ Nursing 100 and more qualified people are needed. I feel there should be more classes than what they have or more teachers to teach these classes! 142 Dear Pat, This is just a note of concern to let you know, lam no longer a Delta College student. I graduated in April, 1991. I currently attend SVSU as a full time Music Education major. ldo not wish to be bothered anymore with this counseling questionnaire. Please understand, it's nothing personal, and if it didn't pertain to counseling I wouldn't have a problem with it. However, the first year and a half I attended Delta, my counselor was H . That particular part of my college career was a nightmare. Not only was I academically counseled wrong. But, on a personal level towards me, H was very rude and unprofessional. l have never had anyone make me feel so down on myself in my life! Thank God, S was there to get my academic future on the right track again. I do not wish to pursue this matter any further. It happened in the past, and that's where I would like to leave it. Many faculty and staff members were aware of the problem I had with H at that time. Also, ldiscussed this matter with the chairman of the Counseling Depart- ment. I believe I have done all I can do concerning this matter; it is no longer in my hands. As you can tell, this matter is very upsetting to me. I only hope no other student goes through the grief l have been through. Regretfully so, I will never recommend H for anything. Sincerely, P 143 RESPONSES OF CONTROL GROUP STUDENTS TO QUESTION OF WOULD YOU LIKE TO PROVIDE ADDITIONAL COMMENTS The Nursing program has 19.0 many ”costs” in it. Someone needs to evaluate the system to up some changing to meet the needs of the people who need to get a career as soon as possible. I think that in some of the heavily taken classes (such as Health oriented courses) there should be more sections offered. Specifically: ST 100. I used the math lab every day for about an hour before class. I found the counselors very helpful to help me plan for my education; however, it was sometimes difficult for me to make an appointment at a convenient time for me. I work full-time, as I did then, and usually had to take time off from work to get an appointment. What all of this means is-l would like to see more evening appoint- ments available for students taking night classes only. Would be beneficial to have a computer lab/learner center with qualified and know); edgeable tutors staffing it. Please make counselors more available to talk to those of us who work full time day jobs and are only able to seek counseling after 5 pm. I recommend E . She is excellent as a counselor and a very genuine person. She even took me to lunch one day! I audited Dr. T Art History and Appreciation Course #152. It was my only class. I am 64 years old and a retired elementary teacher. I took the class to broaden my art interest. I wish you well in your dissertation. M and L help worth using. are excellent with help. They always are positive and offer I'm on the wait list for LPN program. I wish all the departments involved would get their information straight. I've heard different info from all departments. As far as the financial aid dept goes, they need to get rid of everyone in that office (they are so rude and know nothing). 144 Your outline of courses needed to complete classes to obtain an Associate's is pretty much self explanatory. That is why I did not see a counselor as much. I did not use the Teaching Learning Center, but lwould not want it to be discontin- ued because someday I will probably need it. The Counseling office needs to be updated on transfer info. I was given the wrong information on classes that would transfer. Thank goodness I found out that they would not transfer as the counselor told me, before it was too late. Math is a real struggle for me. I don't know what I would have done without the extra help. The Counseling Center helped me line up all my classes so ldidn't waste any time or money toward my degree. Elaasa keep these W. My counselor H is the best. Yes. a grade of a C- should be acceptable. Not just a flat C. Have used counseling service, but not winter semester. They were very helpful. Some of the macro sheets seem vague or unclear on which classes should be taken for the college you are going to transfer to. The Teaching Learning Center is too small they need more tutors to help people