AN INVESTIGATION OF FACTORS RELATED TO THE SUCCESS AFTER READMISSION 0R REINSTATEMENT OF COLLEGE STUDENTS ACADEMICALLY DISMISSED Thesis Ia: the Beam of Ed. D. MICHIGAN SYATE UNIVERSITY Thomas WiIIiam Hansmoiat 1963 IIIIIII IIIIIIIII III IIII III 3 1293 This is to certifg that the thesis entitled Investigation of Factors Related to the Success After Readmission or Reinstatement of College Students presented by Mr. Thomas W. Hansmeier has been accepted towards fulfillment of the requirements for m degree in _Ednc.a.t.ion 7/14/17/6 /// %¢»/16V Major professor ,/ Date July 31, 1963 0-169 LIBRARY Michioan State Dmvcrsuy AN TNVESTTGATTON 0F FACTURS RELATED TO TNIISUCCESS AFTER READHTSSION OR REINSTATEMENT 0F COLLEGE STUDENTS ACADEMTCALLY DISMISSED by Thomas William Hansmeier AX ABSTRACT OF A THESIS Submitted to Michigan State tniversity in partial fulfillment of the requirements for the degree of IXICTTIR OF EDI ’CAT‘IOP; College of Education 1963 AN INVESTIGATION OF FACTORS RELATED TD THE SUCCESS AFTER READNISSION OR REINSTATEMENT OF COLLEGE STUDENTS ACADEMICALLY DISMISSED by Thomas William Hansmeier The study sample consisted of 294 individuals who (1) entered Michigan State University as new freshmen in the fall of 1956, (2) were academically withdrawn by action of the University College between September, 1956 and June, 1958, and (3) who were readmitted or rein— stated as of the fall quarter, 1959. The sample divided into four sub— samples: 90 men and 35 women who were successful after readmission or reinstatement, and 123 men and Q6 women who were unsuccessful after readmission or reinstatement. The academic progress of these students was studied until their graduation, termination, or through the spring quarter of 1960. Seven research hypotheses were postulated, based on the assump- tion that variables which have been empirically proved to be related to college achievementamd persistence in general may also be related to achievement and persistence after readmission or reinstatement. One of the hypotheses was accepted, four were rejected, and the signi- ficance tests of two hypotheses yielded mixed results. Thomas William Hansmeier (1) The proportion of female subjects who were successful after readmission or reinstatement (35 of 81, or 43 percent) was not signifi- cantly different from the proportion of male subjects who were success- ful (90 of 213, or 42 percent). (2) The proportion of veterans who were successful after read— mission or reinstatement (21 of #1, or 51 percent) was not significantly different from the proportion of non-veterans who were successful (69 of 172 or 40 percent). (3) On the average, successful men and women earned higher grades in high school than unsuccessful men and women, but t-tests of the differences in quarter rank in high school class means were not significant. There was no significant difference in the numbers of successful and unsuccessful males and females who ranked in the upper half or lower half of their high school class. However, the rate of success of subjects who ranked in the first quarter was significantly greater than the success rate of subjects who ranked in the bottom quarter, (4) The mean age differences between successful and unsuccessful sub—samples were not statistically significant. (5) Successful subjects, on the average, earned higher scores than unsuccessful subjects on each of the eight college entrance exam— inations. Thomas William Hansmeier For male sub-samples, the following mean score differences were significant: Michigan State University English Test, Michigan State College Arithmetic Proficiency Test, American Council on Education Pscyhological Examination-Linguistic Ability,and Michigan State University Reading Test—Vocabulary and Total. The American Council on Education Psychological Examination—Quantitative Ability and Total and the Michigan State University Reading Test-Comprehension mean score differences were not significant. For female sub—samples, only the Michigan State College Arith— metic Proficiency Test score means were significantly different. (6) First-quarter grade—point average means were significantly higher for successful men and women than for unsuccessful men and women. Cumulative grade-point index means for successful sub—samples at the end of the quarter of initial withdrawal; end of the quarter of initial readmission or reinstatement; after readmission or rein- statement; and at graduation, termination, or the end of spring quarter, 1960 were also significantly higher than the corresponding mean grade-point averages for unsuccessful sub-samples. (7) There were no significant differences in socio-economic status between successful and unsuccessful subjects, using occupation, education of father, and education of mother as criterion measures. Thomas William Hansmeier Coefficients of biserial correlation and Pearson product-moment correlation coefficients were employed as indexes of the predictive power of ten academic variables--eight college entrance examination scores, first-quarter college grade—point average, and cumulative grade— point average at the end of the quarter of initial withdrawal. The entrance test scores proved to be mediocre indicators of post-readmission or reinstatement success, yielding biserial correlation coefficients of .084 to .307 (median = .1885). Pearsonian r's ranged from .043 to .290 (median = .182) using final grade-point average as the criterion; and from -.046 to .249 (median = .043)when grade—point index after readmission or reinstatement was employed as the dependent variable. 0n the other hand, the grade-point measures were relatively potent predictors, yielding biserial correlation coefficients of .513, .532, .546, and .650, and product-moment correlation coefficients of .507, .534, .564, and .686 against the criterion, final grade-point average. The magnitude of the zero-order r's decreased when grade in— dex after readmission or reinstatement was the dependent variable (range = .249 — .510, median = .334). As an exploratory extension of the study, the relationships between certain academic behavior patterns of the sample and success after readmission or reinstatement were investigated. It was discovered that: (l) successful students changed their academic major significantly more times than unsuccessful students; (2) successful and unsuccessful sub-samples did not differ significantly either in enrollment in, or Thomas William Hansmeier achievement in, remedial courses; (3) repeating courses was a critical factor in post-readmission or reinstatement success, the successful subjects gaining significantly more points by this method than the unsuccessful subjects; (4) reinstated students were significantly more successful than readmitted students, but the latter were considerably poorer risks due to their low pre—withdrawal cumulative grade—point average; and (5) only 40 subjects (23 successful and 17 unsuccessful) transferred credits to Michigan State University from other institutions, and the mean number of credits transferred did not differ significantly between successful and unsuccessful sub-samples. The findings of the study warrant the following conclusions: (1) The prediction of college achievement and success is in- precise and uncertain at best, but it is especially difficult to pre- dict the post-readmission or reinstatement success of academic withdrawals. (2) Standardized test performance effectively differentiates between withdrawals and non-withdrawals but does not discriminate well between academic withdrawals who are successful or unsuccessful after readmission or reinstatement. (3) Academic withdrawals are quite homogenous in scholastic aptitude, but successful and unsuccessful withdrawals differ signific— antly in college achievement prior to withdrawal. AN INVESTICAIIUN (IF FAC'IIIRS RELATED TO THE SITCCIEZSS AFTER READMISSIDN UR REINSTATEMENT 0F COLLEGE STUDENTS ACADEMICALLY DISMISSED by Thomas William Hansmeier A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DIICTDR (IF EDUCA'I'IILIN College of Education 1963 ACKNOWLEDGEMENTS The writer is indebted to Dr. William Farquhar, who served as major advisor, and to other members of the Guidance Committee-—Dr. A. 0. Haller, Dr. John X. Jamrich, and Dr. Walter F. Johnson—~for their expert guidance and discerning criticisms. Dr. John N. Winburne and his University College colleagues were most generous in their support of the investigation. Dr. Willard Warrington is gratefully remembered for the encouragement and the advice which he provided. My friends and colleagues at Kent State tniversity, Dr. Ronald Roskens and Dr. Henry Dizney, were willing and able editors and morale boosters. Dr. Dizney was especially helpful in the design and statis- tical analysis aspects of the study. Mr. David Gronauer volunteered the electronic computer assistance which greatly facilitated the pre— sent research. Last chronologically, but of primary importance, the writer expresses profound gratitude to his wife, Jean. It was her lot to endure for five years the temperament of a husband made more irascible by the pressures of a prolonged doctoral program. This she did with understanding and forgiveness, and with remarkable graciousness. ii TABLE HF CONlENlS CHAPTER PAGE I. 'HlE PROBLEM . . . . . . . . . . . . . . . . . . . . 1 Statement of the Problem . . . . . . . . . . . . ' 2 The Problem: Its Nature and Importance . . . . . u Definition of Terms . . . . . . . . . . . . . . . 10 Overview of the Study . . . . . . . . . . . . . . 11 ll. REVIEW OF THE LITERATTRE . . . . . . . . . . . . . 13 Academic Factors in Attrition . . . . . . . . . . 14 Non—Academic Factors in Attrition . . . . . . . . 25 Readmission Studies . . . . . . . . . . . . . . . 34 Summary . . . . . . . . . . . . . . . . . . . . . 42 III. DESIGN OF lNE STEDY . . . . . . . . . . . . . . . . 45 The Null Hypotheses . . . . . . . . . . . . . . . 45 The Demographic Data. . . . . . . . . . . . . . . 60 Statistical Analysis. . . . . . . . . . . . . . . 65 Summary . . . . . . . . . . . . . . . . . . . . . 68 I\ RlSULTS . . 70 Testing the Null H}potheses . . . . . . . . . . . 7O Predicting Post-Readmission or Reinstatement Success . . . . . . . . . . . . . SQ S Ulnllla 1‘}? a . a o o a n o o a o o o , a a o o o I a 96 iii CHAPTER \' ACADEPIIC BEHAVIOR PA'I"I'ER.'~IS Initial Curriculum Choice and Curriculum Change. Readmission or Reinstatement Repeating Courses. Transfer of Credits. . . . . Enrollment and Achievement in Remedial Courses Stnwmary VI. SDDLARY. CHNCLISICUS ASJD IMPLICAI'IHNS FOR FUTURE RESEARCH Summary . . . . . . . Conclusions . . . . . . . . . Implications for Future Research BIBLIOGRAPHY . . . . . . . . . . . APPENDIX A iv PAGE 97 97 101 104 106 109 TABLE 2. 2 L'\ f) L\ l .2 LIST OF TABLES PAGE Intercorrelations Among Variables Used in Studying the Applicants for Readmission to Two Colleges of Ohio State University . . . . . . . . . . . . . . . . . . 38 Relationship of Readmitted Students' Grade-Point Average Before and After Absence from the Univer- sity of Utah According to Activity Engaged in While Away from College, Length of Absence from College, and Performance on the English Placement Test . . . . . . #0 Levels of College Accomplishment Attained by Success- ful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . . . . . . . . . . . . . . . 48 Quarter Rank in High School Graduating Class for Successful and Unsuccessful Readmitted or Rein- stated Males and Females . . . . . . . . . . . . . . . . 51 College Entrance Deficiencies or Conditions of Success- ful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . . . . . . . . . . . . . . . 52 Academic Scholarship Actions,Including Withdrawal Actions, Incurred by Successful and Unsuccessful Readmitted or Reinstated Males and Females. . . U1 U1 Mean Criterion Scores for Male and Female Comparison Groups and F-Ratios Testing Comparison Group Mean Differences . . . . . . . . . . . . . . . . . . . . . . . 58 Mean Age at College Matriculation of Successful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . . . . . . . . . . . . . . . . . 73 College Entrance Achievement and Aptitude Test Scores for Successful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . . . . . 75 Cumulative Mean Grade-Point Average at End of First Quarter in College, End of Quarter of Initial With- drawal, End of Quarter of Initial Readmission or Re- instatement, After Readmission or Reinstatement (Ex- cluding Repeated Courses), and at Graduation, Termi- nation, or End of the Spring Quarter, 1960 for Success- - ful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . . . . . . \l 00 TABLE L\ .I.\ J_\ U1 4.6 4.8 4.9 4.11 .12 4) PAGE Occupational Status of Fathers of Successful and Un- successful Readmitted or Reinstated Males and Females . . . 80 Level of Education Achieved by Fathers of Successful and Unsuccessful Readmitted or Reinstated Males and Fenlales . I O C O O C C O O O O O O O I O I O C I O O O 82 Level of Education Achieved by Mothers of Successful and Unsuccessful Readmitted or Reinstated Males and Females. I I O I O O O O I O C O C O C O I O O I O I O 83 Coefficient of Biserial Correlation Achievement and Aptitude Estimates of the Criterion, Success After Readmission or Reinstatement . . . . . . . . . . . . . . . 85 Pearson Product—Moment Correlation Coefficient Achieve- ment and Aptitude Estimates of the Criterion, Final Grade-Point Average of Readmitted or Reinstated Males and Females . . . . . . . . . . . . . . . . . . . . . . . . 86 Pearson Product-Moment Correlation Coefficient Achieve- ment and Aptitude Estimates of the Criterion, Grade- Point Average After Readmission or Reinstatement of Readmitted or Reinstated Males and Females . . . . . . . . 87 Number and Percent of Readmitted or Reinstated Males and Females Achieving Within Various Grade-Point Intervals After the1?irst Quarter in College Who Were Successful after Readmission or Reinstatement . . . . . . . 90 Number and Percent of Readmitted or Reinstated Males and Females Achieving Within Various Grade-Point Inter- vals After the Quarter of Initial Withdrawal Who Were Successful After Readmission or Reinstatement . . . . . . . 91 Number and Percent of Readmitted or Reinstated Males Achieving Within Various Grade-Point Intervals at the End of the First Quarter in College Who Earned Grade-Point Averages After Readmission or Reinstate- ment (Exclusive of Repeated Courses) at or Above Specified Values . . . . . . . . . . . . . . . . . . . . . 92 Number and Percent of Readmitted or Reinstated Females Achieving Within Various Grade-Point Intervals at the End of the First Quarter in College Who Earned Grade— Point Averages After Readmission or Reinstatement (Exclusive of Repeated Courses) at or Above Specified Values . . . . . . . . . . . . . . . . . . . . . . . . . . 93 vi TABLE PAGE 4.14 Number and Percent of Readmitted or Reinstated Males Achieving Within Various Grade-Point Intervals at the End of the Quarter of Initial Withdrawal Who Earned Grade-Point Averages After Readmission or Reinstate- ment (Exclusive of Repeated Courses) at or Above Specified Values . . . . . . . . . . . . . . . . . . . . . . 94 4.13 Number and Percent of Readmitted or Reinstated Females Achieving Within Various Grade-Point Intervals at the End of the Quarter of Initial Withdrawal Who Earned Grade-Point Averages After Readmission or Reinstatement (Exclusive of Repeated Courses) at or Above Specified Values . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 5.1 First Curriculum Choices of Successful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . 98 5.2 Academic Aptitude and College Achievement Measures for Students Readmitted at Least Once and for Students Always Reinstated . . . . . . . . . . . . . . . . . . . . . 103 5.3 Mean Number of F Credits and D Credits Repeated, and Credits and Points Gained by Repeating Courses, for Successful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . . . . . . . . . . . . . 105 3.4 Total Number of Enrollments in Remedial Courses (Improvement Services) for Successful and Unsuccessful Readmitted or Reinstated Males and Females . . . . . . . . . 108 vii CHAPTin’ l The high rate of student attrition from American colleges and universities has been a grave concern of educators for many years. . Approximately one-half of the students enrolled in institutions of higher learning withdraw, either voluntarily or involuntarily. be- fore obtaining an undergraduate ngJYWB. a situation WQICM lffert describes as "an alarming waste of our most. competent manpower."1 There is an extensive body of literature addressed to the dropout problem. but the research conducted thus far has yielded “neither adequate nor conclusive answers.“2 In an effort to decrease the attrition rate. educators have subjected the admissions process to careful and critical scrutiny. The literature is replete with reports of validity studies of various instruments and techniques which have been de ised to more accurately appraise the scholastic abilities and aptitudes of prospective stu— dents. Howeyer. the prediction of success in college is made diffi— cult by the multiple factors which contribute to academic achievement. l '.' 1.!1g; t t)11 . dents.” ‘. S, Office of Education Bulletin 1958. No. l teas D.C.: ', 8, Government Printing Office, 195/). p. 99. tobert E. Lffert. “Retention and Withdrawal of College Stu— l. f). _, - ..