University of Southern California Dissertations and Theses

The Allied Health Professions Admission Test : its role in selection for physical therapy programs

(USC Thesis Other)

The Allied Health Professions Admission Test : its role in selection for physical therapy programs

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE ALLIED HEALTH PROFESSIONS ADMISSION TEST ITS ROLE IN SELECTION FOR PHYSICAL THERAPY PROGRAMS by Deona M. Lilly-Masuda A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL University of Southern California In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Education) May 1984 UMI Number: DP24996 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dbssrtaîion WblisWng UMI DP24996 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106 - 1346 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA 90 0 0 7 This dissertation, written by Deona M. Lilly-Masuda under the direction of h.^.K,.. Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillm ent of requirements of the degree of D O C T O R O F P H I L O S O P H Y Dean t - I S L i h . tu DISSERTATION COMMITTEE ^ Chdtrman il DEDICATION This manuscript is dedicated to my parents, Richard and Medora Lilly, and to my husband, Eugene A. Masuda, for without their support, prayers, and confidence in me, this dissertation would not have been completed. Ill ACKNOWLEDGMENTS I wish to acknowledge the advice and consultation on the statistical portion of the dissertation by Roberta Madison, Dr. P.H., and Margaret Karagas, M.S. 1 would also like to acknowledge the assistance in typing various drafts of the dissertation and tables by Diane Kennedy and Cindy Chow. Lastly, 1 wish to acknowledge the support and prayers of many friends who have had confidence in me throughout the entire pursuit of this degree. IV TABLE OF CONTENTS Page DEDICATION.............................. ii ACKNOWLEDGMENTS ..... ill LIST OF TABLES........................... vi Chapter I. INTRODUCTION AND STATEMENT OF THE PROBLEM............. 1 Problem Statement Hypotheses Assumptions Significance of the Study Limitations Delimitations Definitions of Terms Organization of the Study II. REVIEW OF LITERATURE....................... 16 Introduction Allied Health Professions Admission Test Professional Education-Admissions/ Selections Education Summary of the Literature Reviewed III, METHODOLOGY,............ ................... 93 Background Population Statistical Methods IV, ANALYSIS OF THE DATA ...... ....... ....... 100 Results of Frequency Distributions Results of Simple Linear Regressions Additional Questions to be Answered Additional Data Analysis: Non- Hypothesized Findings V Chapter Page V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ................ 152 Summary and Conclusions Questions to be Answered Nonhypothesized Findings Recommendations BIBLIOGRAPHY . 165 APPENDIX 173 LIST OF TABLES vi Table P9KG 1. Raw Score Means and Standard Deviation for Applicants to Physical Therapy......... 24 2. AHPAT Annual Reports and Reports on Programs in Physical Therapy............... 25 3. Frequency Distributions, Total Sample...... 101 4. Distribution of Prerequisite Grades........ 103 5. Distribution of Grades in Physical Therapy Curriculum....................... .... 104 6. Distribution of Scores on Physical Therapy Licensure Examination (n=130) .................................... 106 7. Hypothesis One Correlations Between Preprofessional Criteria and AHPAT Scores in Overall Groups (n=194) ................. .............. 108 8. Correlations Between GPA in the Professional Program and AHPAT scores for Total Group (n=194)...................... 110 9- Hypothesis Two; Correlation Between Professional Program Courses and AHPAT Scores When Separated by Sex (Males=67, Females = 127).............................. .... 111 10. Hypothesis Two; Correlations Between Professional Pro gram Courses and AHPAT Scores When Separated by Ethnic Background (Caucasians=173, 0ther=21 112 11. Hypothesis Two; Correlation Between Professional Pro gram Courses and AHPAT Scores When Separated by Marital Status (Single=144, Marrieds38).................................... 114 Table vii Page 12. Hypothesis Two: Correlation Between Professional Pro gram Courses and AHPAT Scores When Separated by Major (Health Science=148, Non-Health=45 ).............................. 115 13. Hypothesis Two: Correlations Between Professional Pro gram Courses and AHPAT Scores When Separated by Number of English Courses (Minimum=66, More than Minimum:109)....... 116 14. Hypothesis Five: Stage 2— Combined Groups, Licensure Examination (n = 130)..................... 120 15. Hypothesis Six: Correlations Between Preprofessional Criteria and AHPAT Scores When Grouped by Sex (Male=69, Female=127)............. 122 16. Hypothesis Six: Differences Between Male and Female Performance........................ .......... 124 17. Hypothesis Seven: Correlations Between Categories of the Time of Application and AHPAT Scores When Grouped by Age......... ................ 126 18. Correlations Between Preprofessional Criteria and AHPAT Scores When Grouped by Ethnic Background...................... 127 19. Hypothesis Eight : Differences in Performance Between Caucasians and Others.................... 129 20. Correlations Between Preprofessional Courses and AHPAT Scores When Grouped by Marital Status...................... 131 21. Hypothesis Nine: Differences in Performance Between Singles and Marrieds........................ 133 Vlll Table Page 22. Hypothesis Ten: Correlations Between Preprofessional Course and AHPAT Scores When Combined by Major (n = 194)................... 134 23. Correlations Between Preprofessional Course and AHPAT Scores When Separated By Major (n = 148) ................ ......... 136 24. Stage 1 vs. Stage 2— Grouped by Sex Professional Curriculum...................... 139 25. Stage 1 vs. Stage 2— Grouped by Ethnic Background, Professional Curriculum .......................... 142 26. Stage 1 vs. Stage 2— Grouped by Major, Professional Curriculum.............. 145 27. Stage 1 vs. Stage 2— Grouped by Marital Status Professional Curriculum ........... 147 28. Preprofessional Courses vs. Professional Curriculum (Stage 1 vs. Stage 2) When Grouped by Number of English Courses Taken 149 29. Differences Between Students* Perfor mance with Low (<3.09) and High 03.09) Prerequisite GPAs............................ 151 CHAPTER I INTRODUCTION AND STATEMENT OF THE PROBLEM "For many are called, but few are chosen." (Matthew 22:14, RSV). There were 88 accredited physical therapy schools in the United States, including Puerto Rico, as of April 1983 (Educational Programs, 1983). Of the 88 programs, 82 offered bachelor*s degrees, six offered certificates in physical therapy, some programs offered both a certificate and/or a bachelor’s degree so a candidate might receive either or both. There were four programs that accepted entry-level master’s degrees that did not require an under graduate degree in physical therapy; 23 programs offered master’s degrees and seven offered doctoral degree pro grams. Most of the programs had more applicants than could be accommodated; these are referred to as impacted programs. Because the programs are impacted, there is justification for more concern about the criteria used in the selection process for admission into the programs. The numbers of applicants exceed the number of vacancies each year. However, it should be noted that many students submit applications to more than one program. With the 2 large number of applicants, great care must be taken in student selection for multiple reasons, many of which have to do with legal factors. More importantly, there is concern that applicants be chosen who will be best suited as contributing members to the profession of physical therapy. The attrition rate is very low in physical ther apy programs; candidates invest a great deal of time, effort, and money in preparation for application to one or more specific programs and, once selected to a program, are highly motivated to complete the curriculum require ments and the internship requirements in their chosen profession. Unlike some professions, such as medicine and dentistry (Aiken, 1971), there is no standardized test used consistently in the selection process by all schools of physical therapy. Whereas, medicine uses the Medical College Admissions Test (MCAT) and dentistry uses the Dental Admissions Test (DAT), physical therapy has no comparable examination to date. In September 1974, the Allied Health Professions Admission Test (AHPAT) was intro duced. It consists of five scores based on the following content areas: Verbal Ability, Quantitative Ability, Biology, Chemistry, and Reading Comprehension. Some schools are using the AHPAT as one of the admissions requirements. Since the AHPAT is a relatively new examin 3 ation, little is known about its significance in the selec tion process, and therefore there is a wide range of uses for the scores in comparing individuals in the applicant pool. The physical therapy curriculum must include, as required by state licensing law, specific content areas for the didactic education portion of the program. Some of the content areas of the curriculum are in the basic, behavioral, and physical therapy arts and sciences. More specifically, the basic sciences must include anatomy, physiology, and neurology; behavioral sciences must include psychology ; and physical therapy arts and sciences must include biomechanics. The physical therapy curric ulum is not limited to the above mentioned subjects, how ever, the sciences mentioned are common to accredited physical therapy programs and also contain elements of areas tested in the AHPAT. Additional selection criteria for students apply ing to physical therapy programs frequently include a minimum prerequisite and/or overall grade point average, biographical information, letters of reference, experience appropriate to health care professions, varied numbers of personal interviews, personality profiles, interest inven tories, or other standardized tests. 4 Selection of students into any professional pro gram has always been a difficult task for the members of the selection committee. With the large numbers of appli cants seeking education in professional programs in phys ical therapy and the large number of capable applicants, the task becomes even more difficult. In addition, the Washineton-Kiplinger Letter (August 13, 1982) indicated there will be a big demand ahead for physical therapists. The Chronicle of Higher Education (1981) indicated phys ical therapy is one of the ten most needed professions projected for the future in the United States. It becomes a matter of extreme importance to investigate various aspects of the admissions process to determine the best methods for selection that will result in selection of the most suitable candidates for the profession, and more specifically to determine if the AHPAT could serve as a viable instrument for making selections. Problem Statement Are there relationships among the Allied Health Professions Admission Test (AHPAT) and age, sex, ethnic background, prerequisite grades in specific subjects, prerequisite grade point average (GPA), success in spec ific didactic and clinically oriented subjects, and/or scores on the California State Board of Medical Examiners Physical Therapist Licensure Examination? 5 Hypotheses The following hypotheses were tested at the alpha level of significance of ,0005 to determine the relation ship between the AHPAT and identified aspects of the physical therapy preprofessional and professional educa tion program. A strict level of significance was applied to the data since multiple tests were performed on the same data, and the probability of finding statistical significance at levels above that is very high when run ning multiple tests (Dunn & Clark, 1974). 1 . There is no relationship between the separate scores and/or total scores on the AHPAT and the individ ual’s performance in certain required preprofessional courses as measured by grades: the separate AHPAT scores which are compared are verbal ability, quantitative abil ity, biology, chemistry, and reading comprehension with the grades received in first and second semester chem istry, algebra, trigonometry, and/or precalculus, biology, anatomy, physiology, first and second semester physics, biostatistics, number of courses in college English, and prerequisite GPA. 2. There is no relationship between the separate and/or total scores on the AHPAT and the individual’s success in the didactic education portion of the profes sional physical therapy curriculum as measured by grades: 6 the separate AHPAT scores which are compared are verbal ability, quantitative ability, biology, chemistry, and reading comprehension with the grades received in anatomy, physiology, neurology, psychological aspects of dis ability, and biomechanic. 3. There is no relationship between the separate and/or total scores on the AHPAT and the individual’s success in the clinical education areas of the physical therapy curriculum as measured by whether or not individ uals successfully completed each internship period in the internship experiences: the separate AHPAT scores which are compared are verbal ability, quantitative ability, biology, chemistry, and reading comprehension with success on internships. 