with questions. Need tutor at one table not helping other tables. Single individual help. The staff and most of the counselors in the counseling center treated me very kindly and courteously. They really want to help you. I feel I must state, however. that Mr. H , counselor, told me during orientation that I really didn't need to worry too much about college and a career as I had a husband to take care of me. I see another counselor. NO The staff at the loan center were very helpful with helping me pick the right classes and loan I needed. BIBLIOGRAPHY 145 BIBLIOGRAPHY Abrams, H. G., and Jernigan, L. P. ”Academic Support Services and the Success of High- -Risk College Students." W 21 (Summer 1984): 261-274. Aitken, N. D. ”College Student Performance, Satisfaction, and Retention. Journamf HigbsLEducaticnsa (1982): 32- 50. Baldridge, V. Kemerer, F. and Green K. ”Enrollments in the Eighties; Factors, Actors, and Impacts. " HIgbaLEducatbnflssaamhfiapQHNm Washington, D. 0.: American Association for Higher Education, 1982. Baumgart, N. L., and Johnstone, J. N. "Attrition at an Australian University." Journal QLfligbaLEducaticn 48 (1977): 553 570 Beal, Phillip, and Pascarella, ErnestT. "Designing Retention Interventions and Verifying Their Effectiveness." W San Francisco: Jossey- -,Bass (December 1982): 73-88. Bean, John P. ”Dropouts and Turnover. The Synthesis and Test of a Causal Model of Student Attrition." Wm 12 (1980): 155-187. Bean, John P. ”Conceptual Models of Student Attrition: How Theory Can Help the Institutional Researcher.” Wren. San Francisco: Jossey-Bass, (December 1982): 17- 33. Bean, John P., and Metzner, BarbaraS. ”A Conceptual Model of NontradItIonal Student AttrItIon BaxiamaLEducaticnalflassambss (1985): 485-540. Billson, Janet M., and Terry, Margaret 8. "A Student Retention Model for Higher Education.” Qnuegaandunjyarsjty 62 (Summer 1987): 290-305. Bishop, J.B., and Breneman, K.A. ”An Initial Assessment of a Counseling Center's Role in Retention.” JanrnaLnfflflagasmdantfiarsnnnejfl (1986): 461-462. Bishop, J.B., and Walker, S.K. ”What Role Does Counseling Play in Decisions Relating to Retention?” JoumalaLCoflacaSiudanLszalcpmem 31 (January 1990): 88-89. 146 147 Bock, Daniel E. and Sullins, W. Robert. "The Search for Alternative Funding: Com- munity Colleges and Private Fund- -Raising." Qnmmunitygoflegaflaxiaw 15 (W nter 1987):13-20. Bogus, Jesse Parker. W. New York: McGraw-Hill Book Company, lnc., 1950. Breneman, DavidW., and Nelson, Susan C. EinanongficmmunmtcclleceaLAn W. Washington, D..:C The Brookings Institution, 1981. Brint, Steven, and Karabel, Jerome. W . New York: Oxford University Press, 1989. Cohen, Arthur M., and Brawer, Florence B. W- San Francisco, Washington, London: Jossey-Bass Publishers, 1984. Cope Fl. and Hannah. W W W New York: Wiley. Cross, K. Patricia. Beynnmnaflpanflnnr. San Francisco: Jossey-Bass, 1971. W. Lansing, Michigan: Michigan Department of Educafion,1991. Drew, Claudine P. "We Can No Longer Love 'Em and Leave 'Em: A Paper on Freshman College Retention." W 17 (Spring 1990): 54- 60. W. Washington DC: United States Government Printing Office, October 11, 1991. Forrest, A. "Increasing Student Competence and Persistence: the Best Case for General Education." Iowa City, Iowa: American College Testing Program National Center for Advancement of Educational Practices, 1982. Fox, Richard N. "Application of a Conceptual Model of College \Mthdrawal to Disadvantaged Students." AmericamEducaflmBasearchcumal 23 (Fall 1986): 415-424. Friedlander, J. "Are College Support Programs and Services Reaching High Risk Students?" JeumaLoiCollecefludenLEeramnal 21 (1980): 23-28. 148 Giddan, Norman 8., and Weiss, Steven J. "Costs and Effectiveness of Counseling Center Drapom Prevention." mm 31 (March 1990): 100-107. Glass, Gene V. and Hopkins, Kenneth D. StatisticaIMothodsjoEduoationand Psychology. 