- " ll ' “.ohn summe‘skill. Dropouts from tollege. Chapter 19 in Ne\dlt:t$anford ‘ewi.). The Anuu‘ican (Rillege fund York: gunnilfiiley and Sons. lnc., 1962). p. 627. Summerskill3 reviewed 23 studies dealing with the relationship between college grades and attrition and found that about one —third ofthe dropouts were attributed to academic reasons. rut Summerskill also observed that many academic failures are undoubtedly caused by “underlying problemsiwhich are psychological, parental. social. or financial.”u Thus far. relatively little attention has been given to the situation created by students who encounter academic failure and are dismissed and who subsequently apply for readmission at the same or another institution. There is a paucity of research data regard- ing the post—readmission achievement of students. Statement of tie Problem lhe study was designed to investigate the relationship between selected academic and non—academic variables and success after read— mission or reinstatement. It was hypothesized that a marked relation— ship exists lretween certain academic and non—academic characteristics of students who were academically dismissed from the Cniversity College of Michigan State Iniversity and their success or failure after readmission or reinstatement to the Kniversity. More specif— ically. the cardinal purpose of the study was to investigate the following research hypotheses: 3lbid.. pp. 636-637. “tbid., p. 637. of succ puprose havior, sion or credits The 835 of -3- The proportion of women students who are successful after readmission or reinstatement is greater than the proportion of men students who are successful. The proportion of veterans (students with experience in the military) who are successful after readmission or re— instatement is greater than the proportion of non-veterans who are successful. Students who are successful after readmission or rein- statement earned higher grades in high school than stu- dents who are unsuccessful. Students who are successful after readmission or rein— statement were younger at matriculation than students who are successful. Students who are successful after readmission or rein— statement performed at a higher level on college entrance tests than students who are unsuccessful. Students who are successful after readmission or rein— statement earned higher grade—point averages during their first—quarter of college than students who are unsuccessful. Students who are successful after readmission or rein— statement came from families with higher socio—economic status than students who are unsuccessful. investigation was concerned primarily with the prediction‘ subsequent to readmission or reinstatement. A secondary the study was to describe some of the patterns of be— i.e., initial CUrriculum choice and curriculum change, readmis- reinstatement after withdrawal, repeating courses, transferring from other institutions, and enrollment and achievement in remedial courses,which may relate to success for the readmitted or reinstated student. The> Prtnilem: ‘Its TVath13 atul fnngort;ancc The American ideal has been to equalize educational oppor- tunities and to broaden the scope of higher education. whereas other nations haVe used higher education for the development of an intel— lectual elite.S A marked differential in student performance may be expected under such widely variant. condit ions. as evidenced by the contrasting rates of student retention in British and American institutions of higher learning. Studies of success and failure at the iniversity of London6 and the I'niversity of Liverpool7 revealed that at least 70 pcmcent of the students graduated in regular progression and that more than 80 percent graduated eventually. In contras.. only atout AU percent of the students enrolled in “nited States institutions graduate on schedule. and an additional 20 percent earn a degree after some delay.8 tbviously. British university students exhibit greater staying power than their American counterparts. The same would hold true if collegians in this country were compared with the university 3for example, Hollinshead states that “European aims and methods are so different from ours that 'he question arises of whether it is even appropriate to try to compare systems which have such different putjx)ses.” fiyron S. Hollinstead. ”is turopean hducation Hetterf.‘ lte Lducational Record, 39;E9-96. April. 1958. p. 91. r‘7I‘-.'icholas Dalleson. ”Stthhntt Performance {n‘ [ti ersity College. London. 1948-1951.” “nirersities Quarterly. 12:288—319. hay. 1938. /Jlanmes PKJUIrtft)rd."Sth:Ceess thJ l aiflutx: at tlxe 'ttixt2rsi.ty." niyersities Quarterly. 11:226-23Q. Bay. 1957. '.l‘ft‘t‘t. up. Cll., P. 11“.), students of continental fiurope or of other nations where higher educa- tion is restricted to a select few. This does not signify that Ameri- can colleges are inferior. Rather. it is a consequence of the Ameri- can commitment to democracy in higher education. Higher education in the 7‘nited States has not been forced to conform to uniform patterns. Nowhere in the world can a wider range of institutional types be found: denominational and non-sectarian liberal arts colleges. state and pri- vately—endowed universities. municipal unive‘sities. land—grant in— stitutions. professional schools. technical institutes, teachers colleges, junior Colleges. and military institutes. The curriculum of higher education is Lroad and diversified. A protiferation of courses are offered which are designed to satisfy the multifarious interests of a heterogeneous student body and the specialized needs of a complex society. The popularity of higher education in the ‘nited States is an indication that the colleges and universities are offering programs which do. in large part. satisfy the demands of American society. Per capita enrollment increased tenfold between 18s0 and 19317.9 The campus population increased almost 1000 percent between 1901) and 1948.10 and enrollment in institutions of higlter learning ex— _ . . . .. 1] . . . . ceeded 3.3 million in 1900. never before in history has a nation 9Richard Hofstadter and C. Hewitt Hardy. The Uevclopment and Scrflxs of ? ighet‘lllucatixnt in tlx)'.nittul Staltus (keel kirk: (kilumriei “niyersity Press. 1932). p. 21. l - H .. . W. - H g . . 1(»ohn :. :arubacher‘euui.3illis hudyz r igher :(hnration in litnisitilni (NEMJ'VOtk: liarpen‘ and térotlw rs ludglileJrs. 1935) . pp. 37u—373. l-lhtnialtl B. TllURH35tHl. l HIW)1llntdll P113] C litHlS ftn‘ l ielulr l dLH - ation, 19r1—l979 tdeport of special committee of the American Associa— tion of Collegiate Registrars and fidmissions Htficers. 19 l). p. r. so nearly approached the ideal of universal higher education. Yet. projected future enrollments are even more dramatic. The U. S. bffice of Education recently predicted that 44 percent of the young people between the ages of eighteen and twenty—four would attend college at some point in the 1960's. But another survey indicates that. arresting as this prediction is. it may be much too low. A broad—based survey conducted for the Ford Foundation in the spring of 1959 by leo Roper and Associates showed that 69 percent of the children below the age of eighteen are expected by their parents to go to college. This would mean almost a tripling of present enrollment—~to a college population of 11 [Ill 1 1 i()Ii. 1 2 Concurrent with the trend toward an unprecedented equalization of opportunity to participate in higher education is a strong cross— current. a striving to maintain high academic standards in colleges and universities. When many are permitted to enter the competition but the> steuidarmi of U; uccess is demanding. it is predictable that sub— stantial numbers may fail. Indeed. the incidence of academic failure is alarmingly high. and the attrition rate is viewed as a significant pronlem by many observers. Critics contend that, a high dropout rate. represents a waste of the limited financial, physical. and faculty resources of individual institutions; serious emotional upset. in many cases. to the students who fail and to their parents; and a general lowering of the quality of higher education. Bell cogently summarizes the objections to a policy of ”admit and flunk.” 121. R. McConnell. A General Pattern for American Public Higher Education (New York: McCraw—Hill Book Company. Jnc.. 19d2). p. l. The whole effort expended in getting the unqualified into college, and then dropping them. must be almost a complete waste. sociologically and economically. The college must gear its staff and plant to take care of the mass of dropouts. even though they are only temporarily there. The instruction in the freshman classes must be conditioned by the stand— ards of a class with a percentage of incompetents. Certainly it is a drag on the teacher to work with them...and the progress of the whole group must be slowed down. This may account in part for the com— mon complaint that first—year students are bored by a repetition of ground already covered.... The system is a cruel one. The unqualified student is taken in full of hope and surely at some sacri— fice in money or lost earning power. and then a few months later he will receiye the flunk notices. This cannot be good for either the student or the educa— tional system. Some students will take this lightly. Some will be stimulated to further effort. but many will suffer from the stamp of failure in a test to which they never should have ween subjected. It is not merely the individual who suffers; most families are disappointed. There must result not a few in— feriority complexes.13 American institutions of higher legunting historically have been. and continue to be. select ive to varying degrees in their ad- missions. Pri ately—endowed schools have exercised the prerogative of statiJm; their (WM) standards ftn' selectiant. Publiclyudylpported institutions. although restricted by law. public opinion. and/or alumni pressure. have been arle. nevertheless. to exclude forth- rightly or to “discourage” applicants with poor prognosis for success. 'Nte CentEW” for tlwi:€tudy (fiftiigher‘lkjucaticnt. ‘nixtnwsity tn? Califtnw1ia. attempted to estimate the selectivity operant in higher education by 13£tairci Be?ll. "Athnit atul Fltntk." At lard ic Tfllntltlyt. \til. 21(). Yo. Q:Su—56. Uctober. l9r2. p. 53. relating the ability. as measured Ly scholastic aptitude test scores, v—-—4 of college freshmen to that of high school graduates and tie general population. It was found that in 1952 about 73 percent of the college entrants from a sample of 200 institutions came from the upper half of high school graduates in ability. In 1959 about 77 percent of the freshmen from 167 institutions came from the upper half of high school graduates. It was estimated that about 90 percent of the 1959 freshmen ranked in the upper half of the general population of seventeen—year-olds in scholastic aptitude.14 The nature of selective admissions in the future remains a topic for (HMTjECtUFB enufl debate. Pcfknnnsll foresees gut hmnediate but short-lived raising of admission standards to keep student intake commensurate with expansion in the number and size of in- 1 Ln stittn ions. _MJt he ccnu21udes 'Wltat in ttwa long tint. Anmn‘ican higher education as a whole will not‘become more selectiye. Mass education is here to stay. American higher education will ,, , ' ' 9’16: - ‘ . ‘ become more rather than less lHClUSLYe due to soc1al and economic forces in American society. As educators probe the problem of excessive dropouts. it is natural and proper that they should focus their attention initially on the admissions process. Ideally. potential failures would be identified before college entrance is effected. thereby averting the O lu‘tct‘jt‘ntnell. op. (‘11.. DP. 15—9. lsibid . up ltd—l2. '1"'Ibid.. pp. 8 18. many difficulties which accrue from termination. Lt must be acknow- ledged. however. that two factors will alumys prevent the total cir- cumvention of student failure. Firstly. as noted above. stringent admission standards, except within a limited number of institutions. are not consistent with the American philosophy of higher education. Secondly. the science of prediction is not well developed. Some of the human variables. e.g.. motivation and creativity. which are known Components of academic achievement have proved too nebulous to be quantified with precision at present. Other factors. e.g., scholastic aptittnhfi. have ftmnttneasured eunj quantifitwl. but the enwngr of unwasure— ment is so great as to leave Considerable margin for error. in short. prediction of academic success. or failure. is fraught with uncer- tainty. Faced with the inescapable assumption that academic failure will always be a part of American higher education. it becomes im- perative that the process of readmission be subjected to critical analysis and systematic research. The readmissions process is con- cerned with identifying those students who have failed initially but who are capable of eventual success in college. The rehabilitation of academic failures is, and will continue to be. an important func— tion in American higher education. Therefore. it is important that research be undertaken to augment the paucity of knowledge availahle to those institutional officers who must formulate policy and make vital decisions in the area of readmissions. _ 10. Definition of lernw Tniversity College. Formerly the Basic College. A major administrative and instruc— tional division of Michigan State Ini ersity. All students who enroll at Michigan State Cniversity are assigned to the lni ersity College and remain there for approximately two academic year“. Students who earn 92 credits or more, with a grade-point average of at least 2.00, are eligible for transfer to the upper college of their choice. .pper College. One of several administrative and instructional divisions of Michigan State University. Fpperclassmen of junior and senior classification are enrolled in upper college according to their fields of specialization. ZJithdrawal. A student whose scholarship has fallen below the minimum accept— able level. and hence is involuntarily withdrawn by lniversity College action. Also, the University College action by which the student's enrollment is terminated. Readmission. A Lniversity College action whereby a withdrawal is allowed to re-enroll. but only after one or more terms have elapsed since the withdrawal action. Also, a student who is the recipient of this act ion. -11- Reinstatement. Immediate readmission; that is. a withdrawal is allowed to re— enroll the term immediately following the withdrawal action. Also. a student who is the recipient of this action. Reinstated students are withdrawn from the L'nixersity College "on paper," but they do not actually lea e the campus. As defined in this study, “success,” “academic success.“ or 9 “college success' after readmission or reinstatement will be used synonomously to mean: (1: graduation with the baccalaureate degree; (2) termination with a grade—point average of 2.00 or better; or. (3) progress toward the baccalaureate degree with at least a 2.00 grade—point average. Overview of the Study In Chapter ll is contained a review of the literature which will focus primarily on {1) the relationship between certain academic variables and attrition. t2) the relationship between certain non- academic varia les and attrition, and (3) previous studies of the readmission process. In Chapter ll: are presented the methodology and procedures employed in the study. The null hypotheses are stated. The statis- tical methods and their underlying assumptions are discussed. -12- In Chapter IV are reported the results of the data analysis. The findings are presented in tabular form, and the results are dis- cussed. Academic behavior patterns of readmitted or reinstated students are described in Chapter V, and the relationships between the behavior patterns and post-readmission or reinstatement success are discussed. The summary. conclusions,and implications for future research may be found in Chapter VI. Cn’A PTICR ll l-lleilj-‘n.’ (ll: THL LITE“. will»- The attrition problem is made complex by the multiple and interrelated factors which are involved. Various investigations have indicated a relationship between college student attrition and a myriad of variables: age. sex. curriculum choice and vocational choice. curriculum change. size and location of home town. size and location of high school, high school performance. scholastic aptitude as measured by standardized test performance. reading ability, study habits. personality factors. family income. occupation of father. educational level of parents, family and/or personal health. marriage. military service. college grades, extra—curricular participation. class attendance. employment. academic load. college residence. and social relationships between the student. and his family. his peer group. and the faculty. There is seldom a single reason for the termination # of a college career. jsually a constellation of factors contributes to the student's final decision. An extensive body of literature devoted to the problem of college dropouts exists. This review was restricted to a summary of findings in three areas of research: fl) the relationship between certain academic variables and attrition. (2) the relationship be- tween certain non—academic variables and attrition. and (3) studies which related directly to the readmission of academic failures. 