4. There is no relationship between the separate and/or total scores on the AHPAT and the individual’s grade point average at the completion of the entire pro fessional program: the separate AHPAT scores which are compared are verbal ability, quantitative ability, biol ogy, chemistry, and reading comprehension with the grade point average at the completion of the professional pro gram. 5. There is no relationship between separate and/or total scores on the AHPAT and the individual’s scores on each section and total score on the California 7 State Board of Medical Examiner’s Physical Therapist Licen sure Examination: the separate AHPAT scores which are compared are verbal ability, quantitative ability, biol ogy, chemistry, and reading comprehension with the separ ate sections of the State Board Examination which are basic sciences, clinical sciences, and theory and proce dure • 6. There is no relationship between the separate and/or total scores on the AHPAT and the preprofessional criteria when grouped by sex; the separate AHPAT scores which are verbal ability, quantitative ability, biology, chemistry, and reading comprehension are compared to pre- professional courses when grouped by sex. 7. There is no relationship between the scores on the AHPAT and the age of the individual at the time the examination was taken; the separate AHPAT scores which are compared are verbal ability, quantitative ability, biology, chemistry, and reading comprehension with the age group the individual was in at the time of the examin ation . 8. There is no relationship between the separate and/or total scores on the AHPAT and preprofessional cri teria when grouped by ethnic background: the separate AHPAT scores which are compared are verbal ability, quan titative ability, biology, chemistry, and reading compre 8 hension with the separate ethnic groups of Caucasian and other, including Black, Mexican-American, and Oriental. 9. There is no relationship between the separate and/or total scores on the AHPAT and preprofessional cri teria when grouped by marital status : the separate AHPAT scores which were compared are verbal ability, quanti tative ability, biology, chemistry, and reading compre hension with the marital status of single or married. 10. There is no relationship between the separate and/or total scores on the AHPAT and preprofessional cri teria when grouped by preprofessional major of the individ ual: the separate AHPAT scores which were compared are verbal ability, quantitative ability, biology, chemistry, and reading comprehension with the preprofessional major of health science or non-health science. In addition the following questions were answered: 1. How are the scores of the examination utilized in the selection process at California State University Northridge? 2. What value is placed on the scores of the examination in the selection process at California State University Northridge? 3. Is there any difference in student performance between means for students in the following categories : 9 (a) GPAs < 3.09 vs. > 3.09, (b) female vs. male, (c) Caucasian vs. others, and (d) Caucasian vs. Oriental? Assumptions The following assumptions were made as applied to this study: 1. The present selection process at many physical therapy schools and other schools in the allied health professions is unsatisfactory to the individual schools. 2. An admission examination could be designed that would test individuals seeking entrance into any allied health professional curricular in an objective and reliable way. 3. The content areas chosen for comparison to the AHPAT scores were representative of common curriculum content throughout accredited programs in physical therapy and were the most closely related to scores on the AHPAT. 4. Preprofessional courses taken at other institu tions were equivalent to those at California State Univer sity Northridge. 5. The student records used from California State University were representative of physical therapy stu dents in the remainder of the accredited programs in the United States and Puerto Rico. 10 Significance of the Study Physical therapy educators and applicants have been concerned about the selection process for admission from the onset of the profession. The findings of this study could have important implications for the schools presently using the AHPAT or for those contemplating using it or other examinations in the selection process. Should the AHPAT be a reliable predictor of suc cess in the didactic and clinical program and the licen sing examination, the selection process could be more objective and equitable for the entire applicant popula tion. On the basis of predictability of the examination, the educational program could weigh the scores from the instrument more heavily, thus reducing the possibility of error that can occur by way of grade inflation tendency as manifested in the preprofessional grade point average. Therefore, the purposes of this study were three-fold: (1) to determine if the AHPAT can be used as a predictor of success in physical therapy programs, (2) to determine if the AHPAT can be used as a predictor of success on the physical therapy licensing examination, and (3) to deter mine if the AHPAT can be used as an objective instrument in the selection process of students for physical therapy professional programs. 11 Limitations The following are those limitations applied to this study; 1. The AHPAT has existed since 1974 and has been administered since 1975; therefore, only six years of scores were used for the purposes of this study. 2. California State Board of Medical Examiners Physical Therapist Licensure Examination scores were avail able for 130 out of 194 student records used in this study ; 64 scores on the examination were not available at the completion of this study. Delimitations The following are those delimitations applied to this study: 1. The student records of those accepted into the professional program, those successfully completing the program, and those taking the licensing examination were utilized as the only source of data for this study. 2. Only two schools out of the seven accredited schools of physical therapy in California used the AHPAT as part of their admissions process: Loma Linda Univer sity and California State University Northridge. Only one school is reported in detail regarding its specific use of the examination scores since only one set of student files was most accessible. 12 3. Demographic data, records of preprofessional grades and GPA, professional didactic grades and clinical success, scores from the AHPAT, overall professional pro gram GPA, and the scores on the licensing examination of physical therapy students at California State University Northridge were used as the basis for the statistical portion of this study. Definitions of Terms The following terms are defined for the conven ience of the reader in order to indicate their particular meanings used in this study. Accredited Physical Theraov School— Presently the accredited school is one which meets certain predetermined standards and criteria outlined by the American Physical Therapy Association, which is recognized as an accrediting agency by the U.S. Office of Education and the Council on Postsecondary Accreditation (American Physical Therapy Association, 1983). Applicant Pool— All applicants who meet the minimum acceptable requirements of the physical therapy program to which they submit application for admission. California State Board of Medical Examiners Phys ical Therapist Licensure Examination— An examination service provided by Professional Examination Services (PES) to the state licensing board for the purposes of 13 licensing physical therapists and providing a common ele ment in evaluation of candidates so that standards will be comparable from state to state. The examination consists of objective, multiple choice questions and includes three parts: Basic Sciences, Clinical Sciences, and Theory and Procedure. Certificate Granting Program— Any program in physical therapy that offers a certificate upon completion of the requirements of the program. Candidates receiving a certificate will have a bachelor's degree prior to the awarding of the certificate or will have met the require ments for a degree concurrently. Clinical Education Success— Success on internship experiences as measured by whether or not individuals successfully complete each internship period. Degree Granting Program— Any program in physical therapy that offers a bachelor of arts or bachelor of science degree upon completion of the requirements estab lished for the program. Didactic Education— The part of the educational process which occurs in the classroom and emphasizes skills and theoretical concepts to be put into practice in the clinical education phase of the educational process (American Physical Therapy Association, 1976). 14 Didactic Grades— The grades received by students in the professional program in their academic subjects. Health Care Facilities— Any group or organization whose major function is to prevent, evaluate, or treat physical or mental disabilities. Impacted Programs— Physical Therapy programs that have more applicants than can be selected for admission at any one candidate selection period. Internship— (Clinical education, clinical affil iations, clinical assignments, practicum, field experi ence, clinical experience). The portion of the student's professional education which involves practice and applica tion of classroom knowledge and skills to on-the-job responsibilities. This occurs at a variety of sites and includes experience in evaluation and patient care, admin istration, research, teaching, and supervision. It is participatory experience with limited time spent in obser vation (American Physical Therapy Association, 1976). Physical Therapy— A dynamic health care profes sion. Practitioners are skilled in planning, organizing, and directing programs for the care of patients of all ages who are disabled by illness or an accident or were born with a handicap. Physical therapists work in hospi tals, nursing homes, schools for handicapped children, private offices, rehabilitation centers, community health 15 centers, industry, and as athletic trainers and educators in colleges and universities offering programs in physical therapy (American Physical Therapy Association, 1983). Physical Therapy Professional Program— Any program in physical therapy, degree or certificate granting, which is approved by the American Physical Therapy Association. Prerequisite Grade Point Average— The numerical average for all courses required prior to admission to a specific physical therapy program. These courses vary with each program. Organization of the Study The study was arranged in the following manner: Chapter I includes the introduction to the topic, problem statement, hypotheses, assumptions, significance of the study, limitations, delimitations, definitions of terras and organization of the study. Chapter II consists of a review of the literature on the topic of admissions cri teria, student selection, and admissions examinations. Chapter III describes the methodology utilized for the purposes of the study. Chapter IV reviews the data col lected for the study. Chapter V summarizes the statis tical findings related to the student records, offers conclusions based on the findings and makes recommen dations for further study as a basis for more equitable student selection. 16 CHAPTER II REVIEW OF LITERATURE Introduction The review of literature is organized to provide first a chronological comprehensive examination of the Allied Health Professions Admission Test from its incep tion. As a basis of comparison of problems in selection, a review of professional education includes Physical Ther apy, Occupational Therapy, Allied Health/Medicine, Nursing and Law. A section on education is included to review admissions criteria and predictive measures for general admissions in higher and postsecondary education. The section includes a brief review of some cultural and social factors. Lastly, a condensed discussion on differ ences in achievement between males and females is in cluded . Allied Health Professions Admission Test History The history of the AHPAT, taken from the Report on Standardization and Preliminary Validation of the Allied Health Professions Admission Test (The Psychological 17 Corporation, December 1974), began in the fall of 1972 when the School of Allied Health at Ohio State University asked The Psychological Corporation for assistance in developing a standardized admissions test particularly for use in admission of upper-division students. There was a general agreement that items should include verbal, quanti tative, reading skills, biology, and chemistry. Questions were developed by psychometric personnel and consultants, including several content experts. Proposed questions were administered on a trial basis to upper-division allied health students at three different institutions: Johns Hopkins, Wayne State University, and Ohio State University. Items were dropped or retained based on sta tistical analysis of the experimental items (The Psycho logical Corporation, 1974). The result of the trial administration and statis tical analysis was to publish the test in five parts based on content areas: 1. Verbal Ability (75 questions), non-technical vocabulary. 2. Quantitative Ability (50 questions), basic arithmetic processes, 3. Biology (50 questions), major emphasis on human biology. 4. Chemistry (50 questions), elementary organic and inorganic chemistry. 