2nd ed. Englewood Cliffs, New Jersey: Prentice- Hall lnc., 1984. Glennen, RE, and Baxley, D.M. "Reduction of Attrition Through Intrusive Advising." NASEAJooLnal 22 (Winter 1985): 10-14. Gold, L.N. "Educating At-Risk Adults: Success Means Breaking Down Federal Policy Barriers." QhangaUuly/August 1990): 6-7. Goldbold, D. H. "The Future of Community Colleges and Their Products." In Ad; 1W pp 145155 Edited by C.J. Smith, Washington D.C.: Institute for the Study of Educational Policy, 1978. Halpin, Richard L. "An Application of the Tinto Model to the Analysis of Freshman Persistence In a Community College." QommonihLQollogofiayiayy 17 (Spring 1990): 22- 32. Harrar, William R., and Ender. Steven C. "Assessing Students' Perceptions of Tutorial Services." JournaLoLCallogonndonLEarsonnal 28 (May 1987): 276-277. Hartle, Terry. "Federal Support for Higher Education in the ‘905 Boom, Bust or Something in Between." Qhanga(January/February1990): 32-41. Hartley, Maurice P. "H.E.L.P. for Students: One University's Action Approach to Increasing Student Retention." Qollogoandllnjmsity (Fall 1987): 80-94. Heath, Peter L., Skok, Richard L., and McLaughin, T.F. "A Proposed Community College Student Retention Model." Wad 25 (March 1991): 506-514. Hodgkinson, Harold L. "Guess Who's Coming to College." Aoadomo 69 (March/ April 1983): 13-20.- House, J. Daniel., and Wohlt, Victoria. "The Effect of Tutoring Program Participation on the Performance of Academically Underprepared College Freshmen." JaumaLQICQlleceStudemDeBlepmemSI (Juli/1990): 365-370- 149 Hudesman, John. "Locus of Control and Allocation Phase: Impact on the Academic Achievement of Special Program Students Over Three Semesters." Collage W19 (Spring 1985): 23-29. Hudesman, John. "Counseling Style: Its Impact on the Academic Performance of College Students In Special Programs." WNW I191 27 (May 1986): 394399. Hudesman, John., Avramides, Basil., Loveday, Charles, Wendell, Anne-Sojourner, and Griemsmann, Russell. "Counseling Style: Its Impact on the Academic Performance of College Students in Special Programs." Journalolfiolleoe Studentflarsonnol 27 (May 1987): 250-254. Kiesler, CA, and Morton, T.L. "Psychology and Public Policy in the Health Care Revolution." W 43: 993-1003. Knapp. JR, and Karabenick, S.A. "Incidence of Formal and Informal Academic Help-Seeking in Higher Education." JournaLonollagoSmdanLDoyalgomom 29 (May 1988): 223-227. Kowalski, 0. W" New York: Philosophical Library, 1977. Landward, Stayner., and Hepworth, Dean. "Support Systems for High Risk College Students: Findings and Issues." W (Winter 1984): 119-128. Lenning, Oscar T. "Variable Selection and Measurement Concerns." NmQILeo tionoiQLJnsmutionaLBasoaLch. San Francisco. Jossey-Bass, (December 1982): 35— 53. Lenning, Oscar T., Beal, Phillip, and Sauer, K. "Retention and Attrition: Evidence for Action and Research." Boulder, Colorado: National Center for Higher Education Management Systems. Levin, Henry M. "Cost- Effectiveness and Educational Policy." EducationaLExalua tionandfiolimLAnalBis100988) 51-69 Lewicki, G.J., and Thompson, D.L. "Awareness, Utilization, and Satisfaction With Student Services Among Freshmen and Sophomores: a Consumer Evalua- tion." JoumaLcLCoJlegeSiudentEarsormaIZS (1982): 477-481- Melia, Patrick., and Goodman, Marshall. "Survival by Controlling Demographic Impacts." Wm" 1988): 51-63. 150 Metzner, Barbara S. "Perceived Quality of Academic Advising: The Effect on Freshman Attrition" Amancanfiducatmrifleseambiloumal 26 (Fall 1989): 422-42. Miller, Maria. "A Positive Approach to Student Retention. the Academic Advising, Intervention and Monitoring System." NAOADAdournal (October 1985): 19— 24. Munro, Barbara H. "Dropouts From Higher Education: Path Analysis of a National Sample" AmericaaEducationalfieseathoumal 18 (1981)= 133141 Nelson, R.B., Scott, TB, and Bryan, W.A. "Precollege Characteristics: Early Col- lege Experiences as Predictors of Freshman Year Persistence." JournaLof W25 (1984): 50-54- O'Bannion, Terry. "Student Development Philosophy: A Perspective on the Past and Future" WWW (Fall 1989) 5- 17 Pantages, TimothyJ., and Creedon, Carol F. "Studies of College Attrition: 1950- 1975." BezimoLEducationalfleseamhw(VWnter1987li 49- 101 Parnell. Dale. W Washington. DC: The Community College Press AACJC, 1990. Pascarella, Ernest T., and Chapman, David W. "A Multiinstitutional, Path Analytic Validation of Tinto' 5 Model of College WIthdrawaI. " Amon’oanfiducational Bosoarohdournal 20 (Spring 1983): 87—102. Pascarella, Ernest T., Duby, Paul 8., and lverson, Barbara K. "A Test and Reconceptualization of a Theoretical Model of College WIthdrawaI in a Commuter Institution Setting." SociologyoLEdnoation 56 (1983): 88-100. Pascarella, ErnestT., Smart. John C., and Ethington, CorinnaA. "Long-Term Persistence of Two-Year College Students" BesoarohjnfljghaLEducation 24 (1986): 47-71. Pascarella, Ernest T., and Terenzini, Patrick T. "Predicting Freshman