'fhe research re\iewed 'here suggested the general approach -13- -14- to the present study, and the research hypotheses stated in Chapter I were derived from the findings reported by tlue researcners cited herein. Academic Factors in Attrition Representative studies of high school grades, standardized test performance. first—term college grades. curriculum choice. and curriculum change as factors in college student attrition were included because of their relevance to the present study. High school grades. A number of studies have established a substantial relationship between admissions data and persistence in college. iffert} in a definitive study of 12,667 men and women who entered institutions of higher learning in the fall of 1950. found that; rank in high school class and college placement test scores were both related to college 0 success. the former being a ”much better indicator' of the probayility of college graduation than was the latter. However. plaCement-test standing rather than rank in high school was the better index of pro— bable length of persistence in college, and the casualty rate was lower anwnug students ydjliltigh test scrntus than among stinkntts with high grades in secondary school. 1 so ert ll. lffert. 'Whetentitnt and gdfilhlrawal (3f Colltgyz Sttnhntts.' S. (lffice of lkhh2ation 'flJllOtlJl 1058. lk). l (Chashingttnt. ).C.: T. F. dovernment Printing office. 1957). pp. 93-97. !_ins guuj Pitt2 fourui higflt schcu)l gtwmjes tt) be nu)re Cl()5€l}’ related than college entrance test scores to persistence in college. Vsing data from 1994 subjects who enrolled as new freshmen at the. Cniversity of Wisconsin in the fall of 1945. they computed correlation coefficients between number of semesters registered and perCentile rank in high school graduating class (.374), percentile rank on the American Council on Education (ACE) Psychological Examination (.288).. and first semester grade—point average (.541). Heintraub and Salley.3 in a study of 10th Hunter College students,investigated the relationship of two entrance test measures and high school achievement to graduation from college. academic withdrawal, and non-academic withdrawal. For the three groups who ranked in the upper half of their high school class and above the fiftietlttxarcentile (N1 the tkfigntts examinations and the ACE thunnination. there was a relatixely high rate of graduation '60.5. t1.2. and 56.3 perrewn_, respecti\wfl¢:) and a ltnu rate of aceuhflnie unjtnjrawal {6.8. 7.6. and 14.4 percent. respectively). There was a lower graduation rate (40.2. 39.0. and 46.3 percent) among students who ranked in the bottom half on the three measures and a higher rate of academic with— drawal I31.2. 31.3. and 23.9 percent). The. incidenCe of non-academic withdrawal was a out the same (range: 22.1 to 25.5 percent) for all six groups. 2 .. . . “H - . . L. J. Lins and My Pitt. lhe staying Power and date ot Pro— gress of gniversity of Wisconsin Freshmen.“ College and ”hirersity. 29:E6-99. Uctober, 1933. 3 . p . . a n. . Hutti C. .veitttrain) arul hutlt L. tralltay. (Jrathlatltnt Prtnq>ect 5 (M7 an I ntewcing I resflnnattfi’ u)urw1al ()f txiuc;atitntal l‘estvarrh. 39; ll€«-120. twcttxher. 1943. and M65 enrolled students randomly and 73 L; studied 55h dropouts State College of Washington. s Slocum 583 at the rank in high school class wa Exam- selected from three clas Median ACE found that the average percentile for enrolled students compared to 58 for dropouts ination percentile scores for the enrolled and the dropouts were 53 discovered a statistically significant . S and u2. respectively. Bragg difference between the high school grade averages of "continued" 45 of and “withdrawn” university students. studied intensively a group of 98 advisees Yoshinob whom dropped out by the end of the freshman year and 53 who remained. .01 level between the two There was a significant difference at the using mean high school grade average and mean ACE Examination groups. rating as the criteria. There was also a significant difference be- tween those who dropped for academic reasons and those who withdrew for non—academic reasons when the same criteria were applied. grades euul there is research evidence of a suhstantial posi— In summary. tive relationship between high school marks and college ". school tend to Individuals who did well in high college persistence. do well in college. Slocum. “Social Factors Involved in Academic Nortality. 1956. . 'w. ' 32:53-64. Fall, 4W L College and “niversity. DEmma N. Bragg. 'A Study of Student Withdrawal at Journal of Educational Psyrhology, 47.199-202. April. 1956. tlte htuj ()f tine } f{‘Slf— 32zu2-48. September. 1959. 61. Roger Yoshino. ”College flrop~outs at man Year.” tournal of Educational Sociology. -17- Standardized test performance. Studies cited in the preceding section indicate that standar— dized test scores may be used to predict probable academic success or failure with some efficiency. The results of other studies serve to reinforce this assertion. Halladay and Andrew.7 in a survey study of dropouts from nine Arkansas colleges. determined that 6& percent of the dropouts scored below the national median on an ”ability test.” Furthermore. 74 percent were below the national median on an "English achievement test," with approximately 60 percent ranking in the lowest quarter. Eurich8 found reading ability to be a potent factor in college performance. When he matched probation students with non—probation students on the basis of intelligence (as measured by the Miller Analogies Test). sex, academic major, and college classification. there was a marked difference between the two groups in performance on the Yinnesota Reading Examination. Gerlach9 studied 208 students dropped for poor scholarship at F-liami T'niversity of Ohio. and discovered that a majority of academic failures could be classified as poor risks at admission on the basis 7. _ . . L J. whitney Halladay and Jean C. Andrew.'Jrop-outs from Arkansas Colleges.“ Personnel and Cuidance Journal. 37 212—213. November. 1958. 8Alvin C. Eurich. ”College Failures.“ School and Society. 37: 092-690. May 27. 1933. \0 Harry N. Cerlach. ”The Attendance Record and Scholastic Apti- tude Test Scores of Students Dropped for Poor Scholarship.“ College and "nive‘sity. 24:370-373, April. 1949. of performance on the ACE Examination. Scores were available for 180 of the 208 subjects. About one-half {50.6 percent) scored at the thirtieth percentile or below. and 70.6 percent ranked in the lower half. Gerlach observed that barring poor risks would be an easy solution to the attrition problem if there were a way to identify those students who score low but succeed. Taylor}0 in a study of 20 all—A students and 20 all—F students: found marked differences in test performance between the two groups. Intelligence test percentile scores ranged from 65 to 100 for the all-A students and from 5 to 78 for the alleF students. The median percentile ranks were 91 and 36. respectively. On a test of reading comprehension, the percentile score range was 66 to 100 for the all-A students and 12 to 83 for the all-F students. ‘he median values were 95 and 38, respectively. Two studies indicated a less pronounced relationship between standardized test performance and persistence. Johnson,11 using as subjects a representative sample of college freshmen who returned and another sample of freshmen who did not return. made comparisons on the basis of ACE Examination — Quantitive Ability (ACE—Q) and ACF Examination — Linguistic Ability (ACE-L) scores. Iowa Silent Reading Test scores. California Test of Personality ratings. and college grade- point average. He found that the only significant difference (.01 level) 10 . . . L. 0. Taylor. “Why College Students Fail.” oournal of Higher qucation, 19:425-427. November. 1948. Llflranville B. Johnson, Jr.. ”A Proposed TechniqUe for the Analysis of Drop-outs at a State College.” Journal of hducational Reseénxflt, 47: 3381-387. .tanuaJWr. 1954. v .. — -19- between the male subjects who stayed and those who left was in grade— point average and that the most significant difference ..02 level) between female subjects who stayed and those who left was in Iowa Silent Reading Test performance. Hunger12 investigated the relationship be- tween Uhio State Psychological Examination f08PE) scores and the firs - semester college grades of 748 students who ranked in the upper. middle. and lower thirds of their high school graduating classes. . and the length of their college attendance. It was found that once students had been categorized according to high school rank. the OSPE offered little as a predictor of persistence. 0n the basis of research evidence presented here and else- where. it may be concluded that performance on standardized tests of scholastic aptitude. achievement, and reading ability is an ef— fective predictor of college academic achievement. length of persis- tence. and probable college success. However. the strength of these relationships varies considerably. depending on the testis) used and the nature of the sample. First—term college grades Research findings indicate that grades earned during the first term in college are potent predictors of subsequent achievement I . 13 , ¢ 0' oo 11 00 and perSLstence. Bragg found that the 'continued group at "N. L. 7 . 1 , ‘9 " ' ' ‘ 0' . 1“Paul F. hunger, btudent Persistence in College. Personnel and Guidance Journal. 35:241—243. December, 1956. 13 Bragg. op. cit. _20_ had significantly higher first—semester grades than the "withdrawn” 4 . - . . . . 1 using 891 subjects who had graduated in the lower group. Hunger, third of their high school graduating classes and who had been ad- mitted to college on a trial basis. discovered a meaningful relation- ship between length of stay at the university and grades in first— semester COU‘ses in tnglish. history. mathematics. social science, and science. He concluded that grades in first college conses may be realistically used in predicting probable persistence. Reference has been made to the Tniversity of Wisconsin study15 which demonstrated that first-semester grade—point average was more closely related to number of semesters registered (r = .541) than either percentile rank in high school class tr 2 .374) or ACE Exami- nation scores (r = .288). Potthoff.1€ in an early study at the Vnive‘sity of Chicago. investigated the feasibility of predicting the ultimate failure of college students on the basis of first-quarter records. His subjects were 482 students who entered as freshmen in 1924 and who continued in attendance for at least one.and for as many as two, academic year (I) l4 - . .. _ . . Paul I. Hunger. Length of Residence and First Courses of Unpromising College Students.” School and Society. 81:120—122. April 1c. 1955. 15Lins and Pitt. 9p. cit. lbEdward F. Potthoff. “Predicting the Vltimate Failure of Col- lege Students on the Basis of their First Quarter's Records.” School and Society, 33:203—204. February 7, 1931. -21... He computed coefficients of correlation between two—year grade— point average and firs —quarter grade-—point a erago .slfi), high school average ;.6201. and intelligence tes scores .433). The multiple correlation derived by combining all three variables was .842. only slightly greater than the zero-order correlation of .810 obtained by using fi‘st—quarter grades alone. Vunger17concluded that it is advantageous to combine high school rank and first—term grade average in predicting persistence. In a study at the tniversity of Toledo. be found that students who grenhlattul iti the tippen‘ tlzird (3f ttneir ltigh i5ChtK)l clx;ss xvill [>robeflvly persist if they obtain a first—semester average of almost b or better. students from the middle third should have an average sorew at Higher than 61(3, and students th t_e lowet‘tfl.ird in high srlxn)l rank will persist if they earn barely a C average in their first semester at college. he inferred that students who achieve in college at a level consistent with past achievement manifest notiyation which will enable them to persist and succeed. Curriculum choice and curriculum change It is logical to assume that there may be a positive relation— siip between curriculum choice and persistence in college. The student who selects a curriculum well suited to his interests and abilities should do better academic work than the student who selects an unsuitable tfor him) major field of study. 17” ,. . . hunger. 'Student PerSistence in College.” . . 18. 19 . . In two different studies, Bragg found a statistically significant relationship between curriculum and both academic de- G ficiency and continuation. Slocum discovered that uncertainty about educational and vocational choice was an important factor in dropouts at the State College of Washington. . Weitz, Clarke. and Jonele investigated the relationship between choice of a major field of study and academic preparation and perfor— manceu 'lne sutjects uumw3 2111 freshumnttdho entered tlm? liberal arts college of Duke University in the fall terms of 1951. 1952. and 1933, and who indicated either an identifiable preference for a major field of study or no preference for both educational and vocational ot— jectives. Three hypotheses were tested: H1 - Students who express a preference for a major field of study in college are better prepared for college than students who express no preference. H2 — When college preparation is held constant. men students wlo express a preference perform better in college ttan those who do not. lbBragg. op. cit. fmma W. Bragg. ”A Pilot Study of Mid-term Marks.” Journal of Higher Education. 28:279—280. Hay. 1957. 20c . Slocum. op. Cit. 2 1 ' II '~ ~ to - y 0 c 0 litenry 1*.81t2, nary Clarke, and tire ._.-ones. Inc Relationsnip Between Choice of a Major Field of Study and Academic Preparation and Performance.” Educational and Psychological measurement. 13:29-38. oanuary. 1955. H3 — Khen college preparation is Leld constant. women who expre“s a preference perform about as well in college as tho 0 who do not. All three hypotheses were supported by the findings: fl) Differences in test performance were in favor of students wto expressed a major preference. For men..mean score differences on tthIACF—L and —11n131; Cooperatixw: anlish. thmjing Comprtdxnision test; ennl the Duke University bknfinnnatics Placemnntt'fest were signi— ficant at tre .Cl level. and tte difference on the Cooperative English. Mechanics of txpression test was significant at tne .65 level. For women, only three differences were significant: ACE—L and (LJopcxwative tkuglistt, Rezxiing (knnprelbensixnt at the. .01 levcd. and ACE Examination — Total score {ACE—T) at the .05 level. .2) A group of 100 men students who expressed a preference were matched on seven levels of placement tcst performance with 109 men who did not express a preference. The same procedure was followed with women students. The men expressing a preference earned a signi— fiCéNltl}7 greuater‘tnnnbet‘ of t5irst-te11n qtualityr poitits. TI£¥ dil“fetkwtce in performance between the female groups was not significant. or Fullmer. in a study of ICES students selected from the fresh- man classes of 1948 and 1949 at the Yniversity of Denver. found that the “declared major“ group had a higher percent81e of graduates than ‘Uaniel N. Fullmer, ”Success and Perseverence of .niversity Students." Jouxnal of Higher tducation. 27;uus_uu7. November, 1936. r __._"_l_ the “undeclared major” group. but he also noted that those who initi— ally stated a preference had a higher index of cFange activity than tlnase whr>vuwwe initially'tnndecided. 131 general. in: concludewl that those who changed objectives were more likely tt persemafgthan hoSe whc>lenu;. arul iifllC€ Iteexl mtntz itnflivithlal. atttnititnt t?.ar1 yrningew‘ students. The studies in which the relationship between age at matri— culation and subsequent college success or failure have been investi— gated yield findings which are contradictory and inconclusive. There . 24 s . H H . John bummerskill. Dropouts from College. Chapter 19 in NevitthSanftnXi ’ed.). 'Nte Anwufican Chillegxz {New qurk: .iohrllfiiley and Sons, lnc.. 1962), p. n31. zaaragg. “A Study of Student Kithdrawal at 'N. “.' A6 , _. — a... are indications that the relatitnhfltip is not a significant one. with some tendency toward a higher attrition rate among older students. The admonition of Summerskill is appropriate. Results attributing a higher attrition rate to the older group should be interpreted with care. Ulder students may have ‘een delayed in entering college for personal or financial or other reasons that pe‘sist and contribute to withdrawal. in fact, Peder showed that a number of students who were de— layed in matriculating had a higher withdrawal rate although their college grades were better than pre— dicted. The general conclusion to be drawn from the litera— ture is that age per se _does not affect attrition although older undergraduates may encounter more obstacles to graduation. A number of studies have revealed marked sex differences in certain areas of academic life. Iffert found that (1) 35.8 percent of the men and 40.5 percent of the women in his sample graduated in regular progression from the institution of first registration. a difference that was not significant.27 .2) Women rated academic reasons as most important in their initial decision to attend college and occupational reasons second in importance. Fen reversed the order. i.e.. occupational reasons were most important and academic 28 . . . . reasons were secondaryn \3) Women and men differed considerably 26. . . :ummerskill. loc. cit. K) x7ffert, op. cit., p. 17. ——-—— ;bid.. p. 101. Y\) ( I” in their reasons for discontinuanceu Txilitary service and financial problems were critical factors in male attrition. while marriage was the major reason for female withdrawal. Concern about academic 29 problems was more common among men than among women. 