5. Reading Comprehension (45 questions), three passages. (The Psychological Corporation, 1974, p. 2) 18 Test Administration There are four test dates for the examination yearly. Candidates must apply to take the examination approximately one month in advance of the test date and the fee is $25,00. There are 134 test centers in the continental United States, Alaska, Hawaii, Canada, and Puerto Rico. However, not all centers offer the examin ation on each of the four test dates. Students in southern California generally take their examination in Los Angeles. Standardization The standardization was based on 1,016 students tested in summer and fall 1974 from 17 campuses across the country. The age range was from less than 18 years old to over 26, three-fourths were women, and the educational backgrounds ranged from pre-college through college grad uates. The Kuder-Richardson Formula 20 which was used to describe the statistics is based on the mean, standard deviation, number of questions and number of examiners (The Psychological Corporation, 1974, p. 2). The relia bility coefficient is a measure of internal consistency of a test. The result was as follows: 1. Mean scores were significantly greater for males in biology and chemistry. 19 2. Mean scores were significantly greater for females in reading comprehension. 3. Observed differences in verbal ability and quanti tative ability were too small to be significant; however, males scores were slightly higher than females in both parts. 4. Seniors and college graduates consistently achieved higher scores in all areas; the pre college group had lowest means in three of five areas and college freshmen scored lowest in biology and chemistry. 5. The age group of 26 or older scored highest in verbal ability and lowest in quantitative ability. 6. The age group of 18 or less scored lowest in verbal ability, biology, chemistry, and highest in quantitative ability. 7. There was relatively little variability of mean scores by age on biology, chemistry, reading comprehension subtests. It was thought that the high scores in quanti tative ability of the pre-college group reflected prox imity of experience with the subject matter and that higher verbal ability scores in older students were con sistent with academic ability levels. (The Psychological Corporation, 1974, p. 3). 20 Caution was advised in drawing conclusions from the data about the mean scores of specific health pro fessions because of the limited and tentative nature of the figures. There were 10 Allied Health Occupational areas reported, 262 of which were tested for Physical Therapy. Physical Therapy students* mean scores ranked eighth in verbal ability, second in quantitative ability, seventh in biology, fourth in chemistry, and sixth in reading comprehension. The other occupational areas were dental hygiene, dietetics and nutrition, medical records, medical technology, nurse anesthetists, occupational ther apy, physicians* assistant, respiratory therapy, and speech pathology and audiology .(The Psychological Corpor ation, 1974, p. 9). Validation Validation was done in fall 1973 on follow-up data from the three institutions originally tested, Johns Hop kins, Wayne State University and Ohio State University, by presenting correlations between grade point average and AHPAT scores based on tryout forms of the examination. Correlations ranged from .24 (lowest, reading comprehen sion at one university) to .66 (highest, reading compre hension at another university) (The Psychological Corpor ation, 1974, p. 9). 21 In the publication, Professions Admission Test (The Psychological Corporation, October 1976), "relia bility refers to consistency of measurement, validity indicates how effectively a test measures whatever it is being used to measure" (p. 1). Reliability coefficients ranged from .83 to .91 on the five AHPAT scores showing the test measured what it was expected to measure consis tently (p. 1). The validity of the AHPAT in the October 1976 report was based on multiple correlation analysis. Cor relations between AHPAT scores and GPA in first year upper- division Allied Health Occupation majors ranged from .23 to .76. Twelve of the correlations were significant at the 1 percent level except one school was significant at the 5 percent level. To establish further validity, scores were compared of students who dropped out for aca demic reasons vs. nonacademic reasons vs. those who re mained the first year were compared. Physical Therapy was one of three largest specialties. Those who dropped out for academic reasons (eight students) scored lower in all five parts of the examination (especially biology) than those who dropped out for nonacademic reasons (11 stu dents) and retained students (284). Those who were re tained had lower verbal ability scores and higher biology scores than those who dropped for nonacademic reasons. 22 Interpretation of AHPAT Scores Scores are reported in percentiles which compare applicant scores with those of entering students admitted to upper-division allied health programs throughout the country. The percentage indicates the percent of upper- division students who were equalled or were exceeded by the applicants in each area (The Psychological Corpor ation, 1974-75a). "The purpose of the AHPAT is to help predict success in upper-division allied health programs" (PED- 1974-75a, p. 1). It was designed to be used as one tool in the complex task of student selections. The Psycho logical Corporation suggested some thoughts regarding discrepancies they had seen. For example, some questions could be posed when there are high scores on the AHPAT compared to low grades in the academic subjects related to the particular score: 1. Did the student come from a highly com petitive institution? 2. In high school, was the student in an honors program? 3. Were lower grades in non-technical or non-academic subjects? (PED-1974-75a, p. 2) The Psychological Corporation recommended that scores on the AHPAT be considered from the standpoint of the specialty the applicant seeks to enter and the 23 competency level required in each area tested for that particular specialty. It was suggested that the AHPAT only be used as a means of comparison, not directly for student selection. Raw score means and standard deviations were reported in "The First Year," (1974-1975b), published by The Psychological Corporation for all candidates of vari ous applicant groups and all candidates to various allied health programs. Raw score means and standard deviations were reported for applicants in 1974-1975 and 1975-1976 and are recorded in Table 1 for Physical Therapy only. Annual Reports from 1974-1975 to 1980-1981 and Reports on Programs in Physical Therapy are summarized in Table 2 with the exception of the 1979-1980 Annual Report and the 1978-1979 Report on Programs in Physical Therapy which are not ‘ available from The Psychological Corpora tion . Research on AHPAT Schimpfhauser & Broski (1976) found the pre professional GPA to be the best predictor of success in the first year of the professional program in physical therapy. In their study the cognitive factors used were five ACT subscores, preprofessional GPA, and five sub scores on the AHPAT. Academic success was defined as first year professional grades. In addition, in physical Table 1 Raw Score Means and Standard Deviation for Applicants to Physical Therapy 24 Test 1974-75 1975-76 Verbal Ability Mean 44.75 44.76 S.D. 11.72 11.85 Quantitative Mean 29.19 29.26 Ability S.D. 7.51 7.50 Biology Mean 30.26 30.43 S.D. 7.29 7.41 Chemistry Mean 27.16 27.13 S.D. 6.85 6.87 Reading Mean 26.58 26.48 Comprehension S.D. 6.97 6.87 Note From "Allied Health Professions Admissions Test, The First Year, 1974-1975" by The Psychological* Corporation, Professional Examinations Divisions, 1974-1975b, and "Allied Health Professions Admissions Test, Annual Report, 1975-1976." 25 oj <D rH g CO 0 LO VO • zf 1 cn m CM to ^ E - * VO 0 C ^ - S • b- <0 in to T — 0 >» CO • CM CL 1 E - I 0 C O cn C O C O C O C O C O • VO c . c ^ - to v- 0) H r4 cn C O c ^ - c a v VO VO # CO 0 1 CO m CM to < H CM •H CO >- S • CM C O >- VO to CM to C CO •H c ^ - CM LO cn • m 1 ro 1 — to ^ E - I VO 5 3 >- s j * Z • r- c ^ - VO to CM c . t a O 0 c ^ - to c ^ - 0 ST C T V • VO 1 ro ro T— to ^ E - h sr C VO CM to • V — 0 >- VO to CM C O 4 - 5 s - 0 VO CL C ^ - 0 ▼ — 0 • vO C D 1 CO ro CM to ^ E - 4 ro to LO 0 1 Z • b — C ^ - r j - C T v to T — •a r- c C O C O LO 4J >- ro ro 0 • C T » s - 1 cn ro CM to ^ M CM 0 S3 " VO 1 to • r- a >- CM c r > to • — C D to pH C O C O C D P C O *H C 4 - > 4 - 5 c C -H C O rH ^ 0 C O +5 • p 4 * r 4 rH « a j pH 0 r - ' C O to d C D < D -H 33 Q, CL 0 0 < « a ; 00 0 C D r-Nr-NS Û4 < * - i Cm C D C D ‘ -x 00 0 0 T3 0 0 0 QrQ 4J CL S - t . S 2: G C O 3 C D C D W 0 C D 0 0 0 -H bO G g g C D X bO G C J 7 3 3 bO C D C D C O 2 2 « a ; CO to ► - ] OvoSS C0CfMx4O C \ | GÜ 3C S to 00 OÜ 2 s u O OVOSZSOOOSO CM • ‘ to ac CM to OvOSto C O O S O CM S ac CM O VO S to CM S O LO S to CM 2: O VO to O cviga I g Q) X3 S m i C O coco S to ac T- s ^ V - 00 00 C J V = 0 “ C G 0 0 •H 0 •H CD X bO C O < D X bO bO CD CD q bO CD CD • a j ooto • a s«4 CO to s 26 3 oo O OOtn o CM ^ 2 u CO CM O O ^ to CO CM • » m o sj- vO o °? cr> h- J L j T- Uj o CO CM g 2 <53 CO O O cn to 2 o on to CO T— OO " » •‘CM • ‘ ,53 w o m m VO I CO o o L tn S 2 CO o in to o CO CM • ‘ • ‘ CM 2 C O ( 0 oo g c in 2: to • ‘CM 2 CO o • - 3 o i n s t o # » o*CM • ‘ a ,53 g in o = r VO I VO CO in 2 to CM 2 L CO CM O in to o CM CO in o = r VO in L CO * ~ 3 CO o # . oo S o c_r o in to CO CM • ‘ g in in = r VO g: C5 CO* o in to o CM 2 in in sr VO 1 I I 0) I C O C O C O 3 3 3 4^ 4 - > ( 0 ( 0 C O >» 4^ CL CO c CO C CO C 2 o O o < D TJ •H ■o •H ■o •H 2 c d< D X t u O C O < DX bO C O( U X bO E- o t a O < D0 ) o bO< D0 ) o bO d >< D < f l<s> C O CJ\ VO VO in o C O 0) h - CJ\ 2 V — 1 — 1 < D -p C O bO 3 4-5 o < D O G •H 1 —1 C O C J v O bO rH II h - CL O O to 1 0) rH o C O CO C O C O C O to g Il > , h - 2 G •* rH 1 — 1 o CO rH G C O >> C O O 3 G O •* G G to G 4H >» OO G O G Or h - «aj g •H < D C O 1 o o in 2 G O G h - VO VO in G 1 • 0 ) h - 4H "-3 2 C O 2 C Q Il 4-5 TJ E-l < D - 3 G C O B - O G rH oo • * 2 C O C O h - G 1 e u II o o h - O O G O II •H 1 i n in in •H OO O 2 O C O VO i n in in C O CT\ S ^ h - <0 o •» * » 2 •H 2 o Ç O to # A " r H G O G O 4-5 4-5 G « a J OO O C O G G O vO 1 •rH Il C O 0) C M C^ (0 C T » C O C O rH O 1 i n in in G l> - •H 4-5 1 00 LO in VO in O CJ\ > 2 0) h - •H •H G 1 -P G (0 Q C O TJ 3 O C O S •rH O O < D OO < 0 2 2 C O O O C m t— G C O II II m O 1 O < D ^ O G h - G •iH G 2 C O C O 1 in in o Ou 4-5 to 0) ad* in VO VO CT\ C O S h - 2 G < D 4H 4-5 4H •rH bO C O C O O 1 — 1 0) C O C D 4-5 C O t- rH qj 3 . +> < D l> - X rH 2 1 4-5 c 2 1 to O 4-5 G G o VO O G C O C D o "O l> - rH C D II O to rH TJ < D C O rH to 2 II C O < D •H G 2 II 3 > 3 > . rH O C O •‘ to to * > H C CL r4 •iH rH C D 2 G •H C O VO C O •rH bO •‘ C J * G t-. C O C O C D •‘ 4-5 C D C 0) 1 Q > > rH C D C O o -G • S in C m C O rH 4-5 (D C D o o O O C O 3 rH 1 ê G CJ\ G 4-5 O 3 1 -H 1 1 — 1 o to to O 1 TJ TJ 4-5 C M C O c _ > G C D C O vH G ü 2 « G G 2 C D < D •H O S O C O to O II ü r— 1 W M 2 s c u l 4^ Il II 3 l . X) > > 2 O to -p C O C O o 2 o C O 2 o 0 2 C L , H Ou 2 C O 2 ü 28 therapy, they found the ACT (American College Testing) composite and the AHPAT chemistry subscore fairly signif icant when the ACT subscores were available. The authors also found the preprofessiona GPA and AHPAT verbal sub score were significant when the ACT scores were not avail able. Schimpfhauser & Broski suggested that multiple factors be utilized for selection of students. They recommended the use of computer generated, institutionally derived indices with a weighted combina tion of variables for a more powerful and generally more efficient and practical method of student selection. Broski, Schimpfhauser, and Cook (1977) compared eleven different variables; five AHPAT scores, five ACT scores, preprofessional GPA with GPA in first year of the professional program. This study included a sample of different allied health professions (N=435). The data showed little correlation between AHPAT scores and pro fessional GPA, whereas preprofessional GPA and ACT math scores were found to be the best predictor followed by AHPAT verbal score. Based on the results of this study, Ohio State University discontinued the use of the AHPAT until the AHPAT could be used as a predictor of success with confidence. Also in fall 1977, Katzell reported on the back ground of the examination. The report included the same 29 information published by the Psychological Corporation plus some of the author's own comments and findings. The exam was devised because of a need for objective, accurate predictors of upper-division performance and because of increased numbers of qualified applicants and limited spaces available in professional programs. The author reported that approximately 12,000 students were tested between 1974-1977; norms were based on students accepted into professional programs. Results of findings indicated that preprofessional GPA correlated with GPA in Allied Health programs at .01 level in five of the six schools reporting pre-physical therapy GPA. AHPAT scores cor related at .01 level in all eight schools having physical therapy programs. The correlation between Allied Health GPA and a combination of AHPAT scores and pre-GPA were significant at .01 level for all six schools reporting pre-GPAs. Katzell concluded, based on results of the study, that "the AHPAT can provide comparable, relevant information for applicants from varied backgrounds to facilitate the process of identifying the most promising applicants to allied health programs" (p. 20). A study done at Northwestern University Medical School (Laurencelle, Kay, & Edelsberg, 1979) included an analysis of the AHPAT based on 312 applicants. Data were recorded on all applicants, enrolled: ranked alternates. 