Persistence and Voluntary Dropout Decisions From a Theoretical Model. " JournaLof tIInbeLEducation 51:60-75. Pascarella, Ernest T. and Terenzini, Patrickl'. "Patterns of Student-Faculty Informal Interaction Beyond the Classroom and Voluntary Freshman Attri- tion" JeumaloLHicbeLEducatioMB (1977): 549552 151 Pedrini, Bonnie C., and Pedrini, D.T. "Attrition/Persistence: Stepwise Multiple Predictors for Disadvantaged and Control Freshmen." .lournaLofJnstruotional Psychology 13 (Summer 1986): 131-4. Perry, R. P, and Magnesson. J. L "Casual Attributions and Perceived Performance: Consequences for College Student's Achievement and Perceived Control In Different Instructional Conditions." W 81 (June 1989): 164 72. Plaucher, Frank E. "A Developmental Model for the Community/Junior College." QnmmuninLinlnnnflnxinw 15 (Wlnter 1987) 26 32 .Edited by David Mitzel. Washing- ton, D. 0.: National Council for Resource Development, 1988. Richardson, RichardC. Jr, and Leslie, LarryL. IhflnoossioloflreamfiLEinannlnn WW. Washington,D.C.: AACJC,1980. Ross, Jovita M. "Recruiting and Retaining Adult Students In Higher Education." NnnLDiLentinnatnLCnminuinnEdunatinn 41: 49- 62 Spanard, Jan- Maria A. "Beyond Intent: Reentrying College to Complete the Degree." WWW (Fall 1990): 309- 344 Stoecker, Judith., Pascarella, Ernest T., and Wolfle, Lee M. "Persistence in Higher Education: A 9-Year Test of a Theoretical Model." .lournalolgollooo Studentfleyoloornont 29 (May 1988): 196-209. Terenzini, Patrick T. "Designing Attrition Studies." WWW Research 36 (December 1982): 55-71. Terenzini, Patrick T., and Pascarella, Ernest T. "Voluntary Freshman Attrition and Patterns of Social and Academic Integration In a University: a Test of a Conceptual Model." WW6 (1977): 25-43. Terenzini, Patrick T. and Pascarella, Ernest T. "Toward the Validation of Tinto' 5 Model of College Student Attrition: A Review of Recent Studies." Research inflcanEdunatinn 12 (1980): 271-282. Terenzini, Patrick T., Pascarella, Ernest T., and Lorang. Wendell G. "An Assess- ment of the Academic and Social Influences on Freshman Year Educational Outcomes." IhnflnxinanLflInhnLEnunatinns (Wlnter1982I 86109 Tinto, Vincent. "Dropout From Higher Education. ATheoreticaI Synthesis of Recent Research" BesnnmnnLEdnnatinnalflnsnarnMS (1975) 89125 152 Tinto. Wncent. "Limits of Theory and Practice in Student Attrition." Jnumalottliohor MS (1982): 687-700. Tinto, Vincent. "Stages of Student Departure: Reflections on the Longitudinal Character of Student Leavlng." JournaloLljjohorEduoatlon 59 (July/August 1988): 439-55. TItIey, R., and Titley, 8. "Initial Choice of College Major: Are Only the "Undecided" Undecided?" IIQumaLQLQQJIfiQiSIudBIJLEflSQDDfll 21 (1980)! 293'293- Voorhees, Richard A. "Toward Building Models of Community College Persistence: A Logit Analysis." BesnathnflinhoLEdunatinnZG (1987):115-129. Voorhees, Richard A. "Student finances and Campus-Based FInancial Aid: A Structural Model Analysis of the Persistence of High Need Freshmen. BesnambinflnanEnunatinn 22 (1985): 65-92. Walleri, R. Dan. "Student Retention and Attrition In the Community College: A Review and Research Design" EBIQDnnumnntBnnLnduntinnSnnLinns NW (April 1981) 2 35 Webb, M. "A Theoretical Model of Community College Student Degree Persis- tence" Cnmmunimcnllnnafinvinn 16 (Spring 1989) 4249 Weiss, Steven J. and Giddan, Norman S. "Applications of Cost Effectiveness to Counseling Center Retention Programs." Journalolfiollooafiudont Personnel 27 (May 1986): 260-268. Whitener, Ellen M. "A Meta-Analytic Review of the Effect on Learning of the Interac- tion Between Prior Achievement and Instructional Support." Boyimof EnunatinnaLannaLnn 59 (Spring 1989): 65-86 Vlmliamson, David R., and Creamer, Don G. "Student Attrition in 2-and 4-Year Colleges; Application of a Theoretical Model." Jnumalmmflmmem QayaloornoanQ (May 1988): 210-17. Wirt, John G. "A New Federal Law on Vocational Education: Will Reform Follow?. " EhLDaltaKaooan 72 (February 1991): 425-433. Wojcicki, T. P., and Kaufman, K. "Vocational-Technical Schools Must Implement a Holistic Counseling Model Before Retraining Adult Workers" JournaLot Industnalleanbnrfidnnatinnfi (Summer 1990) 77-81 Young, Robert 8., Backer, Richard, and Rogers, Gregory. "The Impact of Early Advising and Scheduling on Freshman Success." Journolorgnllogofiludont Doyeloomontao (July 1989): 309-312. "Illllilllillilllllllllilili