4) Women were markedly superior to men when rank in high school graduating Class was used as the criterion. However. the differences in College plaCement test performance were negligible.30 A studfnof l3fl4 men and 630 women who entered the Yniversity of Wisconsin as new freshmen in the fall of 1958 revealed that women. compared to men. ranked considera 1y higher in their high school graduating class, did not perform as well on the ACt Examination but earned significantly higher first—semester grades. and more consis- tently achieved the expected classification, i.e.. made normal pro— grwass tcnearxi a chagreC%. Tlmére ken; litt 1e cliffetwntce th the eittril ion rates for men and women. but the latter as a group achieved at a higher grade—point level. Slocum32 ftnnuj that male drtnants at the State (L)llege of Kashington had lower grade averages than female dropouts; 69 percent of the former and only 43 percent of the latter were below the 2.00 grade-point level when they left. The rate of dropout by sex was 31Lins and Pitt, op. cit. 32 Slocum. op. cit. almost identical during the first year. :ut relatively more women than men dropped out in each succeeding year. The causes of withdrawal differed significantly by sex. the men more typically dropping for academic reasons and the women for non—academic reasons. heitz, Llarke. and Jones. in a Duke niversity study. found that substantially greater numbers of men than women entered with no curriculum preference: men with no preference earned significantly lower grades in comparison to men who expressed a choice. whereas curriculum preference was not a critical factor in the achievement level of women: and that women earned higher grades than men. I , . 4 - ‘ 3S . , . .30 halladay and Andrew, Long; and Perry. and drag; conducted research which indicated no statistically significant relationship . . - 37 between sex and perSLstence in college. However, bragg found that the relationship between sex and academic deficiency at mid-term was signifitmNIt beyond tlme .001 level. The research exidence here presented supports the generali— zation “that the withdrawal rate for woven equals that of men because 39 more women withdraw for non—academic reasons. primarily for marriage. 33 Weitz. Clarke. and Jones. op. cit. Lt Halladay and Andrew. op. cit. 3Diouis Long and games D. Perry. “Lortality Study of College Students.“ School and Society. 77:103—105. February 14. 1953. Dr? 'V O 0v 3bBragg. ”A Study of Student Withdrawal at a. 7Rragg. “A Pilot Study of Kid—term Harks.“ 38Summerskill. op. cit.. p. 632. -29- There is considerable evidence that women earn better grades than men both in high school and in college. although men typically score as high (H ltigher (M1 sctujlastic £u3titude euui achiextmmnrt tests. thnnen more consistently achieve at a level commensurate with their ability while.nmn1 are prtnua to ”undezeufltieve.“ llmnw: is some iJuiicatitnt that this may be caused in part by the primarily academic motivation of fe— males and the primarily vocational or occupational motivation of males. Veteran Status Following World War 11. there was sometrepidation in higher education circles ”that a large number of veterans might enroll in school mainly to take advantage of the provisions of the (1.1. Bill, only to drop when the academic going became tough.”39 Most studies which have investigated the relationship between veteran status and college success indicate clearly that this fear was unfounded. Indeed, veterans, as a group, were eminently more successful than non—veterans. Bragg“0 found a relationship between academic deficiency and veteran status that was significant beyond the .01 level. Halladay and Andrew.Z41 upon discovering that only 12 per— cent of Arkansas college dropouts were veterans and only nine percent married. concluded (without the benefit of comparable data for 39 . . ., . ,, . . Lawrence Lipsett and lec>l-.tnniH1, \eterans Carry lhrough on Operation ’hducation‘,” tournal of Educational Research, 42:395, January. 1949. [4ij '0 ‘ '~ .' \ . Bragg. A Pilot Study of hid—term .1arks.’ \. 1, . halladay and Andrew. op. Cit. ~30- non-withdrawals) that both marriage and military service tend to keep 42 at the Rochester Institute of Technology students in school. A study disclosed that the rate of withdrawalvms less for veterans than for non—veterans, although the difference was not statistically significant. One of tlmatnore scufliisticated stinlies of tliis relatitniship . 43 . . . was conducted by Whittemore. he selected from four institutions-- Boston ‘niversity. University of Michigan. Indiana “niversity. and ‘niversity of Minnesota--four samples of 100 male students who had their college careers interrupted by military service. It was found that 72 percent of the 400 continued their education after service. and that 83 percent of these graduated. The graduation rate of veterans was significantly greater than that of non—veterans. Whitte- more concluded that“the interruption of a college program to fulfill a military obligation increases rather than decreases the chance that . . .. . ”44 the sttthIt will entnitually altEHJltllS degree. Focio-economic status There is considerable research evidence to su:stantiate the positive relationship between the persistence of college students and their social standing. as measured by the educational level and I ’) 4("Lipsett and Smith, op. cit.. pp. 395—397. 3 . . . . , . Irving C. Whittemore. ”Does a Military .nterruption necrease the Chances of (btaining a HegreeR.” Fchool and Society. 78:23-27. July 25. 1933. ““1btd.. p. 27. -31- . .. -. u . the occupational status of the parents. mesearch by sihley S Cited in brimub showed "that both the high school student's intelligence and his father's occupation are related to the amount of education he will obtain; namely. that as both rise. the chances of his com— pleting higher grades of education increase. but notably that his in~ teIligence is more closely related to his educational attainment at the secondary lexel or below. and that for achievement in higher educ- ation his father’s occupation is a better predictor than is his own . . ,47 . . - .. intelligence.' Brim also referred to a follow—up study of gifted ,. -. ,._u3 . A ,, . cnildren by Ierman and tden which demonstrated the faCilitating effect of higher social status on educational attainment?”9 an . ' _‘ l effect which held even for persons with 1.9. of 150 and over. olocumSL found that the level of education attained by parents was positively correlated with the academic survival of students. Similarly. there was a higher persistence rate among students whose parents were engaged in higher status occupations. \4 4S- . ,. ,q . .. - . . . . tlbridge bibley, 'bome Jemograpnic Clues to stratification.' American Sociological Review. 7:322—330. 1942. 6 . . _ . -. - . Urville G. Brim. Jr.. boCiology and the tield of Education (New York: Russell Sage Foundation. 1955), p. 37. 47 . Ibid. 48, . ,. .. ,, . _ . ,. . ,. , , Hewis UL. Ierman anul; elita PL t;den. llu:(3ifted (l ild flroys “p: 'I‘wenty—five Years’ Follow—up of a Superior (Iroup {Stanford ~'niversity Press. Stanford. California. 1947). 49 Uz‘int, ItDC. c it. SO” . olocum. op. Cit. -32- The educational ramifications of social status cannot be at- tributed solely to intelligence, a factor known to be related to ,- . . . . i ,, t 51 educational and occupational aspiration. bewell. caller. and >traus conducted a study in which data were obtained from a one-sixth random sample of all non-farm seniors in Wisconsin high schools. l9u7—1948. Four null hypotheses were tested: these may be stated compositely as follows: Among (males, females) there is no significant association between level of .educational. oc- cupational) aspiration and social status when measured intelligence is controlled. Chi square analyses resulted in rejection of the null hypotheses. Therefore, the authors felt justified in formulating the following conclusion: Because the sample was drawn randomly from a broad population of high school seniors fthe entire state of Wisconsin) and because the effects of measured intelligence and sex were controlled. the present tests lend support to the sociological claim that values specific to different status positions are important influences on 1 rels of educational and . . . .) occupational aspiration. Neither can the relationship between social status and attrition be attributed entirely to family income. although the financial pro- blem is a factor in student mortality. lffert found significant 51.. . H 1 . _ . William m. :ewell. Archie t. daller, and surray A. Straus. ”Social Status and Educational and tccupational Aspiration.“ American SoCiological Review. 22:67—73. February. 1957. 521bid.. p. 73. -33- differences in the family income of students compared on the basis of length of attendance in college. The median family inCome of students who persisted to graduation was more than 31000 greater than that of students who dropped during or at the end of the first registration period. The median family income of all graduates was 5437 more than 53 ” . - that of all non—graduates. however. the complete analysis of length of attendance and family income revealed that ”entering and persisting hi college attendance is only nnxherately dependent upon financial - - -_ - _.. . . . i- . - . “54 resources as represented by the single factor of family income. There are. then. values associated with different status posi— tions which account for the relationship between Social status and attrition not explained by intelligence and family income. Summe“skill refers to these values as motivational forces. ”The trou 1e here.” he avers. ”is that hK’Ch)IN)t know what motivatiinual forces are actually predictive of college success and we do not know how to accurately ’ « fl _ . -. . ,,55 assess such motives in students. in summary. a relationship between socio—economic status and ° college attrition has een established. but the nature of this relation- ship is not well known. 53iffert. op. cit.. p.62. 541bid., p. 63. Si . . . oummerskill, op. cit . p. C39. -3u_ Readmission Studies Whereas the literature abounds with validation studies of ad— missions data, there is a distinct lack of reported research which relates directly to the readmission and follow-up of academic failure. 56 , . Dressel conducted a follow—up study of 171 students ’advised to withdraw“ from the Liberal Arts division of Michigan State College over the two-year period, 1936—1938. At that time. Michigan State students who fell below a C-a erage were carefully studied by a com- mittee at the end of each term and were warned, put on probation. ”advised to withdraw.“ or “asked to withdraw.” depending upon the severity of the problem. The “advised to withdraw” action was used when the committee was quite certain that the student was nt;capable of doing college work. A student so advised could ignore the advice if he wished, and was eligible for readmission on probation if he did leave the institution for a time. A student “asked to withdraw” was required to sever relations with the institution forever. About 81 percent of the 171 subjects were freshmen. some were second—year students who had not attained sophomore rating. and only seven students were of junior or senior status. The latter had lengthy records of previous academic difficulties. The ”advised to withdraw“ action was usually, but not always, preceded by action less severe. Q Some of the sample students were 'advised to withdraw“ more than once; a total of 193 actions of this type were taken. C' = ’OPaul l. Hressel. “Liberal Arts Students Advised to Hithdraw.” .ournal of higher qucation. 142h3-45. 'anuary. 1943. Only 34 students actuall)’ withdrew as advised and seven of those returned after one or two terms. There was some indication that the 27 net withdrawals t15.8 percent) were slightly superior students, with a mean grade-point average of .55 compared to a .44 average for those who did not withdraw and with 18 percent in the highest three—tenths in ACK Examination performance compared to 16 percent of the non-withdrawals. Those differences were not statis— tically significant. The students were quite unsuccessful. Ninety r52.6 percent) were eventually “asked to withdraw,“ 11 (6.4 percent) were removed from probation but withdrew voluntarily, five (2.9 percent) were ad— vised to withdraw” again and did so. and 14 (8.2 percent) dropped out while still on probation. A total of 714 scholarship actions were taken against them, 300 prior to the “advised to withdraw" action and 414 after the action. Hf the 171 subjects. only 21 {12.2 percent) graduated and three (1.8 percent) were still enrolled with graduation assumed. Hence. Dressel concluded that the ”advised to withdraw“ action showed little advantage over the "asked to withdraw” action. On the other hand. it could be argued that a "saving” of 14 percent constituted a significant conservation of human resources. -36— I 57 .. ‘ . . . Warman's study at Ohio State YniverSLty involved 234 students who were ”but Ynder Rules” (academic dismissals). petitioned for re— admission to College A or College B. and were referred to the Tniver— sity Counseling and Testing Center 4;"..‘CTC). At. Ohio State l'niversity, applicants for readmission were often referred to the “CTC. am} counselors' reports were reviewed along with materials gathered as evidence in connection with the petitions: statements from employers. doctors. and parents. and academic and activity records. Action on the applications was taken by the petitions committee of the college in which the student wished to enroll. A follow—up disclosed that: {1) (if 77 students readmitted to College A, four (5.2 percent) graduated; 18 {23.3 percent) were still enrolled and doing satisfactory work; and 17 122.1 percent) did satis— factory'twazk (hiring tlua last tennn of (atrollnmntt. 111(3311ege 15. 40 students were readmitted and the corresponding figures were none; four iten percent); and five {12.5 percent), respectively. A greater incidence of non—registration when eligible and of non—academic dis— missals among College 5 students contributed substantially to the relatively inefficient readmission decisions of College B. ’2) [n College A. 26 students (33.8 percent) were dismissed after one quarter and ten '13 percent) were dismissed after two or more quarters. as opposed to 13 (37.5 percent) and ten (25.0 percenth ‘espectively. S7 40y F. Warman, ”A Study of Applicants for Readmission to Colleuyafl' Perscnnuel and (hiidance IHMJrnal. Tfl42553-558. Year. 1956. in College B. (3) bf the 117 students readmitted to both College A and College 8. about one—half were dismissed again. and one—third were lost in only one quarter after readmission. 74) The counselors were accurate in their prediction of successful readmission ’41 of 4%). but they were inaccurate in their prediction of unsuccessful read— mission 521 of 59). Mata presented in Table 2.1 indi ate that hhio State Psycholo- gical Examination (USPE) performance was weighted quite heavily by counselors and by petition committees in making readmission decisions. however, the relationship between ()SPl-L scores and academic record after readmission was not significant. Harman concluded ”that tested scholastic ability was given too much weight in these cases...hther probably less objecti.e factors are particularly important and need to be brought into sharper focus when dismissed students are being Considered for readmission to...the 'niversity.”58 Campbell and dahn59 studied the relationship between grade— point index improvement after readmission and three variables: fil) activity engaged in while away from College. ’2) length of absence from college. and '3) high school English background as measured by Knglish-Total score on the Cooperative College Entrance Examination. Data employed in the in estigation were obtained from the records 58 . r Jbid.. p. 538. 59- n s - n - ' ‘ ~ L. howard Campbell and Walter Hahn. nkadmljSLUn of tormer dttnhuits afttn‘1Absence fkunn the (Rumpus: Pttddxnus and {tmx)rtunitics. College and ”niversity. 37:126-134. Winter. 19(2. o ..oonHCU ow :chmHEcmo: Low mochWHo1< we xenon < .cmOH .sm: .ammnsm .:mEom3 .. .flmceDCfi .,.._ >6...“ :w m Samba—5.5 ~6me oo:mUH:: Csm #zzzsneom H ._:¥woom mfloyoflEszxo mscflflfiwdoa He. c3. . H3. wd. ozoflofiumooU “cm tam mflmocgoem uofimmssoo .:0HmmflEvm sou boomm Uuooou oHEosmom osmoflwwszm ac: Co. No. mm. ozowowmuooo H:m ©:m mwmozwoea soaomczou #:mfiofiwmooU :oqo .:0qmmetmme eooum secomu osmofiuwzmwm so: mm. osmowmwcwwm #3: ON. nmfioeeoo Hmfieomflz oHEobmom U:m mewomp :mmc ozowoflwwooo cow“ .Ammmmo Hfimv mfimcszea Ho. em. Ho. mm. anoLLOo Hmflemmflm ecflomczoo wzm mzfiome 3mm: .Acmoom omooflaaoo “zofloflwwoco :cflo :oH;3 cog: mommoo mflmccmoen _ . 8 Ho. me. H0. H3. tmfioeeoo Hmfiuwmflm uoamm:30o Usm mcfiome mam: 3 . _ .cofioom mooeflEECU mzofloflooo Eowowwumco so: wzm wit“... Cams: :OA. omzfizm Ho. m3. Ho. mm. umfloeeoo Hmfluomqm xx: Hmoflwcfiozozmm oomem 3H2: oo::oflwwcwflm e oocmowuwcmflm ; tom; «0 H$>QA mi m_”oa_,_._fiouv w: H$>m4 xx m neu*.