30 accepted/withdrew, incomplete/withdrew, and declined, "Because of extraordinary ranges, discriminating ability of AHPAT is questioned" (Laurencelle et al,, 1979). If GPA has good or better discriminatory power than AHPAT, the study questioned why AHPAT should continue to be used. "However, and to date, the grade point averages have proved good predictors of success in the demanding phys ical therapy program of studies" (p. 37). In an attempt to reconcile differences appearing in the literature on the AHPAT, Leiken and Cunningham (1980) formulated multiple regression models to analyze the AHPAT's predictability of success in allied health programs. The model used included entering grade point average (BGPA), final GPA in allied health programs (AGPA), AHPAT scores, both total score and individual scores, plus a variable which weighted the students pre professional institution attended (zero, if attended com munity college, one, if attended a four-year institution). The study included 152 graduates of allied health pro- I grams. The results revealed that BGPA significantly affected AGPA; however the explanatory power was low. When the preprofessional institution attended was added, the medical technology students from four-year institu tions did better, but the physicians* g&sistant students from community colleges did even better. When the 31 individual scores of the AHPAT were added, each section was significantly related to AGPA in physical therapy and physicians* assistants. When total AHPAT scores were used, there were substantial increases in the predictive ability for all programs except medical technology with only a marginal increase. The authors completed one final comparison with the cardiorespiratory science students. AHPAT scores of eight students who failed to complete the program were compared with those who completed the program and were found to have significantly lower AHPAT scores. Based on the results of the study, the authors concluded that the AHPAT was significantly correlated with GPA and improved predictions of GPA when used along with prior GPA and type of institution attended. A fairly recent (date unknown) and comprehensive paper written by Wiesseman of the School of Allied Health Professions (SAHP) at Loma Linda, California, asked the question: "Does the AHPAT add enough predictive ability to the college GPA to justify its use?" (Title page). Physical Therapy is one of fourteen departments in the School of Allied Health Professions. The applicants to the school come from many different colleges and univer sities, most are current students, some have not been attending college for varying periods of time. The valid ity of the application GPA, coming from many sources, as a 32 predictor of performance in the professional program is questionable (p. 3). Some may have been employed for several years and have reached quite different levels of maturity. It seemed logical to the author that the results of the same test given all applicants would be more comparable than GPAs coming from many different colleges. However, logic, even though well conceived, is not always in agreement with reality. Applicants are put under some emotional strain to take the test, it takes time (approximately 3-1/2 hours, Katzell, 1979) and vari ous amounts of money to travel to the testing centers, and there is a $25.00 test fee. Applicants must also plan well in advance of the test application deadline so the scores might be available in time for the professional program application deadline, Wiesseman used a sample of successful students in one of nine departments of the SAHP in 1976 for a more homogeneous group in scholastic aptitude, although because of the homogeneity, correlations of various independent variables were expected to be lower. Statistical analysis was done using the SPSS package; data were analyzed using stepwise multiple regression. The author used science, non-science, overall GPA, age, and the subscores for the AHPAT. Results for Physical Therapy were based on 109-111 cases (range due to missing data in some cases). Results 33 for Physical Therapy at Loma Linda are included with the results in Table 2. Correlations were ranked with overall acceptance GPA, (AGPA) being the highest (rank 1) and AHPAT reading comprehension ranking fourth. Multiple regression equations were done to determine the order in which the variables contributed to predicting the cumula tive GPA. In Physical Therapy acceptance overall GPA was the best predictor followed by AHPAT biology score and lastly, age. When professional board examinations were correlated with the independent variables and ranked by multiple regression analysis, acceptance overall GPA ranked first, AHPAT chemistry score ranked second, and AHPAT verbal score ranked third. The author concluded "that the AHPAT may give more help in predicting success in the board examination than it does in predicting the GPA" (p. 9). As a result of the study, although the AHPAT did not give as much ability to predict cumulative pro fessional GPA, Loma Linda has continued to use the AHPAT as one method of evaluating success of prospective appli cants. Results of a questionnaire on the feasibility of developing and administering a Physical Therapy Admission Test included some comments on the AHPAT (Soderburg, 1983). Ninety Physical Therapy educational programs responded to the questionnaire and one of the common 34 responses was that the AHPAT had little apparent value. Most indicated an interest in development of an exam spec ific to Physical Therapy and expressed an interest in its ability to "predict" success. Based on the results of the report, the author stated that the AHPAT did not have strong applicability to physical therapy and that few schools required the test for selection of students. Professional Education-Admissions/Selections A chapter in McGlothlin*s book, Patterns of Pro fessional Education (1960), was devoted to recruitment and selection of students (p. 137). The author reports the conclusion reached by Ginzberg and associates that choos ing an occupation is a lengthy process and is largely irreversible (other authors take exception to this). Tentative choices are made beginning at age 10 to 12, realistic choices are made beginning at ages 16 to 18. Professional schools must consider both student demand and societal needs but, at the same time, must be sensitive to the needs of the profession (p. 138). Population growth, rise of technical knowledge, and growth of social and economic institutions are all factors which expand needs for professional people. Ideally, needs can be met by selective recruitment of students of high quality who normally would not reach professional schools. The purposes of professional 35 schools are for education of the students and to prepare students to enter the profession equipped with the compe tencies, the motivation, and the ethical understanding needed for successful practice (McGlothlin, 1960, p. 140). Students must be capable of effective interaction and able to grow in directions desired by the profession and the school or the efforts of the school will fail. The dream of faculty is to admit students of high intellectual ability, "deeply interested in the subject matter and practice of the profession, well-rounded in personality, and burning with curiosity to know and with desire to do" (p. 140). To accomplish this, standards of selection are established that would distinguish the poten tially successful students, "An applicant who surpasses the minimum standards must compete with all the other applicants who have surpassed the minimum standards" (p. 141). Professional schools have two major compulsions for recruitment— one, to fill available spaces, two, to raise standards of the profession by raising levels of admission— but also to recruit those highly qualified students who would possibly never get to a professional school. The purpose of recruitment is not only to get enough but to get the best (p. 142). 36 The admissions committee is crucial to the life of the professional school. The characteristics looked for in the applicant are personality appropriate to the pro fession and depth in preprofessional study. Methods of selection used at the time of the writing of McGlothlin's book were use of tests, e.g.. Law School Admissions Test (LSAT), Medical College Admissions Test (MCAT), student scholastic record (most common), interviews, autobiog raphies (used by some), and letters of recommendation. In a letter to McGlothlin from Dr. Katherine A. Kendall (dated October 10, 1958), Kendall stated "intellectual attainments may not always be of primary significance in certain professional fields" (p. 162), Physical Therapy In 1975, Bradshaw conducted a survey which in cluded 12 Canadian schools and 65 U.S. Schools of Physical Therapy to determine methods of selection used in the schools. Of the 77 schools surveyed, 67 replied (87.01%). The results of the survey showed an increase in the number of spaces available between 1974-75, increase in numbers of applicants and accepted applicants, and a lower ratio of applicants to places offered. Of the 67 schools reply ing, 55 used interviews ; 76,35 percent used only one inter view of which the greatest percent (30,90%) used one inter view and one interviewer. Fifty percent used only aca- 37 demie faculty as interviewers. Sixty percent used inter views to measure attitudes and interactions and others (ability to communicate, poise under stress, experience and insight into the profession). Semi-structured inter view format was used by 70,91 percent; 63.64 percent scored overall impression only versus individual points. All applicants were not interviewed in 63.64 percent of the schools. Bradshaw reported other critieria used: GPA, completion of certain prerequisites, satisfactory refer ences, high scores on certain specific psychological tests (19 tests were tested, one of which was the AHPAT). Entrance GPA was used by all 55 schools using the inter view also. Biographical letters and letters of reference were considered to be of value by a majority of the schools. Most schools weighted the interview less than 50 percent and of these, the largest were in the semi structured interview category. Of the twelve schools not using interviews, half used GPA and test scores, extracurricular activities, knowledge of the profession, and references as criteria. Undergraduate overall GPA and grades in prereq uisite science courses were found to be the most signif icant factors in student selection by Landen (1977). Undergraduate overall GPA was the most significant of the 38 two. As reported by Landen, a study by Dworkin (1970) found that overall the DAT (Dental Admissions Test) was not a good predictor of dental school performance but the DAT-chem subscore and DAT space relations subscores were among the best predictors. In two other studies reported by Landen (Hill, 1959, and Stefanau & Framen, 1971), Hill found high MCAT (Medical College Admissions Test) scores were associated with high grades, and high grades with low MCAT science subscore and quantitative subscore indicated a questionable applicant. Stefanau and Framen found a high correlation between MCAT science scores and high pre-med grades. Landen reported a study by Tidd and Connie (1974) which found significant correlation between academic suc cess and clinical success. Another study reported by Landen, Anderson and Jutzen (1965), found that grades and aptitude test scores were good predictors of academic success. Most authors suggested the interview was a use ful tool for academic screening; however, it was pointed out that the interviewer must be well prepared and use appropriate techniques and procedures (Landen, 1977). In 1978, at the National Conference of the Amer ican Physical Therapy Association (APTA) in Las Vegas, Nevada, Green and Cox presented a paper based on research begun from 1973-74 school year. The purpose of the 39 research was to develop an instrument to be used for admis sion procedures at the University of Pittsburgh, School of Allied Health Professions. The research indicated that the typical variables used for admission criteria were test scores, previous academic achievement, letters of recommendation, and interviews or biographical data. A form was developed at the University to stan dardize and simplify application review which was divided into three sections. Section I, Academic Record, in cluded % (1) GPA overall and prerequisite; (2) level of preparation, including number of credits earned (60 credit minimum) and number of prerequisites completed ; (3) con sistency of performance, same level throughout or rising and falling and ; (4) academic area of emphasis. Number one through three were scaled in this instrument. Section II, Letters of Recommendation, was judged by three eval uators on the attributes of the applicants. The letters were judged on the following; reflecting trust, respon sibility, creativity, cooperativeness, and goal knowledge of the applicant. The last section of the instrument. Section III, Biographical Information, was an essay written by the applicant for which a guide for content was provided. There were four areas judged; (2) self-evaluation essay, regarding quality and organization of the essay, content. 40 insight, accomplishments, aspirations in and out of health care ; (2) account of health care experience, P.T, and other, duration, quality, involvement, motivation ; (3) current enrollment and residential status ; and (4) special considerations, e.g., prior formal supportive study experi ences and/or accomplishments, prior educational depra vation . Each section was weighted by numerical values attached to each criterion based on a long-range evalu ation of the instrument (Green & Cox). Reliability was obtained by consistent use from reviewer to reviewer of the form. How well did the items on the application (independent variables) predict certain performance factor (dependent variables)? Dependent variables included GPA in curriculum, clinical performance ratings, and ratings by first post-graduate employers. The results indicated that none of the independent variables were significant predictors of clinical ratings or employer ratings and correlations involving clinical and employer ratings were negative. However, it was noted that the study included only selected students since it was not possible to collect all of the dependent variables on students not selected. Further, the results indicated that for the 1975 class, prerequisite GPA was a significant predictor of GPA in the curriculum, and for the class of 1976, 41 overall GPA was the best predictor of curriculum GPA, Total points on Section I were also a significant pre dictor of GPA. "Validity was studied by determining if the variables measured could identify those students who were successful, as defined by several criteria, and those who were not" (Green & Cox. p. 3). At least two faculty members rated each applica tion. There was 97 percent agreement between the two raters on classification of applicants' overall GPA. There was no variability on "health status," high reli ability on academic data, and high agreement on personal evaluations with greater consistency of agreement on the ultimately accepted students vs. the applicant pool. Occupational Theraov Lucci (1974) suggested that an approach to selec tion of occupational therapy students include a ranking of applications and interviews. The application required an indication of how specific requirements would be met, work experience, educational financing, transcripts, three letters of reference, and a minimum GPA of 2.5 for pre requisites and 2.5 overall GPA. The application included a summary sheet to be filled out by the applicant. Inter views were conducted by two staff members and included such items as: appearance, speech, general impression, most and least impressive characteristics of the inter 42 viewee, areas of greatest strengths and supplemental observable or verbal behavior based on Smeltzer's Inter view Chart (Lucci, p. 92). There were four scores (highly recommend, recommend, hesitate to recommend, do not recom mend) with comments on each item. Applicants were ranked based on the following factors: scholastic achievement, communication skills, knowledge of field, related work experience, interaction during interview process, independence and assertiveness displayed, variety of experience with people and different lifestyles, sincere desire to work with others, experience with persons with emotional and physical problems or the disabled, initiative to extend beyond minimal require ments, self-awareness, character references, voluntary service rendered, good health and personality traits observable. Lucci reported a high correlation between faculty members on ranking. Later in 1974, a study was reported by Johnson, Arbes, and Thompson on selection of Occupational Therapy students by reviewing several previous studies on the topic. The American Occupational Therapy Association (AOTA) reported a high correlation between the score on the Occupational Therapy (O.T.) Career Inventory and clinical performance. Sens, at Texas Women's University (cited by Johnson et al.) found the ACT (American College 43 Testing Examination) scores correlated significantly with academic course performance and scores on the AOTA Regis tration Exam, High correlations were also found between required freshman, sophomore courses and junior-senior courses, the core curriculum and scores on the registra tion exam, e.g., Anat-Physiology, Kinesiology- Neurophysiology. Both sophomore courses correlated highly with junior-senior level courses and registration exam scores. Johnson et al. (1974) reported a study by Booth done at San Jose State, indicating personality and interest profiles were inefficient predictors of registra tion exam scores and clinical ratings. Another study reported (Maynard, Bilkey, & Hyre, 1972) indicated that previous experience was moderately associated with the performance of students enrolled in a training program for O.T. assistants. The methods used in selection by the various schools were: interviews, college grades (generally cumu lative GPA for freshman-sophomore years), aptitude and achievement tests (SAT, ACT, Otis Test of Mental Ability, AHPAT, etc.), letters of recommendation (less than half the schools), application letter or essay (10 out of 40 schools required), personality or interest inventories (seven schools used), high school grades (seven schools 44 used), biographical questionnaire (seldom used), and one school used a lottery system. The personality characteristics that were evalu ated were motivation, personality, academic ability experience in health-related activity, knowledge of O.T., health, social achievements, state residency, minority student, and one school used first B.A. degree. As a result of the study, Johnson et al. recom mended a systematic development of selection criteria which would include cognitive variable (e.g., GPA, test scores), affective variable (personality and interest measures) and past experience, i.e., "those variables which tap those qualities that most closely resemble pre ferred outcomes, e.g., successful school and job perfor mance" (p. 601). Allied Health/Medicine A number of studies both published and unpub lished, have shown grades to be the best single predictor of success in medical school according to Stevens (1973). The author cautioned the reader, however, that courses attempted but not passed may not have been recorded depend ing on the policy of each particular institution. To assist schools of health professions in the dilemma of the "admissions squeeze," Chaisson (1976) 45 studied the long-term experience of medical school student selection. Four issues were addressed: 1. Personality issue— In an article by Korman, achievement tests are effective in identifying the "smart, achievement-oriented, rather aloof individuals who know how to get good grades" (p. 8), The question arises whether selection keeps out "the most humane and dedicated individuals while admitting the aggressive, science- oriented scholar" (p. 8). The issue of low utilization of medical care facilities is mentioned as it is perhaps related to physicians with poor interpersonal skills which may be a function of the selection process. 2, Numbers issue— The deluge of applicants makes the selection issue even more important. 3. Legal issue— Increased litigations, and affir mative action CDeFunis vs. Odegaard, a non-minority vs. minority) are a growing problem. 4, Lottery issue--This method was thought to be no more acceptable than any other method; however, it was proposed that this method might be used on a trial basis for a segment of the selection and then compare the results with traditional selection procedures. Chaisson (1976) listed criteria measures which included cognitive domain, GPA, MOAT scores, and non- cognitive measures which included personal interview, 46 letters of recommendation, personality testing, and bio graphical data. The author reported a study done by Ham burg in 1971 where criteria were ranked: first, overall GPA, second, science GPA, third, MCAT science scores and fourth, committee letters. A later study, in 1973 by Wingard and Williamson, reported by Chaisson, indicated that grades, MCAT and Board scores were not good predic tors of physician performance. Grade inflation decreases the ability to determine differences in students based on GPA according to Chaisson ( 1976, p. 9). He reported a 1974 study done by Cervo Juola showed grade inflation from 1960-1968 (post Sput nik). Many schools have converted to a minimal GPA level, and then judge applicants based on other criteria. Chaisson suggested that the MCAT be used as a good predictor of academic success in the first two years of medical school ; that it not be used as a sole predictor ; that it can be used as a predictor of the specialty the graduate is likely to enter; and that future editions of the test assess clinical performance, e.g., problem solving skills. Personal interviews must be structured with predetermined questions and criteria with a well- defined point system. "Interviews are expensive and time consuming" (p. 11). There is increasing difficulty in defending the interview format and decision criteria. 47 A 1972 article by Wiseman, reported by Chaisson, suggested a computer-based selection technique. However, there was no evidence that selection by this method cor related with physician success. A study done in 1977 by Williams et al also on a computer-assisted admission pro cess, reported no conclusive evidence of predictability. Additional problems with interviews mentioned were biases of interviewers and either interviewees who were able to portray only their best or those who lacked the social graces who might be disadvantaged in the interview. The interview can measure non-cognitive characteristics ; however, this has not been supported in the literature. The Medical College Admissions Assessment Program (MCAAP) recommends a training program for interviewers (p. 12). Letters of recommendation are receiving less empha sis due to bias and the implications of the Family Educa tional Rights and Privacy Act of 1974 which allows stu dents and parents to read the student files. Letters can no longer be guaranteed to be confidential unless the applicant waives the right to read the letter. Personality testing was not used by many schools for admissions purposes at the writing of Chaisson's article. Some interpretations of this fact were that the tests are threatening to some applicants, others seek tutoring in "response techniques for mastery over the 48 instrument" (p. 14). New ways of gaining objective infor mation or personality characteristics are being studied by a committee of the Association of American Medical Col leges (AAMC), It was found that biographical data were valuable mainly from the standpoint of predicting the future area of specialty and/or practice of an individual based on background. In 1974 a report was approved by the AAMC recom mending revision of the MCAT to include expansion of the test into non-cognitive areas. Other suggestions for admissions were to define good versus bad physician char acteristics and exclude applicants based on bad charac teristics or to develop specific measures to select for achievement. Is the student reasonably capable of professional studies?, screen for psychopathology, select for char acteristics which should include practice-choice and location of same (practice defined to follow the philosophy of the particular medical school, and to select for professional dedication. Society needs life-long learners. (Chaisson, 1976, p. 15) Medical care, or health care, is no longer the responsibility of the physician alone— if it ever was. The last two decades have seen a phenomenal increase in numbers and variety of health professions and health occupations which, with the established health professions, are ideally envisioned as a team." (Miller, 1976, p. 8) 49 Miller discussed the problems of selection ; e.g. , the issues of females in great numbers seeking careers, expecting to be judged on the same selection criteria, along with the issue of increasing numbers of Black Amer icans, American Indians, and Mexican Americans applying, but lower percentages being accepted (1970-73). The GPA and MCAT's verbal and science score increased between the years 1953 to 1974, but "the pre dictive validity for such a selection method is high when the criterion is academic performance in the first year" (p. 4). Only half of the students excelling in basic sciences also excel in clinical performance (p. 4). How ever, 70 percent who excel in clinical performance have not excelled in basic science. The differences in minor ity students from disadvantaged backgrounds gradually decreased but never disappeared. Miller mentioned a few studies that failed to demonstrate a significant correla tion between academic achievement and later performance. Historically, the Flexner Report written in 1910 had a major impact on medical education by establishing the importance of scientific medicine through the addition of basic sciences as fundamental to clinical education according to Gellhorn (Miller, 1976). Previous appro priate higher education was also stressed. Then World War II had a major impact on medicine and medical education. 