~-_AomV mofiemqomom mwfinwfiemy :onmHEUmoz 33H. 5:. ZCHmmHZC<;z ~.r~_mmfr/Azs or~<9wwflve:: e2. mgnfifif~20 2:; WEZ2 Em ZH :zmg ~.N 2;:flr m;;m ”ozixaxe WZAerzaiezvroowirHZg -_-"I of 538 students at the tniversity of .tah who were placed on pro- batixnt. wetwa abserni front the ittiverisity ftu“ at ‘least.t)ne (piarter‘, and who were readmitted sometime between 1952 and 1959. The principal findings are summarized in Table 2.2. All groups showed suostantial gains in mean grade—point average after readmission. but there was considerable individual variation in the academic per- formance of the readmitted students. Some far exceeded the average improvement while others actually performed at a 1rmer level. Fur- ther, the rate of post-readmission success was not given for the total sample or the sub—groups. Results of the Ytah study prompted the following conclusions: ’1) The activity of students while away from college appeared to make an important difference in the quality of academic performance after readmission. Successful work at another institution and military service proved to be especially valuable interim experience. ’2) “Other things being equal... one would be justified in favoring readmission of a student with two or more years of absence over that of a student who had interrupted his college career for a short time only.”60 (3) There was a positive correlation Letweer English placement test performance and grade-point improvement after readmission. regardless of activities engaged in by students during their absence “gig” p. 131. TABLE 2.2 RELATIONSFIP or REAnslrtrn ST DENTS' CRANK-P0131 AVERAGE QEFHHE AND AFTER ArSENCE rRon THE UNIVERSllY or UTAH ACCORDING To ACTIVITY ENCAGLD 13 whitr AWAY FRow COLLEGE, thcrn or ARSEVCE raon CHLLHGE. AND PERFHHEANCE ox 1h£ ENGLlSH PLAchrxl‘irstl N CPA Before CPA After Improae— Withdrawal Readmission ment in GPA All students engaged in “actiyi- 392 1.416 2.374 0.958 ties of import“ while away from college Vilitary service 193 1.384 2.332 0.948 Military service and church 29 1.411 2.558 1.147 mission work Military service and good aca- 12 1.311 2.435 1.124 demic work at other institu- tion Military service and change in 12 1.528 2.360 0.832 marital status Good academic work at other 33 1.341 2.437 1.096 institution Church mission work 64 1.510 2.385 0.875 Change in marital status 26 1.459 2.284 0.825 Responsible employment 18 1.533 2.460 0.927 All students not engaged in ”acti- 146 1.411 2.156 0.745 vities of import” while away from college All students who were readmitted 538 1.415 2.315 0.900 Number of quarters away-—students engaged in “activities of import” 0ne quarter 11 1.595 2.301 0.70h Two quarters 15 1.525 2.133 0.608 Three quarters 10 1.396 2.244 0.848 Four quarters 11 1.588 2.1r8 0.550 Five quarters 6 1.435 2.248 0. Six quarters or more 339 1.401 2.400 0.999 1Taken from Tables 1. 11. and 111 in L. Howard Campbell and Walter Hahn. “Readmission of Former Students after Absence from the Campus: Problems and 0pportunities.“ College and ’nirersity. 37:130- 132. Minter. 1962. _u1_ TABLE 2.2 (Continued) K GPA Before GPA After Improve— Withdrawal Readmission ment in GPA Number of quarters away--students not engaged in “activities of import” Hne quarter 12 1.413 1.993 0.580 Two quarters 30 1.449 2.118 0.669 Three quarters 14 1.110 1.638 0.528 Four quarters 25 1.452 2.254 0.802 Five quarters 4 0.995 2.301 1.306 Six quarters or more 61 1.471 2.276 0.805 English Placement Test rating—-stu- dents engaged in “activities of import” 32nd percentile and below 129 1.432 2.277 0.845 33rd percentile through 66th 126 1.359 2.373 1.014 percentile 67th percentile and abore 58 1.493 2.538 1.045 English Placement Test rating-—stu— dents not engaged in “activities of iflnxart” 32nd percentile and below 36 1.372 2.007 0.635 33rd percentile through 66th 50 1.433 2.177 0.744 percentile 67th percentile and aboxe 39 1.407 2.210 0. (D 0 LA) from school. (4) There was a definite likelihood of academic im— provement after returning. Therefore. in the opinion of the authors. it would be unwise to bar from return arbitrarily all students whose previous grades were unsatisfactory. (5) The individual dispersion of grade-point averages which was observed would make it inadviseable to permit automatic readmission of all such students. (6) Inasmuch as objective data did not indicate with sufficient clarity which stu— dents will. and which will not. improve after readmission. decisions must be based on all pertinent circumstances involved in each case. including non-academic, subjective data. Sununaryr There is a vast body of research literature dealing with the relationship between college achievement and persistence on the one dand and multifarious academic and non—academic factors on the other. Although the results of these investigations are not perfectly con- sistent. there is sufficient agreement to justify the assumption of a substantial positive relationship when high school grades, standar— dized test scores. first—term grades in college. curriculum choice. curriculum change. veteran status. occupation of father. and educat— ional level of parents are employed as predictors. There is some consensus that older students and female students achieve commensurate with their ability, but that non-academic obstacles cause their mort— ality rate to equal or exceed that of younger students and males. -43- The findings gleaned from the research literature afforded a rationale for the research hypotheses stated in Chapter 1. It was assumed that factors which contribute to student success or failure prior to withdrawal may also be instrumental in success or failure after readmission reinstatement. Other researchers have recognized that “in order to deal realistically with readmission applications. the causes for discontinuance of college attendance must be'well understood."61 Only five studies were discovered which dealt directly with the readmission problem. The rate of success among readmitted stu- dents is generally quite low; it is not unusual for 50 to 75 percent to fail. However, this attrition rate can be lessened to some degree if applicants for readmission are carefully screened individually on the basis of both objective and subjective, academic and non-academic data. There is some indication that readmission decisions based on academic factors alone are subject to considerable error, and are somewhat more effective when information non-academic in nature is introduced. The studies reviewed in this chapter were subject to various methodological limitations. .‘fiany of the investigations were not aimed toward the testing of stated hypotheses. but were survey studies of certain groups, e.g., academic failures. non—academic withdrawals. 611mm,, p. 127. -44- readmitted students. Too often. the group surveyed was described without reference to a comparison or norm group, a shortcoming which rendered the results difficult to evaluate. Almost universally. the studies which established relationships did not inquire into the nature of the relationship. This is a major limitation from the Point of . 62 View of research method. but Campbell and Hahn defend the practical value of such findin;s by suggesting that for the readmission officer the causal relationship is not as important as knowing to what extent the factors point to improved academic work after readmission. Another limitation. also widespread, was the failure to study the interrela- tionships of the various independent variables which were related to college achievement and persistence. CHAPTER III DESIGN OF THE STFDY The design of the study entailed the selection of the sample and four comparison groups from among the individuals who enrolled in Michigan State University as new freshmen in the fall of 1956. Data were collected, codified, and punched into cards to facilitate the use of an electronic computer. Several statistical methods were employed in data analysis: the t-test of significance, the chi square test of significance, analysis of variance, biserial corre— lation, and Pearson product-moment correlation. The Null Hypotheses The research hypotheses for the present study were stated in Chapter 1. These were formulated into null hypotheses to be tested statistically. H01 - There is no difference in the proportions of males and females who are successful after re- admission or reintstatement. H02 - There is no difference in the proportions of veterans and non—veterans who are successful after readmission or reinstatement. -45- -46— H - There is no difference in high school grade 03 , . average between students who are successful or unsuccessful after readmission or rein- statement. H04 — There is no mean age differential between students who are successful or unsuccessful after readmission or reinstatement. H - There are no differences in entrance test 05 ~ scores for students who are successful or unsuccessful after readmission or reinstate— ment. HO — There is no difference in the fi“st—quarter 6 grade—point averages of students who are suc— cessful or unsuccessful after readmission or reintstatement. H07 — There are no differences in socio—economic status between students who are successful or unsuccessful after readmission or rein— statement. The Sample The subjects for the investigation were selected from among those individuals who enrolled at Michigan State University in the fall quarter, 1956. The sample consisted of 213 men and 81 women who (1) entered Michigan State Kniversity as new freshmen, (2) were -47- withdrawn by action of the t'niversity College due to deficient schol— arship between September, 1956 and June, 1958, and (3) who wert re- admitted or reinstated subsequent to their involuntary withdrawal. Only those individuals readmitted or reinstated as of the fall quarter, 1959, were included. The academic progress of these students was studied until their graduation, termination, or through the spring quarter of 1960, The total sample of 294 students divided into four sub—samples (1) 90 men who were successful after readmission or reinstatement, (2) 35 women who were successful after readmission or reinstatement, (3) 123 men who were unsuccessful after readmission or reinstatement, and (4) 46 women who were unsuccessful after readmission or reinstate- ment. Delineation of the sample With the exception of 12 married men (five successful and seven unsuccessful), all subjects were of single marital status at the time of college matriculation in September, 1956. Forty-—one male subjects had served in the military prior to their enrollment. The varying degrees of college accomplishment achieved by the 1 of the sample are summarized in Table 3.1. Fifty-nine percent subjects (174 of 294) terminated their attendance at Michigan State Yniversity while still in the Kniversity College, 52 percent (153 of 294) with a grade index below 2.00. One—third of the sample (98 of 1 Reported percentages are rounded to the nearest whole number. oo.mo 30m oo.oc~ o: oo.oo~ mm oo.oo~ mma Ho.coH oo mqm ocqoaaocmew dc.mm mmH mc.mo a: mo.mw 00H .mjomom mmofifioo zowmoo>flcs N .2 3 L W h C oo.m us wmmoo>m ocwoanwpmom es.“ Hm oe.es e Kc.cs ms .mssmsm mwmssco ssdmoasaso oo.m 33Hmn vwmom>m pCHOQ dd.m 9H mm.:. N mm.- :H nmompw .msumom owwaaoo soda; & o>com .4 ho oo.m w: mwmoi>m ocfloo mm.mm mm om.mo mm No.50 oo umomow .msomom wwwfiflou smog; :oflmmmew m:.m mm oo.om m mo.©H ma iota emaswmo :w poemstmoc PM... Z «we 7. at 7 no Z ...L Z Abduzo Ammnzv Ammanzv ”Gonzo 52:2:m_gmzcou< Aqomuzv Hammmmoozmc; Hammmmoosw Haummwoojmc; asmmmmoozm 24m22 H . m 21:2... 2..._ZHeom ,trzeosoea 15 wzm mas. o dm.© .H Nd.o H Hm.o H uononm: pch omooH>omm 3m.c H Nd.o H Hm.o H osmEm>OLQEH poemsmoewm o~.H m oh.m m mm.e m 30.0 N me.s m mmmssco mmHoon : Hume oHC: dd.m 0H mm.H H om.m H do.m mH :0.w HH 33.: : ecm mcHomoe Hoocom :wHL Seem mmHocmHonmt Homooma oomw :w.w cm mm.H H NH.N H 3m.HH mm m:.mH 0H No.0 0 nesme oHCQ mmzHomoo we uHsmmo cc.ccH 30m 00.00 H . & mo.mH mm mm.mH HH 0m.mH m mm.@ m mm.0H N: mo.mH Hm mm.mm Hm mm emooH87< Hm.mo NOH mm.cm mo H0.mm 3m mm.wm Hm No.0m mmH mm.mm mo om.mo 0m mzez r. Z n.3, a, 7..H surf Z Wu Z rm»: x cm ., Z LPN“ z . . . mZLHHCZCD z: Hd0muzv HHw n20 Hednzv Ammuzv HmHmuzo Hmmanv Ho0u20 meHuégHUHssz ¥H&2H:; momom :meonz ocm Hcozom :wHL HmooH may :mm3omn ocosomowm cm as mHmmn m:o co mmeHOU 0o nouoHEUm mmz ozovsom oz“ omzo moHMszHm n .mcoHpmcHmem UoNHUemwcmom :HmHooo :3 mocmELOWme 2o0uomumHHmm we ozooH> an omoHHOc ob :onmHEwm emchw ecm mmemom Hooxom 50H: wemnszmunzm rm: ocmezom mzH omzo mmHMchH Wm. 82:: s :08 m .m ”Ema: a— L b4.. Most of the subjects enrolled in a four—year degree program. Six women and three men entered the University College on a two—year terminal program, and seven women and three men were candidates for a provisional teaching certificate. The “average" Michigan State Unive‘sity career of a student in the sample spanned eight quarters (mean = 8.1) of enrollment, in— cluding summer sessions. The mean number of quarters enrolled was 11.1 for successful men, 6.3 for unsuccessful men, 10.8 for success— . ful women, and 5.0 quarters for unsuccessful women. 0n the average, unsuccessful sub—samples experienced academic withdrawal sooner than successful males and females (t = 5.349, df = 211 and t = 4.463, df = 79, respectively, p <:,01). Eight successful men, 33 unsuccessful men, two successful women, and 12 unsuccessful women were withdrawn at the end of their first quarter in college. The number of unsuccessful students withdrawn at the end of the fi‘st quarter was significantly greater than the corresponding number of successful students ()(2 = 16.41, p <<,Ol). As indicated in Table 3.4,the sample of 294 students incurred a total of 1,948 academic scholarship actions? including 497 with- drawal actions. Some individuals were involuntarily withdrawn only once and were the recipients of but two academic scholarship actions totally; others were withdrawn as many as six times and incurred a total of 17 scholarship actions. 3Actions incurred by students in academic difficulty, e.2 -" placed on probation, continued on probation, removed from probation, withdrawn, reinstated on probation, readmitted on probation, final warning 1, final warning II, transferred to unclassified. . hind“ .UoHuHmmmHocz 0o emoemum:meo .HH wzHcomB Hmsz .H w:H:em3 Hmch .coHHmDOoo :o moHpHeemoo .coHuwQOoa :0 emumomszo ”:3msesz3 .:oHomooe0 Ecow eo>oewc .chomncbm :o noszHozoo .onHmQOoa co coomHa ..w.m .szsonmwv oHEoemom CH musmczom so woousocH monoo0 0....Ex..~UZH .mZCHhUsq. ‘H<3<~.:::.H3 QZHHEHUZH mmZA.0H.H.D< mH:.f.N._/.30:Um OHZV._:ewmcmswcp otmpm :mwweodz occom sms.om enm.m Hm mmo.cm mHH ems.:a Hm moe.mH meHmase stmssemucs - Emma weoemme Hos.mm HHe.m mHm oem.mH mes mem.mH mom Neo.mH meHmz stwmsasoz: ssmfm :meequ accom HmHCH I :oHom: em\.cm ehm.m Hm omm.:HH mHH mmo.oo Hm mes.mo mmHmEme -Hgmx; Hmowwcsoeosmm cows moe.om HHe.N mHN eec.mHs mes “HH.Nm mHm sme.mo mmHmz -musee :o Hooesoo smoocse< meoom zeHHHn< owumstcHH : :OHHm: mem._m emm.m Hm mwo.om msH ssm.mm 3m mmH.sm maHQEse idsmx; smoouofioeosma cows mem.oe HHe.m mHm mmq.eo ems omm.em mom HeH.em myHmz -musee so Hwoesoo :muwsmee afoom sowswse . o>HpmoHu:m:O I ccHHm: eeo.mH :NN.N Hm mom.ma mHH mem.mm Hm «mm.em mmHmsos -Hgmxo Hmoomcsoeusmm code MNH.oH HHe.N mHm mee.me smH omm.sm mum mm:.o: msHmz -moseo :o kucsco :mowtmgm mocom umob ems.em ssm.m Hm NHO.:m mHH eme.mm Hm mNH.om monays soeooowoc.m ooomeesou< ch.mm HHe.m mHm mmm.tm me eco.Hm mHm osm.mm memz mwsHsoo memsm emwweudz esm.sm :em.m Hm Hee.0m mHH emh.es Hm Hmo.eH moneae steam swap emMHweo mmm.em HHe.m mHm mmH.RH ems oas.mH mom Ham.eH mmHmz seemeeswes womem seaweed: .m on m: me we me He Hx mozzmoH Ho. om u:mofl0H:me mosHm>IL HHwuomAmoo .m0200w :quHB moomzvm we 82m ecm mQDOem :ooBoon mmumjvm we 93m so“ Ecwooum mo mmoowmw H mm m 030ew :H copes: as“ u m: N 030cm :H doses: 05o n N: H azoew :H ooDED: ozw u H: Am QSOewV HmBmoeonB oHeoUmom wcwoonoooxo osoonB :onmoewOLQ cmstoo CH wcmezemow mucoespm ecu moooum cofleouwco some n mx Hm asocwo empmom % IcHo; Lo eoooHEUmmo us: mHmBmoeonB UHEmcmom ecu moocom :OHLmoHco :moe u mx 1) ll _ H H nw~s.o,_mw V GeomemcHoo e0 tooHHEUmmp mHmchezqu oHsocmom.e00 mmeOom :mHeouHoo some n Hx H 3030 , mmeHou HJN.HQH dnm.m Hm m0n.m mHH mum.H Hm mm©.H monsoL CH cmwcmsv omeHm mo owe.sem HHe.N mHN oom.m mmH ohm.H mHN mem.H thmz new .mwmttsm oCHOQIGUmeo KHH.Nm :mm.m Hm dom.md mHH 0M0.Hm Hm mmw.dm monEmw mucom HmoOH I ummh weaemmz 0nd.o: HH©.m MHN 0mm.o: me. mmm.Hm mHm 03H.mm mmHmZ zuHmom>sz swmpm cmenon \ me me mx m: m H: H e mmmDm and the null hypothesis is accepted. Null Hypothesis: There is no difference in the proportions of veterans and non—veterans who are success— ful after readmission or reinstatement. -70- -71- For this particular sample, 51 percent (21 of 41) of the veterans and #0 percent (69 of 172) of the non—veterans were successful after readmission or reinstatement. The difference in the proportions of veterans and non-veterans who were successful was not significant ()(2 = 1.677, p:> .05), and the null hypothesis is accepted. Null Hypothesis: There is no difference in high school grade average between students who are successful or unsuCCessful after readmission or rein- statement. Quarter rank in high school graduating class for 256 subjects is summarized in Table 3.2, p. 51. Sixty-one percent (19 of 31) of the male subjects who ranked in the first quarter were successful com- pared with 33 percent (Ufl of 30) of those in the bottom quarter who were successful, a significant percentage difference (7(2 = 4.786, p (:05). The chi square test of differences in high scnool rank for successful and unsuccessful females was not significant (jflz = 1.398, df = 2, p:>.OS). The mean quarter rank in high school graduating class was cal- culated for successful men (2.3), unsuccessful men (2.6), successful women (1.9), and unsuccessful women (2.2). On the average, successful subjects earned better high school grades than unsuccessful subjects, but the mean high school ranks were not significantly different for either male or female sub—samples (t = 1.766, df = 180 and t = 1.435, df = 72, respectively, p> .05). -72- The numbers of successful and unsuccessful students who ranked in the upper and lower one—half of their high school class were not significantly different for men or for women 7:?