50 GPAs and MCATs became the two most important factors in selection (66% of the decision, Oetgon, in Miller, 1976). Since then, GPA scores have risen rapidly until in 1975 more than one-thiKd of the applicants had straight A aver- \ ages. The majority of pre-med students take numerous science courses and the concern for high grades in each one is such an obsession that cheating, sabotage of class mates* experiments, and other dirty tricks are common knowledge according to Alton Blakeslee (Gellhorn in Miller, 1976, p. 9). Considerable literature agrees that GPA and MCAT scores accurately predict grades in the first year of medical school. Thereafter, it drops off until by the fourth year the correlation is very small, and none accu rately predict achievement in the clinical years (Gellhorn in Miller, 1976, p. 9). Performance on Part I of the National Boards cor related well with the MCAT which, according to Gellhorn, was less than flattering to the Boards. Gellhorn cited four studies which demonstrated little correlation between the personal interview and medical school performance, general competence at gradu ation, or clinical effectiveness as measured by intern ships. "Perhaps it is not too surprising that the inter 51 view predicts little more than the personal preference of the interviewer" (cited in Miller, 1976, p. 11). The author * s suggestions for change in the selection included : 1. Define the objectives of medical education. 2. If that is accomplished, it would indicate the premedical requisites in a more rational fashion (e.g., decrease chemistry and math which are not contributing to vital bodies of knowledge, increase social sciences and emphasize humanities) 3. Select basis of cognitive and non-cognitive criteria with more testing in the area of problem solving 4. Nonacademic criteria should include personality, attitudinal, self-analysis, and other forms of examination to acquire conclusive information. "The best predictor of future activity is past performance" (cited in Miller, 1976, p. 13). Gellhorn made a final statement regarding the need for reform in the selection process to respond to the needs of the present; There are many signs that society is finding the medical profession too insensitive to health care. The emphasis on education for allied health professions and the resurgence of faith healers are clear indications that traditional doctors are not meeting societies needs. (p. 15) 52 Marston (Miller, 1976) listed three major problems in student selection for medical professions : 1. Public frustration, with a profession restricted to a fraction of applicants who are fully quali fied 2* Student confusion regarding job market, selec tivity, and admission of students experiencing unreasonable blocks to goals pursued 3. Social purpose versus individual rights. According to Margaret Mead (Miller, p. 31), stu dents should consider the possibility of three or more careers in a lifetime and should develop skills and atti tudes that would transfer across several fields of employ ment • The trend is toward purchasing a product with tax dollars that will match those products to the job oppor tunities actually available (Miller, p. 31). Marston summarizes by stating that we must do a better job in terms of defining reasonable criteria for selection, more effective communication of the criteria, and adherence to the criteria once it is established, "One problem about which we must do a better job is being sure that alter native pathways are not eliminated for those who are unsuccessful" (p. 33). 53 Physicians are becoming more oriented toward high ly specialized tertiary diagnostic and therapeutic tech niques, and other health care professionals are assuming the role of interpreting for patients their disease(s) and the meaning or purpose of diagnostic and therapeutic pro cedures being applied (Grove 1976, p. 35). Grove (1976) believes this is not a healthy trend and suggests rever sing the trend back to a broad role for the physician. The author recommends that selection committees need to select those who will become sound practitioners since few go into research or teaching facilities. Among all means of selection, Grove recommends random selection because it is less discriminatory; however, it would require educa tion of and acceptance by the public. Proceeding random selection would be the establishment of an applicant pool using established criteria e.g., GPA, MCAT scores, inter views and a baccalaureate degree, Baird (1976) studied the GRE (Graduate Record Examination), LSAT (Law School Admission Test) and the MCAT (Medication College Admissions Test), and found that students from the more highly educated families wealthier homes, and Jewish background tended to score higher on the exams ; women occasionally scored lower than men. The author stated that the admission tests were only a tech nical tool for use in evaluation of academic promise but 54 added that the tests were particularly efficient in pre dicting grades. Interviews for admissions were discussed in an article by Weinstein and Milgron (1977) from the stand point of the applicant’s perception. The authors devised a questionnaire to be given to every third interviewee. The study included 150 applicants to the Washington School of Dentistry, Seattle, Washington, between 1974 and 1976. The results indicated a high level of satisfaction of the interviewee with the interview. Interviewers were pre sented three two-hour workshops prior to the interview and were also given feedback following the interviews about the applicants* specific comments which had a positive impact on the 1975-76 interviewee responses compared to 1974-75 responses. An organizational model which incorporated an admissions committee for the entire School of Allied Health Sciences was developed at the University of Texas Health Center at Dallas (El-Din, 1977). The purposes of the model included: coordination of admissions process to all programs in the school, evaluation and recom mendations regarding general admission policies, advice to individual programs, facilitation of communication between departments, programs admissions committee and registrars office, consult with departments, programs and Deans regarding special admissions and adherence to requirements regarding equal opportunity admis- 55 sions, and communication of federal, state and local regulations regarding admissions, (p. 16) Each program used the same admissions form; a form was completed for each applicant which was sent to the admis sions committee * Included on the form was the category of admission which was completed at program level to be sub mitted to the committee for review. Data were not avail able regarding the number of infractions averted by this process. Experiences and comments were positive regarding the model*s use in the beginning stages. An apparently very successful automated admissions system was initiated at State University of New York at Buffalo (SUNY/B) by Robinson, Braaten, and Bailey (1979). The system was found to serve many functions regarding recordkeeping. However, its main functions were to de crease time for evaluation of records ; establish a data base ; provide more accessible information ; include more admission criteria ; and decrease faculty time spent on administrative routing. The plan for future development of the computer-assisted system included: longitudinal studies of alumni records, maintenance of academic per formance files, facilitation of degree audits, and compu tation of GPA in specific course content areas. In 1980, Shepard found that preprofessional GPA, MCAT scores, and AHPAT scores correlated positively with 56 successful achievement in pre-clinical work. Academic variables exhibited less relationship with third and fourth clinical years of medical school and performance in clinical internships. Several studies showed pre admission interviews to be better predictors for these clinical years than one or more traditional academic pre dictors, Little or no correlation was found between aca demic achievement and career performance. "GPA (as a single predictor of admission) may have little or no rela tionship to success in the clinical aspects of health care educational programs or to success in the job" (p. 86). The interview was discredited for not being predic tive of measures most often used to define academic suc cess, and for being subjective. Researchers attempted to validate the interview by correlating it with academic measures (e.g., GPA, scores on standardized tests), how ever these criteria failed. A broad point-range for interview scales (greater than four points) was suggested to further delineate among applicants. There was uncertainty as to inter-rater reli abilities on identifying traits or what impact certain traits should have on determination of final interview scores. Problems thought to exist in interviewer errors are : emotional prejudices, halo effect, stereotyping, rating toward the mean, generosity effect, unsystematic 57 use of suggestive questions, and erroneous belief in impor tance of first impressions. Certain skills such as ability to interact respon sively and communicate effectively are usually not demon strated in the one-to-one interview situation (Shepard, 1980, p. 87). Small group interviews were found to be more satisfactory, according to several authors cited in this study, allowing interviewers to view the applicant in a situation similar to that of the target occupational setting. Shepard found that the similarity hypothesis (the more similar an interviewee is to the interviewer, the more likely an assessment favorable to the interviewee will result) was not supported by data. The most relevant was how close the candidate came to exemplifying an "ideal health care professional." Interviewer status was opera tionalized according to the status hypothesis (the inter viewer with higher status will exhibit more influence on the determination of the final interview ratings than those with lower status) since most interviewers are faculty members. The study included 21 faculty interviewers and 296 applicants in three physical therapy programs. Group interviews were used (at least two interviewers and two interviewees). Applicants were rated on a ten-point scale (ten was highest). Each applicant was discussed and a 58 group score was assigned. Each faculty member's individ ual score was subtracted from the group score and a rate- diff was established. Ratediff was used as the dependent variable (values four to one, four being no difference in individual versus group score, etc.). Independent vari ables were faculty age, academic rank, number of years on faculty, and subjects taught (coded-TEACH with values five to one, five being assigned if subjects were in essential core area, three if taught in both essential and less essential areas, etc.). Discussions of applicants were taped and viewed; as a result of the viewing a new vari able was added, TALK. It was evident that the faculty members who gave the most input on a candidate had more influence on final group scores. The results of the study revealed a high correlation between the independent vari ables, so only age was used in a multiple regression anal ysis between individual and collective impact of faculty status variables. The author made several recommendations regarding continued use of the interview process; (1) small group interviews; (2) more discriminatory and broader scoring scale ; (3) agreement on weighting of traits prior to interviews ; (4) discussion of candidates by all interviewers ; (5) final scores to be composite scores. Dietrich (1981) described four sets of decisions 59 to help ensure that admission processes were rational, fair, and humane. The first was to refine the criteria on which applicants were being evaluated. Schema were devel oped to process each applicant through a series of deci sions which included : definition of selection criteria, sources of information, evaluation of data from informa tion sources, measurement of interview outcomes (based on quality of integrity, motivation, and empathy), and a réévaluation of admissions criteria. Additional factors were considered which would modify the decision making, e.g., rapid change in scope of practice to delineate entry- level skills. The second set of decisions was to choose informa tion sources which could be quantified. Examples were admitting GPA (particularly in the sciences for students in medical and dental education), standardized cognitive tests (AHPAT, IQ scores, Watson-Glazer Critical Thinking Appraisal, SAT, ACT, etc.). Reliance on these measures falters if there are "large numbers of non-traditional applicants such as Adults and Minorities" (p. 229). Quan tifying affective behaviors was identified as being the most difficult task (self-concept, self-esteem, values, attitudes, and interests). Affective standardized test development is still in the beginning stages in allied health education (Myers-Briggs Type Indicator, MBTI) (p. 60 . . m . . 230), Assessing psychomotor skills was mentioned as prob ably the most neglected area. Identifying which skills are necessary for a particular discipline is important from the standpoint of Section 504 of the Federal Rehabili tation Act of 1973- The third sj:ep was to transform data from informa tion sources into measurable form. To do this requires weighting of all applicant data. For example, in the area of demographic/biographical data, an applicant with with drawals on the transcript has ,5 points subtracted for each withdrawal (p. 233). The Q-Technique, as first described by Trotter and Fordyce (1975), is a means of ranking applicants on the basis of total behavioral and biographical/demographic profiles by placing the applicant on intervals and summing the intervals to provide a total score. The fourth and final set was to conduct evaluation of the validity and reliability of the admission criteria. Information on applicants should all be quantified so it can be placed in at least one of three levels of variables (nominal, ordinal or interval). Alumni follow-ups could be included as a variable and contain such elements as measurement of formal and continuing education, member ships in professional organizations, job mobility as indi cators of cognitive and affective growth of graduates, and 61 employer evaluations to assess the psychomotor performance of the graduate. Comparison of variables could be accom plished through use of various statistical tests as a means of evaluating admission criteria. Student selection was based on interviews, the Watkins-Glaser Critical Thinking Test, and various aca demic indicators in a study by Seymour, McDougall, Wads worth and Sanders (1982). The Watkins-Glaser test in cludes five areas : (1) inference; (2) recognition of assumptions ; (3) deduction ; (4) interpretation ; (5) evalu ation of arguments. The academic indicators included the ACT, grades in biology, chemistry and physics and overall GPA, The items were based on a 100 point system ; 45 points for interviews, 5 points for the Watkins-Glaser test, 50 points for academic indicators (ACT-30 points, grades in sciences-20 points, overall GPA-10 points). In a previous study by Wadsworth in 1975 at the University of Kentucky (Seymour et al., 1982), the ACT and GPA were found to be the best predictors of academic achievement for the last two years of professional education. The study referenced others: one from the Journal of Dental Education indicated that the primary problem with inter views was the lack of training and evaluation of the inter viewers (Weinstein & Milgron, 1977), and another by Morse in 1971 recommended that the interview be structured with 62 some predetermined questions, and criteria (cited in Seymour et al., 1982). Based on the results of the study, the University of Kentucky included the following nine areas in the inter view, each with specific criteria : (1) experience and knowledge of Physical Therapy; (2) initiative ; (3) motiva tion and perseverance ; (4) leadership and responsibility; (5) communication skills ; (6) community service ; (7) self perception; (8) interaction with others; (9) professional judgment. Each were rated 1 to 7 with 7 being the high est. The interviewers were trained, scores could be arbi trated and the interviewers had no advanced knowledge of the academic record. In 1980, the correlation between the two interviewers was excellent, with no rating difference of greater than .75 out of 65 applicants. The computer was used for analysis of data and the Watkins-Glaser test was required. The academic ranking was weighted 50 per cent, critical thinking test was 5 percent, and the inter view was 45 percent. Carline, Cullen, Scott, Shanon, and Schaad (1983) reported a study done at the University of Washington School of Medicine to examine the predictive ability of the new MCAT for course grades and ratings in clinical years, scores on comprehensive examination of clinical knowledge, and scores on Part II of the National Board 63 Medical Examiners (NBME), A secondary purpose was to discuss the MCAT scores for use in admission procedures. This study included 146 students at the University of Washington, Measurements taken were scores on MCAT, under graduate science GPA, undergraduate overall GPA, measure ment at the end of first clinical year (year three of curriculum) consisted of six subscales, total of Part II of NBME, points on patient management problem (PMP), and multiple choice of third year comprehensive (TYC), plus ratings and scores received in five basic clerkships (including clinical knowledge skill items, interpersonal relationships, and educational attitudes rated on a four- point scale). Correlations and regressions were applied to the data : t tests were used to compare MCAT and NBME Part II for groups of higher and lower rated students. There were many statistically significant correlations between MCAT scores and measures of clinical education. Although the correlations were small to moderate, correlations of under graduate GPA with clinical education were even lower. Multiple correlations of MCAT scores resulted in higher correlations than single MCAT scores ; however, correla tions were still only moderate in size. Four clinical knowledge and skills items were averaged into lower or higher ratings. These were com 64 pared to MCAT scores. Sixty comparisons were made using tests and only one comparison (MCAT reading subtest with lower vs, higher rating groups) was found to be signif icant, which was considered a chance event. As a result of the findings, the MCAT scores and student performance in the first clinical year indicated only a weak association. "The MCAT initially appears to be of little utility in predicting these outcomes of medi cal education" (Carlene et al,, p. 23). The homogeneity of the population was thought to be an explanation for the small validity coefficients ; however, even though there were no large correlation coefficients, the Carline et al,, did not believe evidence was sufficient to discon tinue to use the MCAT as an aid in student selections. More importantly, the question was raised whether the MCAT and the clinical ratings and performance measures ade quately measured the expected competencies of the first year clinical training (p. 25). Nursing "Rate of attrition from schools and colleges con tinues to present a challenge to selectors and educators" (Plapp, Psathas, Caputo, 1965). Large numbers of dropouts occur in nurse's training particularly in the first year. "Academic factors (secondary school performance, SAT scores, and academic performance in college) proved the 65 most effective predictors of dropouts" (p. 565). The authors indicated that in general correlations between intellective predictors and criteria of clinical perfor mance were lower than were correlations between intellec tive predictors and academic performance. In this study the intellective predictors selected were high school rank (HSR), Gamma Am Form of the Otis Quick-Scoring Mental Ability Tests (OTIS), and the SAT. The performance cri teria of the nursing students were continuance, academic performance, and clinical performance. It was expected that the highest correlations would obtain between variables having the most in common, e.g., HSR and academic grades in nursing school. Plapp et al, (1965) suggested that adequacy of performance of the student bears little relationship to adequacy of perfor mance of the professional, and further that this problem exists in general, "since often selection for entrance into training rests solely on measures of intellectual performance with little concern for the individual's abilities in the applied area" (p. 567). The study included 79 students of the 1962 fresh man class at the hospital nursing school in St. Louis. Significant correlations were obtained between each of the intellective predictors and at least one of the criteria of nursing school performance. None of the predictors 66 bore a relationship to fourth quarter academic grades. Correlations between intellective predictors and first quarter grades were significant at the .01 level except the SAT (significant at .05 level). Marked reductions in correlations occurred between first quarter and fourth quarter grades in all but one (between self-rating and the performance criteria). HSR did not prove to be a more effective predictor of scholastic performance, particu larly of first quarter academic performance. Correlations between intellective predictors and performance criteria revealed that SAT was the best single predictor, and HSR was the weakest. The SAT was a successful predictor of fourth quarter clinical grades: "It is of interest that a test requiring two hours and twenty minutes of performance is a more successful predictor of clinical grades than is a measure representing four years of high school perfor mance” (p. 575). Positive interrelationships occurred among the four intellective predictors. Plapp et al. (1965) concluded that tests were valid in predicting per formance in early stages of training, but there was need for development of tests that could predict performance in later stages of training and performance in the pro fession . Law A heuristic model as an initial screening device 67 for admission was proposed by Watson, Anthony, and Crowder (1973) at the University of Georgia Law School. The authors suggested that the role for a model of admissions screening was to provide information from which sound decisions could be made. The heuristic model was defined as that which simulates human thought or decision-making processes. Validity was defined as how well actions of the human counterpart could be duplicated. Computer-generated rankings were used; the depen dent variables were GPA, probable future plans (used by University of Georgia), and distribution of racial groups in the profession and between sexes. The four variables used were LSAT, undergraduate GPA, undergraduate univer sity attended, and extracurricular activities. The other variables were used in the final acceptance/rejection decisions. Weighted values were placed on the variables and applicants were ranked on one of three branches, depending on the consistency of the variables (e.g., if applicants LSAT, GPA, and weight of undergraduate school attended were consistent, the applicant would be ranked on first branch). Values were normalized by dividing the applicant score by the maximum score possible in each variable, then multiplied by the weighted value; then weighted values were summated. Watson, Anthony and Crowder (1973) concluded that the heuristic model was a 68 successful, faster, and more consistent approach to ini tial screening for law school admissions than the human thought process involved in screening applicants. Education Achievement and Predictions Giristi (1964) reported 20 separate studies that concluded the high school grades were the best single source of predicting college success. Hills and Klock (1966) studied a group of college/ university students to determine if total GPA vs academic GPA was significantly different in predicting college grades. The results of the study found very little differ ence between the two, and based on the finding, determined that total GPA was safe to use. The authors added, how ever, that it may not be safe to use averages only, but to add SAT scores to the multiple correlations. Application blanks used as predictive instruments were studied by Willingham (1965). The author reported four distinct functions of the application blank: (1) a necessary administrative document; (2) yields noncognitive information which may be necessary even though it may have no bearing upon academic performance ; (3) serves as a source document for information about matriculation ; and (4) non-cognitive information that may be useful in pre dicting the student's performance after matriculation. 69 The study consisted of two parts. One part was concerned with a scoring system for the application blank to predict grades and for use in selection; the other dealt with the prediction of voluntary withdrawal. Each item on a 35 item application blank was analyzed to determine whether there was a significant relationship between the item answer and freshman grades or if the student withdrew voluntarily. Part I demonstrated that 15 of the 35 items were significantly related to freshman grades. Achieve ment scores seemed to measure "academic motivation" more than anything else. Good items reflected a willingness to work and self-assurance. Willingham (1965) proposed an equation which yielded the best estimates : (3) HSA + (OT)CB-Sc + (.02)CB-Math + (2)AB-8.9= 10 P.G. Where HSA=high school average CB-Sc=College board science achievement CB-Math=College board math achievement AB-application blank score PG-predicted grade. (p. 277) Willingham recommended that the above equation be used only for marginal cases simply because of the increased clerical time necessary to score the application. An alternate equation was suggested for all other cases: (4)HSA + (.ODSAT-V + (.02) SAT-M - 7.5 = 10 P.G. where SAT-V and -M = Scholastic Aptitude test Verbal and Math scores, (p. 277) 70 Crossland (1965) reported on the increased costs of handling admissions from the standpoint of application fees and time for recruiting, redundant testing, filling out forms, and shuffling papers for "ghost applicants" (p. 229). The author suggested that because enrollment was rising, the public institutions might adopt the British system where a student submits only one application, list ing six institutions in order of preference to a central committee. Institutions could be grouped by geography, tradition, educational philosophy, and area of speciali zation or specific profession, and could agree on a common admissions timetable, a common application form, common set of procedures, common entrance credentials, recom mendations, secondary school transcripts, test scores, and standard pattern in awarding financial aid. All applica tions would be processed by a central agency. Cattell and Butcher (1968) wrote a section on hereditary and cultural determinants of ability in The Prediction of Achievement and Creativity, which included a study of twins, identical, fraternal, and of unrelated children. It was concluded that there was considerable hereditary determination of the "g" factor, a single