(:l). For the entire sample of 256 men and women, the rate of suCCess by quarter rank in high school graduating class was: first quarter, 58 percent (29 of 50); second quarter, 39 percent (37 of 96); third quarter, 42 percent (32 of 77); fourth quarter,30 percent (ten of 33). The subjects who ranked in the fi‘st quarter were significantly more successful than those who ranked in the fourth quarter (7(2 = 6.120, p (.02). The null hypothesis of no difference in the high school achieve- ment of successful and unsuccessful sub—samples is supported by three tests of significance and is rejected on the basis of two tests. Null Hypothesis: There is no mean age differential between students who are successful or unsuccessful after readmission or reinstatement. Data regarding the age of the sample at the time of their college matriculation are presented in Table 4.1. The male subjects (mean age = 19.3) uere about one year older, on the average, than female subjects (mean age = 18.2). Mean age differences between successful men and successful women and between unsuccessful men and unsuccessful women were both significant (t = 5.991, df = 123 and t = 5.059, df = 167, res- pectively, p <:.01). The ages of the men were significantly more variable than those of the women (F = 8.995; df = 212, 80; p <<,01), and age variability was somewhat greater among unsuccessful sub-samples, male and female, than among successful sub—samples (Y = 1.204; df = 122, 89; 13> .05, and r = 1.616; as : as, 3n; p> .05, respectively). -73.. TABLE 4,1 MILAN AGE AT COLLEGE MA'I'RICII‘IATIOX 0F SI'C(_TLLSSI“1§L AND lfIx’SlTCCIISSFKL l-ll’INS'I'A'Il‘ZI‘) NAMES AND FEMALES REA J )MI T'I'} II) (JR MEAN ACE STANDARD I)l:\-"[A'I‘l(h\‘ Successful Men (N290) Unsuccessful Men (N2123) All Men (N=213) Successful Women (N235) Lnsuccessful Women (N246) All Women (Nle) 19. l9. 19. 18. 18. 378 244 .399 .507 .466 -74- The mean age differences between successful and unsuccessful males (t = .489, df = 211, p:> .03) and females (t = 1.010, df = 79, 1):} .05) were not statistically significant. Therefore, the null hypothesis is accepted. Null Hypothesis: There are no differences in entrance test scores for students who are successful or unsuccessful after readmission or reinstate— ment. The college entrance test performance of the sample is summarized in Table Q.2 Lnglish, ACE—L, Reading—V, Reading-C, and Reading-T scores were higher for females than for males. The men earned higher Arithmetic, ACE-Q, and ACE-T scores. In general, then, female subjects exhibited relatively superior verbal ability and the men excelled in quantitative ability. Without exception, successful students had higher entrance test scores than unsuccessful students. For successful and unsuccess- ful men, there were significant differences in English (t = 2.010, df = 211, p <m2 Utmtcm-m fim-oancoflomCM .Lemxm fimunOHnZfiozmm :3“- mmm.mo mmm.mo Hmfi.oofi www.mo Hmo.mm HHH.H©H mesom Bmx :mm? roost; :3 Hwo:sou cmoweee< e-HH-n< 033.3H ddm.dH m~m.da 0mm.mH ~0m.3~ oom.c~ :oflomflye: tswccmom oH-m-swzwwncowomcfl :mem Hmuflwoaozozmm :ofio mmfi.0m mflm.mm omc.Ho HJH.Nm Hfim.mm nmx.om oecom 3mg :mmz umozcg :o Hmozsou :mowcoa< >--n< mmm.oH moo.oH mco.oH ooo.oH Nmo.HH mmo.ofl soflomfl>wc tcmocw-m o>HumoHpsmsoucoHomcH nsmx: Hmowwofiosozmm cog“ mmw.om mom.mm mdm.mm mnq.os o~m.om mmm.H: mucom 3m: :mwz twoscu :o Hflocsou :mowcm5< mmm.m oc~.o :mw.: oym.m oma.m oo:.m :oH-m-wm: pemozm-m omww mozofloflwocm oHowE:-Hc< mR-.cm mma.mm oom.-m osm.mm omm.-m Rom.mm mecca 3me see: emu-Hoe m-mum ewe-sow; moo.d omo.d omo.a mmm.d cod.d 0mm.d :oHomH>m2 tumpzm-m “mob :mHHwC; Hmo.©~ omm.c~ Hum.m~ dom.:fi www.mfl Nmo.mH mLOom Bmz cmmz %owmew>wcp m-mom cmmwzu-Z Aflwuzv Aodnzv Ammuz- Amflmuzv Ammauzo Aomuzv 3 move 3 (a. H m , m m mennwo HDW “nub-0.1.5 q.-Uz<....-,~.z: fimcoe Iosmze ummwoosw Hm-oe nozmc; ummwocsm 2:223 2;: mmu<22n :z< m;qu-:U< :ozm: mwd.nH d:o.m~ meowm ofim.o mac.m :owomw>m: dmo.dH mm:.o~ mucom >-Hmum>fl:- m-mom Cemfizm-m Icozmemsoolpmwe zpwmuv>wzs momum newts-mom >Hcmfisnmoo>I-mmk zowmcw>fi:; mompm wzfismmz :mmwzo-Z :on wzwemmm cmwweo-z mcflomox gem-eU-z Aodnzv Ammnzv stmmmo Ham nozmz; Immmoosw 7mm_vafiqw- Ammfinzv Acouz- Haummwu Haw looms: ummmuosm zmca ;02x$+rzm Cow-:3. each: m . d ”-125. _77_ Null Hypothesis: There is no difference in the first- quarter grade—point averages of students who are successful or unsuccessful after readmission or reinstatement. College achievement indexes of the sample at five stages of their college careet are given in Table 4.3. The first-quarter grade- point average means for successful and unsuCCessful males and females differed significantly (t = 6.375, df = 211 and t = 4.235,df = 79, respectively, p (’01). Hence, the null hypothesis is rejected. ‘ Also significant at the .01 level were the mean differences in cumulative grade—point aVerages for successful and unsuccessful men and women at the end of the quarter of initial withdrawal (t = 7.086, df = 211 and t = 5.435, df = 79, respectively), at the end of the quarter of initial readmission or reinstatement (t = 8.755, df = 205 and t = 6.926, df = 74, respectively), after initial readmission or reinstate- ment (t = 11.625, df = 205 and t = 7.817, df = 74, respectively), and at graduation, termination, or the end of the spring quarter of 1960 (t = 16.718, df = 211 and t = 20.500, df = 79, respectively). It should be observed that all sub-samples significantly im- proved their cumulative grade-point index during the one quarter after initial readmission or reinstatement. The amount of improvement was: successful men, .303 (t = 9.6, df = 178, p <:.01); unsuccessful men, .285 (t = 5.8, df = 232, p 4:.01); successful women. .234 (t = 4,9 2 df = 66, p (:lfl); unsuccessful women, .225 (t ; 3,2,df 2 82, p (5,01); .mtioa .5 manco o: Cozumo vsm ozezaomumzflou so :cflmmHEUmo; owozo wsfi30HHOm zfimomflcwssfl emuumzv may wzflcso mowmuw>fl:; comom :mmfiLUflZ Seem Bocncofiz zfiflpmoczac> 2mzo wmsmomn UmUJHoxm oom3 :mEoB m>flm ficw cos XHm :Nm.H Hm oom.a ,eN mom.a em o©:.H on mmo.~ Hm eweoz afi< m0m.~ o: m~d.~ m: :0:.~ m: mom.~ o: mod.H o: :mscz HawmmeQSm:; m3m.m mm dc:.m 3m moo.H 3m Hmm.H mm oom.H mm :6503 Hammmooesm omm.~ mflm “mm.fl New mmm.~ .Nom mm:.H mfim mum.~ Mom cm: HHAN mom.a mNH mum.H “My oom.a NHH Hmm.a mmfi mmm.H mmH :mz Haemmouusmzs mam.m so oflm.m co moo.H so moo.~ oo Nom.H om cw: Hzommmousm e... 7 _ < ez~bXTlie_;>WH:¢;U m. d 2.322. -79- The amount of grade-tn3h1t improvement effected dudeu; the entire period after initial readmission or reinstatement was also significwtt: successful men, .523 (t = 12.4, df = 178, p <:.01); unsuccessful men, .282 (t = 6.5, df = 24a, p <<.01); succe sful women, .512 (t = 7.8, df = U) 68, p <<.01>; unsuccessful women, .236 (t = 4.3, df = 90, p (CADI). It is evident that both successful and unsuccessful sub-samples raised their grades substantially after readmission or reinstatement. However, the unsuccessful students, with low grade—point indexes at the time of withdrawal, were unable to attain the grade—point average required for graduation. Null Hypothesis: There are no differences in socio-economic status between students who are successful or unsuccessful after readmission or rein- statement. Three indexes of the socio—economic status of the sample were employed in the present study: (1) occupation of the father, (2) education of the father and (3) education of the mother. Occupational status of the fathers of the sample is summarized in Table 4.4. Fifty—three percent (131 of 245) occupied positions in the median occupational group composed of skilled workers, small businessmen, and higher-level white—collar workers. More than one— fourth of the fathers (65 of 245, or 27 percent) held higher status professional, semi—professional, or administrative positions (groups 1 c:.oo~ mdm oo.oo~ mo oo.oo~ mm oc.oo~ mm 00.0@ RAH co.oe~ MOM o©.om 3K m4mflueozwflz ”mummmcflmsL HflmEm m3 maoz3c ”memes m:.mm HmH mm.mm o: 3H.mm 0m Ho.o© 0m H:.Hm Ho Hm.©m mm om.:q mm ILmH Hmssz owfifiqu _ mcoflomasuoo % 0.3 F... 7% 52:5 _ Ho>oHIEDHUmE nzm :~.mH mm m©.m~ NH 3H.mfl o w~.m~ c NH.3H mm wd.NH mH o:.o N HmchmmouooonHEom mSCH H Imozooo oywumeumfl u:HETm ~o>manooooz m:.HH mm mm.cH m m:.HH d 00.0 m ow.HH Hm Hm.m CH cw.d~ HH Uzm fimcoflmeMCLQ a 2 fl 2 J z W z i z U z 6 2 Am. lane/Hy ”Z A ”2 H a H H H « 3c Awe V mm /v Amm 7e ARNH zo Amofi zv Adm 7e mm;CzQ q>> .05). The chi square tests of differences in the occupational status of fathers of successful and unsuccessful males (7(2 = 5.344, df = 2, p) .05)1 and females (X2 (1) were also not significant. . The level of education attained by the parents is summarized in Table 4.5 and Table 4.6. The mean educational-level ratings of fathers by sub—sample were: successful men, 4.5; unsuccessful men, 4.6; succ- essful women, 3.7; unsuccessful women, 4.1. The mean ratings of fathers of successful and unsuccessful males and females were not significantly different (t = .477, df = 196 ani t = 1.098, df = 74, respectively, 1);) .05). The mean educational—level ratings of mothers by sub—sample were 4.7 for both successful men and unsuccessful men, and 4.3 for both successful women and unsuccessful women. 1The non—significant tendency was for male subjects with fathers in group 4 and group 5 occupations to have a higher success rate (23 of 40, or 58 percent) than men whose fathers were engaged in group 1 and group 2 occupations (18 of 56, or 39 percent) or in group 3 occupa— tions (33 of 91, or 36 percent). oo.oo oHH O0.00H mm mg<+:9 . CD 0‘» v—-¢ oo.oo 3mm oo.oo on Ho.ooH d: mo.oo mm Ho.ooH HmmmH cc memaz szHmv :oHomo: 00.:H o: mm.cs m so.mH N N~.m H 02.0H mm mm.oH as. mm.mH ms -em acceum eaw;-otm momstmcmuzo: Hd.mH 3m Hm.o m mm.d m m©.mH m :©.mH mm mo.mH :H mm.mH mHnHoczom :ch poczuo~< dm.mm mo mo.mm Hm $0.3m mH mm.mH o om.©m mm ©©.om o: Hm.Hm om momjvmsw Hoesom :uH: wesHm: otcwwtrco: m we :ijuzpw mm.m m cm.m : mm.3 m mm.© m mm.H m NN.H m NN.H H Hoesom :chupmom L. mmHmspmuwnzc: flw do.mH mm Nd.:H HH mm.o m oo.mm w HH.HH mm 3m.oH NH om.mH OH nwmeHoU newswoo< co.Hm ow om.om Mm mw.Hm 3H NH.xm o oo.mH mm OH.wH Hm Hm.oH. cH womssmew mmeHoo . u:me I>Hsvm muH be cmcw mm.m m mo.m m km.m H NH.m H mm.m m mm.H m co.m m new momsvmemuemcm .w 2 fl 2 9 2 fl 2 f z 6 Z w, z Hammnzv Homnzo Asquzc Hmmuzv HmoHuzc AcHanv Ammnzv mgz>gq Hq.::o< ZCH.rH,._1H m. . d ..._1Hm<.H. -83— Hc.coH mom Hc.ooH ox oo.ooH m: oo.ooH mm 00.9: HoH oo.ooH NHH oo.coH om mHH3Um moH co mwcw mm.o N mo.H m om.o H Rm.s a .me msmzsmcm-umsm fl 2 P z w e z . z a z r z Aweeuzs Acmuzc Amsuzv immuzo ismsuzs Amssuzo gawnzo mos>gs s...:_HU< “H: 4 ”+5: -84- The average father or mother was a high school graduate who had some education past the secondary level. The 274 fathers, as a group, had a mean educational—level rating of 4.4 compared to a mean rating of 4.6 for the 267 mothers. The parents of female subjects had more formal education (mean rating = 4.1) than the parents of male subjects (mean rating = 4.6). On the basis of criteria employed in the present study, there were no significant differences in social status between successful and unsuccessful men and women. Therefore, the null hypothesis is accepted. Predicting Post—Readmission or Reinstatement Success The efficacy of ten academic predictor variables, eight test score and two achievement measures, was investigated. Coefficients of biserial correlation between each of the predictor variables and the criterion, membership in the successful sub—sample, were calculated. The results are reported in Table 4.7. Pearson product-moment cor- relation coefficients were computed between each of the ten independent variables and the two dependent variables, final cumulative grade- point average and grade-point average after initial readmission or reinstatement, exclusive of repeated courses. The results are pre- sented in Table 4.8 and Table 4.9. _S)_ TABLE 4.7 CUEFFICIENT 0F BISERIAL CURRELATIUN ACHIEVEMENT AND APTITVDE HSTlEAlES (1F "111E. CRI’HLRION, SL'CCIZSS AFTER i-ZEALWISSION ("R REIXSTATIDIL‘NT , . -. i . HEN whim TNDLPLRDLNT VARIABLES N (32213) (N281) Michigan State tniversity .173b .192 English Test Score . . . . b b Michigan Statt College Arithmetic .191 .307 Proficiency Test Score American Council on Education Psy- .084 .176 chological Examination-Quantitative Ability Score . . . - b American CounCil on Education Psy- .186 .191 chological Examination Linguistic Ability Score American Council on qucation Psy- .161 .203 chological Examination—Total Score Michigan State Tniversity Reading .208b .200 Test—Vocabularly Score Michigan State tniversity Reading .160 .099 Test—Comprehension Score . . . . . . b , Michigan State tniverSity Reading .211 .169 Test-Total Score Grade-point average. end of first .513a .546a quarter in college Grade-point average, end of quarter .532a .6508 of initial withdrawal 8Significant at .01 level bSignificant at .05 level -35- Results of the correlation analyses support the following general conclusions: (1) Entrance test scores were of limited usefulness in pre- dicting success after readmission or reinstatement. Biserial cor— relation coefficients for male subjects ranged from .084 to .211 (median = .1795) and the coefficients for women ranged from .099 to .307 (median = .1915). Ten of the 16 coefficients were significantly. different from zero. Zero—order correlation coefficients between entrance examination scores and final grade—point average ranged from .139 to .290 (median = .223) for men and from .043 to .263 (median = .105) for women. Although nine of the 16 coefficients were statistically significant from zero, the practical significance for individual prediction purposes is negligible. The correlation between entrance test performance and grade— point average after readmission or reinstatement was also low, with co- efficients ranging from -.O46 to .249 (median = .043). (2) First—quarter grade-point index and grade—point average at the 9nd of the quarter of initial withdrawal were effective pre- dictors of success. The biserial correlation coefficients ranged from .513 to .650, and the range of Pearsonian r’s was .507 to .686 against the criterion, final grade—point average. Zero-order coefficients were smaller (range = .249 -.510) when grade-point average after read- mission or reinstatement was employed as the dependent variable. -59- Empirical probability Tables 4.10 and 4.11 were derived to show the relationships between first-quarter grade—point average and cumulative grade-point index at the end of the quarter of initial with— drawal, and post—readmission or reinstatement success. The relationships between the above grade—point measures and grade—point index earned after readmission or reinstatement are illustrated in tabular form in Tables 4.12, 4.13, 4.14, and Table 4.15. These data indicate possible critical "cut-off“ points that could prove useful in formulating readmissions policy and in making decisions or individual petitions for readmission. (1) Only 32 percent (66 of 208) of the academic withdrawals who had first—quarter grade-point averages below 2.00 were successful, and 41 percent (82 of 199) earned grade—point averages above 2.00 after readmission or reinstatement. The percentages dropped to 24 (31 of 127)and 34 (42 of 122), respectively, for students below 1.50 at the end of their first quarter in college. . (2) Only 21 percent (31 of 145) of the sample with cumulative grade-point averages below 1.50 at the end of the quarter of initial withdrawal were successful and 30 percent (41 of 138) compiled a grade—point index of at least 2.00 after readmission or reinstatement. There was a sharp drop in rate of success for students within the 1.00—1.49 grade—point interval compared to those within the 1.50-1.99 category. -90- TABLE 4. 10 NVMBI‘R AVID PFRCITNT 0F READFTI'I‘TLI) ()R REINSIATED MALY'S AND FEMALES AClllli‘x'INCl WITHIN VARIOUS ('IRADt—POINT TN’I‘IIIZVALS AFTER 1111; FIRST Ql,’AliIlZl-3 l.\' CULLIXLL WHO Width. Sl‘CClSSFl'I. AFTILR l-IliAl)‘-TISFTUI‘I UR REINSTATLL‘l.‘JT CHIAHE—PUIVT AVFRAGFZ, 1ND ()F F1831" Q.)l7.»\8'l‘til'3 [3i COLLTt/Ilf 0 00-0.49 0.50-0.99 1.00-1 49 1.50—1.99 2.00—2.49 2.50-2.99 3.00—3 49 MEN 11.11 17.86 31.03 43.86 61.70 84.62 (100.00) N = 213 N =1 N =5 N =18 N =25 N =29 N =11 N = l WOMEN (50.00) 15.38 23.53 41.67 68.18 (100.00) (100.00) 33:81 N=1 .\'=2 .\‘=4 \210 \‘=15 .\'=2 N=1 TOTAL 18.18 17.07 29.33 43.21 63.77 86.67 (100.00) SAMPLE N = 294 N =2 N =7 R :22 N =35 V =44 V =13 \ = 2 ’Percentages in parentheses are based on small N's (1 or 2). and may be quite unreliable. -91.. TABLE 4 . 11 NLNBLR ANT) PLRCLNT (1F R1.AI)\11TTT.1) ()R THINSTATLI) MALLS ANT) FTDIALLS ACTTTLVTNG WITHIN VARIOUS GRADE-POINT IXTTLTU'ALS AFTER THE QVATE'IT’TTZ (11.7 INITIAL \x’ITTHNIATa’ilL NIH-) WITH”. ST'CCLSSLT‘L AITTLR Ill-AINISSIUN (1R liTL'If-JS'I‘A‘ITIMTNT C'.‘.‘"IY_7LA'I‘1 \T? (HLAHL-PUINT AH’RACE. TINT) 017 QlfAi-‘(TTTR (113‘ 1‘11 TIAL W 1T1 {DRAMA L 0.00-0.49 0.50-0 99 1.00-1.49 1.50-1.99 2.00—2.49 2.50-2.99 MEN 11.11 20.00 24.64 56.57 100.00 N = 213 N = l N = 5 N = 17 N = 56 N = 11 WOMEN (50.00) 9.09 20.69 70.27 (0.00)” (100.00) N = 81 N = l N = l N = 6 N = 26 N = 0 N = 1 TOTAL 18.18 16.67 23.47 60.29 91.67 (100.00)‘ SAMPLE N = 294 N = 2 N = 6 \ = 23 N = 82 N = 11 N = 1 'Percentages in parentheses are based on very small N's (1 or 2). and hence may be quite unreliable. oHanHoecs mHHDU on zma tzm Hm ec Ho m.z HHmEm co wommo mom mHmmzHcoemm CH mmumocooewm . mmszp UonHoozm msH $>onm Lo Hm mowmeo>m HzHoaummew casemm 0:3 Hm>owsz some :H osmoomaum...FH.:LmHQ.~a mwsHm> onuHooam ago o50bm e0 Hm mommem>m ocHoosoUmom pacemo 0:3 Hm>oonH some :H omngsznm..;m...hm~u.flm HmywazH some :H ooLES: HmHOH u um Hmcou HocooHv H Hoo.ooHo H Hoo.ooHV H .Hoo.ooHs H H o:.m-oo.m oc.ooH mH co.ooH mH Hm.mo NH mm.oo o mm.mH N mH oo.m-om.m em.om m: mm.Hw H: em.mw am we.:: Hm RN.NH e om.: N we a:.m-oo.m _ em.co mm 33.:o Hm No.3m on ce.mm om Hw.sH m o“.m N am om.H-Cm.H 2 a no.3o Mm mo.Ho Hm oo.me m: em.mm Hm em.m m ON.H H aR.H H em o:.H-oo.H om.mo mm om.mm em om.om eH me.om x cm.m H em oa.o-om.o om.mm m cm mm m mm.mm m HH.HH H o o:.o-oo.c NeumNHe NememmHo ememmHe omsmmHo mammHa msmNHo smmHa :mNHo mmHammHo NHQ mHo He H.H so ...:./.w:,1_CU Om.o oo.H cm.H co.m em.m oo.m em.m 2H zeex< ._.z Tomnwzzzc his .m:,v< gifilufizxo mallow) :m:r.:0m_mm mH>CI< ".5 .~.< AmLWNSHO 2.,.._r_.<....amz LC mH/Hmzqoxm: .H.7m:§.~.<9mz7E z: ZHZWmCAQ<2~H MERLE m23z:HZH HZHlezz ZHZHHZ UZH>L~ZU< mmg<2 :;Hw Hchouocmew eczema 0:3 Hm>uoHcH some :H ownch um...Hu.:.mHm.Hu mosHm> UonHoyam mzH opoLm so Hm mcumoo>m HzHoaloCmow tozbwo 0:3 Hm>eonH some :H Hm>oonH zoom :H ooLEDC HchH om “wwou Hoc.ccHH m Hcc.coHV m .Hco.coHv m Hoo.omo H Hoo.omv H m oo.muom.m cc.ccH Hm 3m.mo cw Hm.mm mH No.0o 3H om.:H m om.: H Hm o:.muoo.m mu mo.mo mm co.cw om H©.mw oH mm.Kd HH md.om m mm oo.Huom.H _ mn.mo mH Om.mm 3H mm.cm m oc.om w oo.mm : Om.mH m 0H m3.Huco.H oc.mN m mm.rm m mm.mm : oc.mm m NH oo.ouom.o Hcc.ocHe m Hcc.ocHH m Hcc.Cmv H Hoo.omH H m o:.cuoo.o 3-32s $32: £325 9385 3mm: 38$ :25 3mm: 25 83 NE NE HpH HH Ho AH TH::._\.Hw Om.o cc.H Om.H oo.m em.m oo.m Om.m . Hmymu us :23 .mQ<~E>< HZ? E15223 .H.Z :./.H.H.H.<,rm_/HHH:_ z: ZHHHmmHHAH_<.H.HH._ x??? :3 D._.,_;< b2 HCnHIMEHEU WHHH<> ZQHLHUHQW m>31< z: H< ngmx;:U 2:%<;m;z L: z>HmHHngv %Z;7;E~223< m..._-H<7_.__r.H H_W.H.H.<.H.T.ZHH.E z: 2.....H..H.HZH._<...E .HC .hzflfivzma :7.< 3.27:2 mH 3: 2.37:. . mosHmy memHoogm yzH . . 1.. J . . u .. - ,.. :H . :HNH . oHCLm e0 Hm vowmo®>m HcHoaumwmom Uvzemm 2:3 Hm>eoHCH some :H HzocHo; um Q a Ha mmsHm; @oHHHosom mgr 932$ .5 to mmwmeo>m HHHHoanowmow HooHZmo 0:3 HWCmHHHH comm :H .3253: um .35... NHLHH Hm>oonH :omo :H uwpasc HmHQH n HM “mesa so.ooH HH oo.ooH HH oo.ooH HH Hm.oo oH HN.NN m oc.a H HH as.m-cc.m _ do.eo mm oo.:o mm mm.mh Rm sc.mm Hm mH.o. a No.H H mm oo.H-om.H m He.oo mm mo.Hm Rm mH.e© m: Hm.mm Hm RR.QH H N©.: m am.H H mo o:.H-oc.H em.mm Hm mm.mm em mm.nm :H “H.0m H NH.: H em as.o-om.o em.mm m em.mm m mm.mm m HH HH H o o:.o-oc.o HemsmmHa RomsmmHo omemmHs omqmmHo memNHa mammHo smmHa smmHo mmHQ mmHo NHQ NHe He He oe H<3< .H.z_.__H...mHH.H...H.mz2e HE 2,852.73: 39%. H_:< .H...H.:.H-EH a: .H.zH_E-W,_:
:mHnHHUHHmm .4382< :C .H.< Hmm_....z::U :.r._.H.<...._mm_z “HG .._.H>Hm_H‘HU.x.H_: ._.z.,.:sz:.<.H.mZHm; HE ZCHmmHZHEmT/H N._H.H..H:.H< mHHHU<~E>< HzTHmIEH/bfiv 2:752: H_::,HI»H ZH:.H.H3 CZHH/HHHZU/H. m...H-Hm HcHoalomem nosemm 0:3 HmkomHzH :omm :H Hcoooomlm H; NE a msHm> UoHWHounm 32H H .1 I). - . .. x . | I Q - Q s c n h o>onm so Hm moumeo>m HcHomnmthu become 0:3 Hm>bmosH zoom :H omoazzn m Hm mflw u Hm>ochH some :H bo253: HwHoH u um “meow Hcc.coH3 H H o:.m-co.m oc.ooH mm oo.ooH mm :H.mo am mm.:H om em.mm oH mm om.H-om.H om.mm om 3H.mm mm :H.em 0H 3H.mm o ew.eH m Hm.oH m mm o:.H-oo.H oo.cw H co.0m m co.om m oo.om N oH oo.c-om.o Hoo.coHc N Hoo.ooHc N foo.0m3 H Hoc.om3 H m m:.o-oo.o nomqmmHa HomsmmHo omemmHa omsmmHo mammHa msmmHo emmHa :mmHo mNHs mmHo NHQ mHo Ha HM Ho H<3< HzH:m-:e<:o ezHcm-e:3:< :3 :.< Ammammgu :...E.<...:mm_: “H3 ...:.:.r.f..:.HUx:V .HLZ...:/m..:.H.<.H.m./HHz: :3 ZH3HmmHZH_<:: :HHELE m:3< .H.ZH3HHI..._2<:C ::z:<: 3:3 4<3zzm UHHmzm cz< mu;zmex OH.H m mm.H H NH.N H mm.H a mo.H m mm.m m :oHHm>eom:ou 3H.m HH 0H.m HH mm.H a mm.m N mesHHsoHew< me.ov;H0mm- 3:.m oH mN.H H RH.N H no.m mH :o.m HH #3.: : :zHeHgonoe Mmmomnav 9 z w z N z e z a z e z . -1 goes :m.-o zH-o :- xmo,,. 3 somnz Hmuz osuz mmuz mHmuz mmHuz oouz mm z-H- H -H /< J- e-HHr onEmm HmHCP HmHOH Hsummouosmc: Hammmooosm Z 3 Z 33 212 Hooch Hammmmoozmcz Hammmooosm mug<2uh QZ< mmAch onHm cmeonz cmoopco 0:3 :oEmeom HHm Lew oem mHmwzHcoema :H MO HHQH 35H :4 mmmezootwm U:m mownesx, AHd.mH3Hoom3 HH.HH om oo.om HH HH.mH H Hm.mm oH Hm.oH NH oH.HH mH mH.H H eozeeeeezm c2 so.m o mm.H H om.m H mm.H m ::.H m HH.H H eooymm oH.H m oH.m m mm.o m 30.0 H Hm.o H HH.H H gmHHmcrsoe Hom.m3 HomHv :H.m HH 30.: : Hm.e m cm.m H om.m H mH.m a mm.m m mez< chH-utm Hee.mv -mmH- . ; . . - Od.m OH aim N merH N om.m m 00.: m mm.m m H._ZHCH_H_.,_>- >H,:<7H-._.._.H..._> :c.H o Hm.m e ::.H m mm.m m HmoHemz-otm oH.H m HH.H H HH.H H mm.H : me.H H HH.H N Hchm:-mHm mm.m H 30.: a oH.m : H:.H m Hm.o H HH.H N oocwHom HmHeom so.H e H:.H H mm.: H mm.H : mo.H H HH.H H mucmHum Hmonsem so.m e Ha.H e oH.m : HH.m H :oH-museo wsHmusz e z z e z e z e z z Hmmomnzo ammuz Hmuz esnz mmuz mHHuz mmHuz cone maesee z;H:quz:o 22< m-o-HHzo wHQEmm HmHCH HmHOH Hsmmmooosmc: Haummooozm HmHCF Hammmouosmc: Hammmooosm zszcz zoz ACOJCHHCOUV H.m :Hm .05). Readmission or Reinstatement Most of the sample were permitted to continue their college careers without interruption after academic failure and University College withdrawal action. Seventy-six of 81 women (94 percent) and 183 of 213 men (86 percent) were reinstated after the initial withdrawal. Furthermore, 84 percent (68 of 81) of the females and 73 percent (156 of 213) of the males were reinstated after subsequent withdrawal actions as well. Some educators are advocates of the theory that an enforced delay in readmission and absence from the campus are conducive to sub- sequent success, because the student has an opportunity to mature while engaging in worthwhile interim activities.2 For this sample, however, reinstated'students had a higher rate of success than readmitted stud— ents, but it should be noted that the latter group had’lower grade—point 1An investigation of the pattern of academic major changes was considered to be beyond the scope of the present research. 2See, for example, Howard L. Campbell and Walter Hahn, “Read- mission of Former Students after Absence from the Campus: Problems and Opportunities," College and Fniversity, 37: 126-134, Winter, 1962. -102- averages at the time of withdrawal and were poorer readmission risks. Thirty men and five women were readmitted after the initial withdrawal, and nine (26 percent) were successful. On the other hand, 116 of the 259 reinstated subjects (45 percent) were successful, a significant increase in rate of success (7(2 = 4.586, p (:05). The success rate of individuals always reinstated (108 of 224, or 48 percent) was significantly greater (:X2 = 12.491, p <:.01) than that of students readmitted at least once (17 of 70, or 24 percent). To ascertain whether the success rate differentials were a function of group differences in academic aptitude and/or college ach— ievement prior to withdrawal, the 70 subjects readmitted at least once and the 224 subjects always reinstated were compared on the basis of college entrance examination scores and two college grade— point measures. The results reported in Table 5.2 indicate that the two groups were quite similar in entrance test performance, but the reinstated students earned substattially higher grade-point averages. Again, there existed a relatively high correlation between college achievement before, and college achievement after, the initial re- admission or reinstatement compared to a rather tenuous relationship between test scores and college success. —103— TABLE 5.2 ACADEMIC APTTTTDE AND CDLLECE ACHIEVEMENT MEASERES FOR STLDENTS RE- ADMITTED AT LEAST ONCE AND FOR STUDENTS ALRAYS RFINSTATED Readmitted Students Reinstated Students CRITERION MEAST'RES 31:70 N=224 Michigan State University English 14.686 15.129 Test mean raw score Michigan State College Arithmetic 32.171 31.799 Proficiency Test mean raw score American Council on Education Psy- 98.543 96.746 chological Examination—Total mean raw score Michigan State University {eading 33.543 33.647 Test-Total mean raw score Mean grade—point average, end of 1.373 1.653 first quarter in college Mean cumulative grade-point average, 1.229 1.531 end of quarter of initial with- drawal —104— Repeating Courses The data contained in Table 5.3 indicates that many individuals in the sample attempted to recoup academic deficiencies by repeating courses in which a D— or an F— grade had been earned previously. The male subjects gained 1,521 credits (mean = 7.1) and 4,393 points (mean = 20.6) by this method,and the females gained 401 credits (mean = 5.0) and 1,080 points (mean = 13.3). i The role that the repeating of courses played in the success of readmitted or reinstated subjects may be illustrated by citing some spec- ific individual cases. One student, a successful male, raised his grade— pointaverage from 1.24 at the end of the quarter of initial withdrawal to a final average of 2.07 with a grade-point index after readmission or reinstatement, exclusive of repeated courses, of 1.81. He gained a total of 13 credits and 42 points by repeating courses. Another successful male raisrd his grade-point index from 1.44 to 2.01 with a grade—point average after readmission or reinstatement, exclusive of repeated courses, of 1.77. He gained a total of ten credits and 44 points by repeating courses. A third successful man raised his grade- point average from 1.69 to 2.00 with a grade index after readmission or reinstatement of 1.47. He gained 20 credits and 56 points. Several other similar examples could be presented, but it is sufficient to state that 20 men and three women achieved success with a grade—point average after readmission or reinstatement, exclusive of repeats, of less than 2.00 —103~ TABLE 5.3 MDAN NlNBER 017 F CREDITS ATEID D CREDITS RIIPI’ATFD, AND CRITDITS AND P(",)INTS GAINED BY REPEATING COTRSLS, FOR SUCCESSFLL AND UNSUCCESSFVL READMITTED OR REIXSTATHD MALES AND FEI'IALIIS MEN WOMFN Success-Unsuccess— Success- nsuccess- Total ful ful Total ful ful Total Sample 5:90 $2123 N=35 N=46 N=294 Credits of F 9.711 10.089 9.930 6.543 7.174 6.901 9.095 Repeated Credits of D 7.856 8.943 8.624 6.857 5.565 6.123 7.935 Repeated Credits gained by 8.011 6.504 7.141 6.057 3.717 4.951 6.537 repeating courses Points gained by 26.011 16.683 20.624 17.943 9.826 13.333 18.616 repeating courses -106— It may be noted in Table 5.3 that the credits of F repeated consistently exceeded the credits gained, which indicates that courses were failed more than once. The disparity was substantial for both unsuccessful men and unsuccessful women, and was considerably smaller for the successful sub-samples. Successful women, on the average, gained a significantly greater number of credits and points (t = 2.197, df = 79, p .<:OS and t = 3.070, df = 79, p (:111,115wmctively). than unsuccessful women. Successful men gained more credits (t 21.863, df = 211, p:> .05) than unsuccessful men and significantly more points (t = 4.704, df = 211, p <:.01). Transfer of Credits Leaving hichigan State University temporarily to earn credits at other institutions was not commonly done. Only 24 males (12 success— ful and 12 unsuccessful) and 16 females (11 successful and five un— successful) had credits from other colleges or universities listed on their official records at Michigan State University. The mean number of credits transferred was: successful men, 15.5, unsuccessful men, 21.7; successful women,13.3, unsuccessful women, 12.2. The mean differences in number of transfer credits between successful and unsuccessful males and females were not statistically significant (t = 1.001, df = 22 and t = .453, df = 14, respectively, p:> .05). -107— Enrollment and Achievement in Remedial Courses Michigan State University offered remedial instruction in reading, writing, speech,and arithmetic for students who were deficient in those skills. The remedial instruction constituted the Improvement Services. Data reported in Table 5.4 indicates that there was a total of 253 enrollments, including 61 repeats, in the Improvement Services: writing, 109; arithmetic, 57; reading, 56; speech, 31. Forty—nine per- cent (98 of 200) of the total enrollments for men were in the writing Improvement Service and 43 percent (23 of 53) of the enrollments for women were in the arithmetic Improvement Services. Thirty—nine of 90 successful men (43 percent), 64 of 123 unsuc— cessful men (52 percent), 14 of 35 successful women (40 percent) and 25 of 46 unsuccessful women, (54 percent) enrolled in one or more of the remedial courses. The number of successful and unsuccessful sub— jects who enrolled in Improvement Services did not differ significantly (7&2 = 3 032, p:> .05). A composite grade—point average earned by each of the sub-samples in the four Improvement Services was computed.3 The Improvement Ser- vices achievement index means between successful and unsuccessful men and women were not significantly different (t = .554, df = 144 amj t = .3us, df = 44, respectively, p:> .03). 3 , . . hhen an Improvement Service was repeated, the highest grade earned was used in the calculation. —108— TABLE 5 . 4 TOTAL .‘CI'FIRER ()1: lLNl-I(')LL:\IENIS [7‘3 REMEDIAL COT'RSES (.1I~'1Pl-i()\'E.\llL‘JI‘ SERVICES) FOR SUCCESSELL AND ETSECCESSFIL READMIITED OR REINSTATED MALES AND FEMALES IMPROYEFHINT S ERVICES ' Reading Writing Speech Arithmetic Total Successful Men tnsuccessful Men Successful Women Unsuccessful Women TU 1A1. SAMPLE 5 20 5 ll 51 (0) (9) (7) (1) '£112 15 29 l2 12 68 27 39 7 22 95 ’3) (30) (4) (0) (37) 30 69 11 22 13 6 1 4 (0) (1) (1) r0) (2) 6 7 2 4 19 5 3 4 17 29 (0) (l) (2) (2) (5) 5 4 6 19 34 53 68 17 54 192 (3) (41) (14) ( ) (61) 56 109 31 57 253 ”Xumbers in parenthesis indicate Improvement Services repeated ~109— The findings indicate that many academic withdrawals were aware of their limitations and attempted to improve themselves by enrolling in remedial courses, but there is no evidence that those efforts produced success after readmission or reinstatement. Summary The relationship of certain academic behavior patterns of the sample to post-readmission or reinstatement success was investigated. Successful students changed their academic major more frequently than the unsuccessful, the difference between male sub-samples being significant at the .01 level. The group of academic withdrawals never readmitted (reinstate— ments) was Significantly’ more successful than the group that was read- mitted one or more times (readmissions). The two groups were quite similar in their college entrance test performance, but the rein- statements earned substantially higher grade-point averages prior to withdrawal. It was concluded that the higher success rate of rein- . stated students was due primarily to their relatively superior pre- withdrawal achievement. The repeating of courses played a major role in success after readmission or reinstatement. The SUCCESSfUl student Of both sexes gained a significantly greater number of points by this method than did the unsuccessful. For female sub—samples, the difference in number of credits gained was also significant. -110- A substantial number of the sample, a non—significantly larger proportion of unsuccessful subjects than successful subjects, enrolled in one or more of the remedial courses, or Improvement. Ser— vices. The differences between successful and unsuccessful men and women in mean grade—point average earned in the Improvement Services was not significant. Only 40 subjects, 23 successful and 17 unsuccessful, transferred credits to Michigan State University from other institutions. This did not appear to be an important factor in the college success of re— admitted or reinstated students. Cl M P l'lLR \" I SUMMARY. CON USIONS, AND IMPLICATIONS FOR FUTURE RESEARCH Academic failure is a common phenomenon in American higher educa- tion; literally thousands of college and university students are dismissed each year. Some of these individuals terminate their college careers at that point, but many others seek readmission at the same or another institution. There is,therefore, a need for valid criteria which might be used by college and university officials to assess the merits of in- dividual applications for readmission. The present investigation was an attempt to identify such criteria. Summary The study sample consisted of 29h individuals who (1) entered Michigan State University as new freshmen in the fall of 1956, (2) were academically withdrawn by action of the University College bet— ween September, 1956 and June, 1958, and (3) who were readmitted or reinstated as of the fall quarter, 1959. The sample divided into four sub-samples: 90 men and 35 women who were successful after re— admission or reinstatement, and 123 men and 46 women who were unsuccess- ful after readmission or reinstatement. The academic progress of these students was studied until their graduation, termination, or through the spring quarter of 1960. —111— —112- Seven research hypotheses were postulated, based on the assump— tion that variables which have been empirically proved to be related to college achievement and persistence in general may also be related to achievement and persistence after readmission or reinstatement. One of the hypotheses was accepted, four were rejected, and the significance tests of two hypotheses yielded mixed results. (1) The proportion of female subjects who were successful after readmission or reinstatement (35 of 81, or 43 percent) was not signifi— cantly different from the proportion of male subjects who were success- ful (90 of 213, or 42 percent). (2) The proportion of veterans who were successful after read- mission or reinstatement (21 of 41, or 51 percent) was not significantly different from the proportion of non-veterans who were successful (69 of 172 or 40 percent). . (3) On the average, successful men and women earned higher grades in high school than unsuccessful men and women, but t—tests of the differences in quarter rank in high school class means were not significant. There was no significant difference in the numbers of successful and unsuccessful males and females who ranked in the upper half or lower half of their high school class. However, the rate of success of subjects who ranked tithe first quarter was significantly greater than the success rate of subjects who ranked in the bottom quarter. —ll3— (4) The mean age differences between successful and unsuccessful sub-samples were not statistically significant. (3) Successful subjects,on the average, earned higher scores than unsuccessful subjects on each of the eight college entrance exam- inations. For male sub—samples, the following mean score differences were significant: English, Arithmetic, ACE-L, Reading-V, and Reading-T. ACE-Q, ACE-T, and Reading-C mean score differences were not significant. For female sub-samples, only the Arithmetic score means were significantly different. (6) First-quarter grade—point average means were significantly higher for successful men and women than for unsuccessful men and women. Cumulative grade—point index means for successful sub-samples at the end of the quarter of initial withdrawal; end of the quarter of initial readmission or reinstatement; after readmission or rein- statement; and at graduation. termination, or the end of spring quarter, 1960 were also significantly higher than the corresponding mean grade—point averages for unsuccessful sub-samples. (7) There were no significant differences in socio—economic status between successful and unsuccessful subjects, using occupation, education of father, and education of mother as criterion measures. -114— Coefficients of biserial correlation and Pearson product-moment correlation coefficients were employed as indexes of the predictive power of ten academic variables—-eight college entrance examination scores, first-quarter college grade-point average, and cumulative grade- point average at the end of the quarter of initial withdrawal. The entrance test scores proved to be mediocre indicators of post-readmission or reinstatement success,yielding biserial correlation coefficients of .084 to .307 (median = .1885). Pearsonian r's ranged from .043 to .290 (median = .182) using final grade-point average as the criterion; and from -.046 to .249 (median = .043) when grade—point index after readmission or reinstatement was employed as the dependent variable. On the other hand, the grade—point measures were relatively potent predictors, yielding biserial correlation Coefficients of .513, .532, .546, and .650, and product—moment correlation coefficients of .507, .534, .564, and .686 against the criterion, final grade-point average. The magnitude of the zero—order r's decreased when grade in— dex after readmission or reinstatement was the dependent variable (range = .249 - .510, median = .334), As an exploratory extension of the study, the relationships bet- ween certain academic behavior patterns of the sample ani success after readmission or reinstatement were investigated. It was discovered that: (l) successful students changed their academic major significantly more times than unsuccessful students; (2) successful and unsuccessful sub-samples did not differ significantly either in enrollment in, or ~115- achievement in, remedial courses; (3) repeating courses was a critical factor in post—readmission or reinstatement success, the successful subjects gaining significantly more points by this method than the unsuccessful subjects; (4) reinstated students were significantly more successful than readmitted students, but the latter were considerably poorer risks due to their low pre-withdrawal cumulative grade—point average; and (5) only 40 subjects (23 successful and 17 unsuccessful) transferred credits to Michigan State University from other institutions, and the mean number of credits transferred did not differ significantly between successful and unsuccessful sub—samples. Conclusions The findings of the present study warrant the following con- clusions: (1) The prediction of college achievement and success is inprecise and uncertain at best, but it is especially difficult to predict the post-readmission or reinstatement success of academic withdrawals. (2) Standardized test performance effectively differentiates between withdrawals and non-withdrawals but does not discriminate well between academic withdrawals who are successful or unsuccessful, after readmission or reinstatement. (3) Academic withdrawals are quite homogenous in scholastic aptitude, but successful and unsuccessful withdrawals differ signifi- cantly in college achievement prior to withdrawal. —116- lmplications for Future Research The present study was more accurately described as an investi— gation of factors which may be related to the success after readmission or reinstatement of academic withdrawals. The investigation revealed certain areas which should be subjected to comprehensive study in the future. (1) The assumption that readmission is more conducive to ultimate college success than reinstatement needs more adequate empirical veri— fication. An experimental study of two groups of academic withdrawals similar in intellectual ability and in college achievement prior to withdrawal, one group readmitted after a year and the other group immediately reinstated, would permit a test of this prevailing assump- tion, (2) lhe present research included an investigation of the re- lationship of first curriculum choice and frequency of academic major change to post-readmission or reinstatement college success. A study of the pattern of curriculum changes is needed to determine why the academic withdrawals who change curriculum frequently are more successful. (3) An investigation of the relationships of non—academic behavior patterns of academic withdrawals, e g., extra—curricular activities, peer group relations and college residence experiences, to college success after readmission or reinstatement is needed. —ll7- (4) An intensive study of the non-cognitive characteristics, e.g. motivation, of deviant academic withdrawals, i.e., those who should have succeeded on the basis of academic aptitude and pre— withdrawal college achievement but did not and those who were poor risks but did succeed, could add substantially to the understanding of the dynamics of college persistence and achievement. (5) The repeating of courses was practiced extensively by the sample of academic withdrawals. What are the characterisitcs of in— dividuals who graduate from college due, in large part, to dogged per- sistence in the face of repeated failures? How successful are they after graduation from college? What affect does the extensive repeating of COUrses by marginal students have on the academic environment of a university community? Future researchers miglt well seek information pertinent to the above questions. (6) Socio—eConomic status as a factor in post—readmission or reinstatement success merits further study. BIBLItERAPlIY BIBLIOGRAPHY Aaronson. “.8. ”Factors Influencing Veterans to Leave School," Journal of Higher Education, 21: 96—97, February, 1950. “Lack of Money and the Veteran's Withdrawal from School," School and Society, 69: 28—31, January 8, 19u9, Bell, Laird. “Admit and Flunk,“ Atlantic Monthly, Vol. 210, No. 4: 54-56, October, 1962. Bragg, Emma W. “A Pilot Study of Mid-term Marks,“ Journal of Higher Education, 28:279-280, May, 1957. , "A Study of Student Withdrawal at 'W.U.',“ Journal of Educational Psychology, 47: 199—202, April, 1956. Brim, Orville 6., Jr. Sociology and the Field of Education. New York: Russell Sage Foundation, 1958. Brubacher, John S., and Willis Rudy. Higher qucation in Transition. New York: Harper and Brothers Publishers, 1958. Bryant, Girard T. “Why Our Students Leave School,” Junior College Journal, 212217—220, December, 1950. Campbell, L. Howard, and Walter Hahn. ”Readmission of Former Students after Absence from the Campus: Problems and Opportunities,“ College and Vniversity, 37: 126-134, Winter, 1962. DiVesta, Francis J., Asabel D. Woodruff, and John P. Hertel. "Motivation as a Predictor of College Success," Educational and Psychological Measurement, 9: 339-348, 1949. Dressel, Paul L, “Liberal-Arts Students Advised to Withdraw,” Journal of Higher Education, 14: 43-u5, January, 1943. Dwyer, P.S. “Some Suggestions Concerning the Relationship Existing between Size of High School Attended and Success in College,” Journal of Educational Research, 32: 271—281, 1938. Edwards, Allen L. Statistical Analysis for Students in Psychology and Education. New York: Rinehart and Company, lnc., 19h6. Eurich, Alvin C. "College Failures,“ School and Society, 37:692-696, May 27, 1933. -119— -120- ,"lmprovement in Scholarship During the Probationary Period,” School and Society, 35: 129-134, January 23, 1932. Fullmer, Daniel W. "Success and Perseverance of tniversity Students," Journal of Higher Education, 27: 445-447, November, 1956. Gerlach, Harry M. "The Attendance Record and Scholastic Aptitude Test Scores of Students Dropped for Poor Scholarship,” College and University, 24: 370-373, April, 1949. Giesecke, G. E., and John W. Hancock. “Rehabilitation of Academic Failures," College and Pniversity, 26: 72-78, October, 1950. Grace, Harry A. ”Personality Factors and College Attrition,“ Peabody Journal of Education, 35: 36-40, July, 1957, Griffiths, George R. ”The Relationship Between Scholastic Achievement and Personality Adjustment of Men College Students,” Journal of Applied Psychology, 29: 360-367, 1945. Hahn, Walter, and Boyd V. Sheets. "The Role of an Administrative Com— mittee in Assisting Failing College Students," College and Lniversity, 29: 378-385, April, 1954. Halladay, D. Whitney, and Dean C. Andrew. "Drop-outs from Arkansas Colleges," Personnel and Guidance Journal, 37: 212-213, November, 1958, Hardaway, Charles. ”Study of Official Withdrawals at I.S.T.C.," Teachers College Journal, 19: 72, December, 1947, Hofstadter, Richard, and C. DeWitt Hardy, The Development and Scope of Higher Educatior in the United States. New York: Columbia University Press, 1952. Hollinshead, Byron S. "ls European Education Better?” The qucational Record, 39:89-96, April, 1958. Hurd, A.W. "Why Graduate and Professional Students Fail in College Courses,” School Review, 57: 282—285, May, 1949. lffert, Robert E. “Retention and Withdrawal of College Students: F.S. Office of Education Bulletin 1958, No. 1 Washington, D.C.: U.S. Government Printing Office, 1957). Jenkins, Leo W. "High Schools Can Reduce Freshman Failures in College,“ Nation's Schools, 44: 50—51, December, 1949. Johnson, Granville B. , Jr. "A Proposed Technique for the Analysis of Drop-outs at a State3(3311ege,“ Journal of Educatirnual Research, 47: 381-387, January, 1954. -121- Johnson, Palmer 0. Statistical Methods in Research. New York: Prentice—Hall, lnc., 1949. Jones, Edward S. ”Why Students Fail in College,“ Association of American Colleges Bulletin, 39: 282-287, May, 1953. Kahl, Joseph A. The American Class Structure. New York: Rinehart and Company, Inc., 1957. Kirk, Barbara A. ”Test Versus Academic Performance in Malfunctioning Students," Journal of Consulting Psychology, 16: 213-216, June,l952. Koelsche, Charles L. “A Study of the Student Drop-out Problem at Indiana Fniversity, ” Journal of Educational Research, 49:357—364, January, 1956. Leffler, Emil. ”A Motivation Technique for Aiding Freshmen in Academic Difficulty," College and University, 26: 280-284, January, 1951. Lins, L. J., and Hy Pitt. ”The Staying Power and Rate of Progress of University of Wisconsin Freshmen," College and Eniversity, 29: 86-99, October, 1953, Lipsett, Laurence, and Leo F. Smith. ”Veterans Carry Through on Operation 'Education’,” Journal of Educational Research, 42:395-397, January, 1949. Long, Louis, and James D. Perry. ”Mortality Study of College Students” School and Society, 77: 103-105, February 14, 1953. Malleson, Nicholas. "Student Performance at Tniversity College,flondon, 1948—1951," Eniversities Quarterly, 12: 288-319, May, 1958. Marshall, M. V., and E. W. Simpson. "Vocational Choice and College Grades,” Journal of Educational Research, 37:303-305, December,1943. Matteson, Ross W. “Changes in Curriculum Preference Made by Basic College Students,” College and University, 26: 257-264, January,1951. McConnell, T. R. A General Pattern for American Public Higher Education. New York: McCraw-Hill Book Company, Inc., 1962. Mountford, James. "Success and Failure at the University,” Universities Quarterly, 11: 226-234, May, 1957. Mumma, Richard. "The College Record of Students Admitted on the Basis ' of G.E.D. Tests," College and University, 26: 79-86, October, 1950. Munger, Paul F. "Length of Residence and First Courses of Unpromising College Students,“ School and Society, 81: 120-122, April 16, 1955. , ”Student Persistence in College," Personnel and Guidance Journal, 35: 241-243, December, 1956. “A -l22— Potthoff, Fdward F. "Predicting the Ultimate Failure of College Students on the Basis of Their First Quarter's Records,“ School and Society, 33: 203—204,February 7, 1931. Quarles, Benjamin. ”Student Separation from College: An Overview," Association of American Colleges Pulletin, 35: 404-409, October,, 1949. Roth, Julius A. "A Study of Academic Success and Failure,” Educational Research Bulletin, 35: 176-182, October 10, 1956. Sanford, Nevitt (ed). The American College. New York: John Wiley and Sons, Inc., 1962. Schwebel, Milton. "Guidance for the Withdrawing College Student,“ Occupations, 25: 381—382, April, 1947. Sewell, William H., Archie O. Haller, and Murray A. Straus, ”Social Status and Educational and Occupational Aspiration,” American Socio- logical Review, 22: 67-73, February, 1957. Sheets, Boyd V., and Walter Hahn, ”Scholastic Probation as a Tool in Raising Academic Standards,” The Educational Record, 35: 150-154, April, 1954. Shuman, R. Baird. ”College Drop—outs: An Overview,” Journal of Educa- tional Sociology, 29: 347-350, April, 1956. Slocum, W. L. "Social Factors Involved in Academic Mortality,“ College and University, 32: 53-64, Fall, 1956, Smith, Leo F. “Student Survival in a Technical Institute,” School Review, 53: 294—299, May, 1945. Taylor, L. 0. “Why College Students Fail,” Journal of Higher Education, 19: 425-427, November, 1948. Thompson, Ronald B. Enrollment Projections for Higher Education, 1961— 78. Report of Special Committee of the American Assocation of Collegiate Registrars and Admissions Officers, September, 1961. Walker, E. T. “Student Housing and Vniversity Success,” School and Society, 42: 575-577, October 26, 1935. Warman, Roy E. "A Study of Applicants for Readmission to College,” Personnel and Guidance Journal, 34: 553-558, May, 1956. -123- Weintraub, Ruth C., and Ruth E. Salley. ”Graduation Prospects of an Entering Freshman,” Journal of qucational Research, 39:116— 126, October, 1945. Weitz, Henry, Mary Clarke, and Ora Jones. "The Relationship Between Choice of a Major Field of Study and Academic Preparation and Performance," Educational and Psychological Measurement, 15: 28-38, January, 1955. Whittemore, Irving C. "Does a Military Interruption Decrease the Chances of Obtaining a Degreeff School and Society, 78: 25-27, July 25,1953. Wooster, George F., and W. Wallace Stover. “Lost-Students,” Educational Research Bulletin, 37: 85-90, April 9, 1958. Wrightman, Lawrence 8., Jr. “The Effects of Anxiety, Achievement Motivation, and Task Importance Vpon Performance on an Intelligence Test,“ Journal of tducational Psychology, 53: 150-156, June, 1962. Yoshino, I. Roger, ”College Drop-outs at the End of the Freshman Year,“ Journal of Educational Sociology, 32: 42-48, September, 1958. .I“ - id 1 [J 5'." APPEZJDIX A -l25- afflum osa.ma www.mm new mmm.oaom mmm.wmo m omm.okmo maa.kmmoa myamats -ewesassco-omte wzoemtz 30:.mm mo:.mm Haw mem.~_aom mo:.mwc~ N oom.omam mom.mmmmm moon: zewmcm>oco oomom :mwflzoor macaw ox:.om Rmk.mm 3km mmw.ommoa owm.0moa m oem.oomm mmm.o:©~H madmaym semasamuc>lsmae weacmtz seq.mm mam.o: woo omfi.~maom mom.mmma m mmN.Noam :Ha.w:mmm meow: sawma¢>acs momam :mmae0az mesom Amos-Ichpmcfiem wok.om ame.~©: new Ham.qamoma eem.oamqom ome.om¢mm no.3mmzmamyamaus -x: amoamcaczusmm aeoom moa.om :Mm.mmo Hoe oo~.:mo:nm me:.moomam 3:0.3smmm QmH.o:m-: meow: loose so Haozsco smoosag< wsoom soasaa< D l - _ mom.om osc.mma saw www.mamam mea.aamm m mam.~mmaa o:.oo:mo mmamste -xmowmmemfiwmoMmmommwmm mom.c: Nmm.moH Haw JOH.NM~HNH moH.oo@m N owm.owmmH ood.N-mmH mmHmZ lost: :0 Hwozsco :mofloo§< woCom kogawo< o>womo«#:w:O.COHom:HEm :No.mfl moc.mH~ 3mm mm.m~mom QMm.00mH m mno.fimmm mmm.:omdm mmfimeou uxm Hmuflwoaozummm :Oflfim m-.o~ mmm.mHH ago 003.c0m0m 0mm.mfimm m 00m.:mdn ooo.:modm mofimz loop: :0 fiwuzzoo :moHLmE< mocum Home www.mmmH mm~.:mooH mvaEmm xozofionOCA aflomezofig< ooo.mozm msk.wqemm meow: massaco utmom cmmaeoaz :swm ©m3.mm mam.mm 3mm m©:.Ho~o omm.-o mmo.mm Hem.mm Hfio odo.©mmom mmm.HmmH m m wom.~m mmd.©m 3mm N3m.do©m 3©©.mmd N mmm.mmm dmm.omd© mmHmEoL . meoom wme mefiwcm mmm.©m HHH.mN flaw mmm.oNHdH 0mm.o:m m oom.om©H mom.Home mmfimz >pwmum>ficg momum :mmHSUHZ ma::Lo mascos b oozmflom> up Easy“: mosmwom> up :ym3owm mmomsvm mmtmsvm mmemacm mo mozzm L: mHm>q .m:z<;0m LC mZLW < xH-Xgmm< -l26— .~o>o~ Ho. om ozmowwflcwflm mos-m>|u HH< mmeHOU HJN.HOH oodm. 3mm wm0m.mm mmom.mm m mumm.0m mmmm.o©H moemeom :H omoomsc umowm no om:.omm flown. Haw Hono.mo- mmds.om m Ommm.mmH Hmmm.:mm mwfimz Use owmam>m ocfiomumvmou NHH.Nm www.moH 3mm m-n.mmwom cofi.mm3m N o~m.dooo mmo.0mm©m monEQL wecom HmDOHlammH wsflvmmm mmd.o: Hm:.oNH Haw Nmm.:momm moo.oamm m oma.owsoH mmm.dmmow mofimz xowmoo>fisg oomam :mwflzoflz mag/.0 mazoho u mozmflom> up :Hzowz wozmwcm> up cmozamm mwpmsvm mwomscm mmemscm mo mozzmfiz 225350 mo 53m mo 92m Esm Hmuoe Acozcflazoov < xHQ2mmm< ROOM uss Ottt‘r’ drawn-Inca" IL "lttlll‘llllllll‘lllf