general factor. In 1937 Burt, Newman, Freeman, and Halzinger (cited in Cattell & Butcher 1968) suggested that 80 percent of the variance in intelligence-test perfor 71 mance was due to heredity and 20 percent to environmental factors. Later this was questioned ; however, more recent evidence indicates a preponderant influence of heredity. Environmental variance is due to intellectual stimulation and other causes, e.g., diseases. The child develops through various stages of maturity in a style referred to as problem solving or an "instrument" or "aid." The earlier the use of problem solving, the earlier the learn ing which results in better scores on intelligence tests. Cattell and Butcher (1968) indicated that tests based on perceptual material and/or performance tests were reasonably culture free and good measures of intelligence compared to traditional verbal type of intelligence tests. Cattell and Butcher reported results of intelligence tests based on the typical life curve and found that there was a decline in scores after age 25; however, when there was no limit on time, the decline was much less. Research done by Professor L. L. Thurstone (cited in Cattell & Butcher, p. 33) listed five primary mental abilities: (1) Verbal-meaning, (2), reasoning, (3) space, (4) number, (5) word fluency. Of the five, verbal-meaning and reasoning were considered the best predictors of high school work. The Differential Aptitude Test indicated that verbal-reasoning was the best predictor of grades in science, social studies, and history, and the numerical 72 test was the best predictor of grades in mathematics and languages. At university level there is a greater need to work with little guidance and to form habits of work and study and independent thinking. Previous attainments and special aptitudes are more important than pure abilities (Cattell & Butcher, 1968). Weiss (1970) suggested a multifactor admissions predictive system using the high school rank (HSR) plus test scores to increase predictability. Formulas were used to produce non-linear results (accomplished by squaring the numerators). Each item was given a percent weight (SAT-Verbal, HSR, and IQ). The results indicated a high correlation between admission data and GPA for first year of college. The advantages of the system were that it showed a difference between students, it was a rela tively complete profile, it was a means of evaluating large groups, and it developed a unique score. The ques tions unanswered were the nature of specific weights to use for each item and other factors that might need to be included in the system. The interview was studied by Burgess, Calkins, and Richards (1972) as a selection device. The authors indi cated that interviews were an integral part of most if not all medical school admissions procedures; however, the 73 interview had a higher correlation with internship grades than with general achievement in clinical medicine. Bur gess et al, recommended a structured interview with spec ific format and content areas for either physician or non-physician interviewers. Both formats included the applicants perception of medicine. The physician format included ten item in personal attributes and the non physician format included ten items in work habits. Appli cants were scored on a scale of one to six, six being the highest (outstanding) score, plus a score reflecting whether the interviewer would recommend acceptance, rejec tion, or was undecided for each applicant. The results revealed rating of applicants was on the basis of a more global impression rather than individual items. The level of agreement between interviewers was low on all items. Sternburg (1973) did a cost benefit analysis of interviews at Yale. Costs in money were found not be be a major problem; however, costs in time were high. The author suggested that one of the functions the inter viewers attempted to perform was to try to provide infor mation about applicants which was useful, unique, and valid. The results indicated low correlations between interview ratings and other admissions variables, high correlations between interview ratings and student folder reader ratings, and fairly high correlations with the 74 recommended actions. Following the interviews applicants were asked to respond to a questionnaire, most were satis fied with the interview process. Fifty-two percent indi cated at least a moderate amount of information was exchanged, most felt comfortable in the interview, and most felt the interview made a favorable impression on them. In summary, the interview provided potential bene fits to the extent that the author recommended that it be continued. A study was done at the U.S. Air Force Academy by Westen and Lenning (1973) to compare the predictive abil ity of the American College Testing Program (ACT) versus SAT. The differences between the two batteries are many; one is curriculum oriented the other is factor oriented. The ACT has an entire subtest on writing skills, the SAT does not. The ACT attempts to measure complex reasoning emphasizing rigorous, precise, quantitative, and objective reasoning and reasoning leading to value judgments, where as the SAT is more largely concerned with abstract reason ing. The authors found the ACT to be as predictive, if not more predictive, than the SAT at the Academy. A study of Munday (1965) was cited by Westen and Lenning where the ACT was found to be even a better predictor than the SAT for women in particular. The ability to write, express oneself, and recognize biases were emphasized at highly 75 selective colleges as well as more typical colleges (p. 75). The relationship between cognitive style and achievement was studied in a sample of 275 fifth-grade children (Gray & Kneif, 1974-76). The Lorge-Thorndike Intelligence Test, the reading and mathematics section of the SAT, the Conceptual Style Test, and the Sigel Cogni tive Style Test were used in the analysis. Girls obtained higher scores than boys in six of eight variables, the exceptions were descriptive style and mathematical skills. Of interest also was the statement in the summary of the article "That the relationship between cognitive style and school achievement changed from one classroom setting to another" (p. 71). Barriers of admission to higher education were identified to the Council on Higher Education in American Republics (Crossland, 1976) as follows; financial re sources, excessive distance from home to institutions of higher learning, sex discrimination, inadequacy of lower schools to provide academic preparation, prejudice against certain racial, religious, or political minorities, unfair culturally biased standardized entrance exams, insidious counseling of secondary students, physical (not mental) disabilities that inhibit locomotion, age discrimination, and undue emphasis on communication skill requirements. 76 Crossland stated that there were two valid reasons for entrance exams: one, if there are more applicants than available spaces so it can be a screening device : two, it may be used for diagnostic and placement purposes. All tests have in common the following; measurement and high importance on verbal skills, scores reported on a relative scale, and abilities and skills purported to be measured distributed on a bell curve. The author's criticisms of tests were that ; they do not measure certain skills, e.g., manual dexterity, physical coordination, or crea tivity ; they have an unfair emphasis on communication skills ; and that rank ordering is unfair in that indi viduals should be compared against themselves and prior performance rather than compared with others. Precautions to be taken when utilizing an examin ation according to Crossland are; (1) does the test actually measure what it purports to measure, (2) admin ister test in secure and appropriate conditions, (3) test items worded fairly so as not to discriminate, (4) proper interpretation of test, (5) modify interpretation as cir cumstances require, (6) recognize certain test taking skills and that prior experience may be an advantage, (7) do not overstate the utility of the test, and (8) do not attach undue significance to the test (p. 14). Kerr (1978) stated that there was a serious issue 77 of "reverse discrimination" in college issues even before the Bakke case (challenged the right to equal opportunity vs. right to equal protection under the law) (p. 4). The admissions policies to the so-called 'gate keeper* schools, such as law, medicine, den tistry, exert a powerful and often controlling influence on who may practice certain critical professions in this country, and also control access to positions of influence and higher economic and social reward, (p. 4) The Carnegie Council believes that an applicant who meets the required academic standards for success in college or university is relevant to admissions criteria. Race should be considered in selection. "In any race, if each runner begins at a different starting line, how can choosing the winner by judging who finishes first be a fair measure of accomplishment" (p. 5)? Judging how far a person has come and what obstacles have had to be overcome in the process deserves consideration. Kerr recommended a two-stage admission procedure: first, eliminate appli cants who do not meet the minimum standard (should be the lowest level possible to have a reasonable chance to succeed); and two, consider the racial experience along with scholastic test scores, grades, special interests or abilities, special skills, special identities, e.g., relig ious or alumni affiliation, ability to contribute to the diversity of the student community, etc. Kerr did not recommend using quotas or goals, and did recommend flex 78 ibility in admissions policies for judging individual applicants. Consideration of special characteristics is a major means for society to become more just and more inte grated . Culture and Socialization Cattell and Butcher (1968) reported a study done by McClelland (1961) on strength of achievement motivation in a culture by rating the content of stories found in school textbooks and assessing the extent to which the content reflected a preoccupation with achievement meas ured by the economic growth. The conclusion was that genetic factors and climate were not sole causes of differ ences in cultures, and that pattern of child rearing and socialization within a culture were also causes of differ ences. There was high correlation between progress and achievement of cultures and temperate climate conditions. Strong evidence was found cross-culturally that an author itarian and restrictive approach by the father lowers the need for achievement. "Need for achievement" includes self-assertive erg, self-sentiment, super ego, and some times pugnacity (p. 222). Related to achievement levels are three culture pattern dimensions; cultural pressure, enlightened affluence (or affluence-education), and morale. Cultural pressure, which included urbanization, complexity of 79 living, suicide rate, and frequency of involvement in war, was especially correlated with creativity in the arts and sciences. Innate differences, biological variations, natural selection, and migration accounted for some corre lations of intelligence test scores, standard of living, and school achievement (p. 235). A "gesellschaft" group, a German word meaning organization or society, includes schools and non-family units, which are formal and impersonal groups (Schwartz, 1975). "The prime, formal socializing agents within schools are teachers" (p. xvi). Classmates, particularly high-achieving classmates, also have an impact on social ization of an individual. School allocation processes provide social mobility for a few but at the same time maintain the status quo for many. Schools sort and select and, by so doing, provide personnel to fill needed roles in society, although, in the process, "define the life chances of individuals" (p. 161). Schools, depending on the neighborhoods in which they are located, differ in their socialization objectives. Some place emphasis on high levels of cognitive achievement, others emphasize employability immediately upon leaving school (p. 169- 170). Access to higher education has increased for all with the growth of community colleges. Many institutions actively recruit students whose subpopulations have been 80 under-represented, "Graduate schools are also attempting to provide fuller educational opportunity" (p. 177) • "Socialization in families and schools determines more than any other factor a young person's capacity to benefit from both the democratization of higher education and its resultant opportunities in the larger society" (p. 178). The results of a study done by Gordon, Arvey, Daffron, and Umberger (1974) indicated that whites bene- fitted more than blacks following a training program in mathematics at a manpower development program in eastern Tennessee. Several factors were suggested as possible explanations; both races did profit from training, how ever, variables that need to be examined are the role of socioeconomic status, nature of the learning environment, and motivation to perform well on tests, and/or the will to achieve in the training program. The authors refer enced Jencks' study on social stratification and higher education (1968) from the standpoint of the significant correlation between educational achievement and socio economic status. Lyman (1975) studied the Oriental cultures in North America contrasting the historical patterns of Chinese versus Japanese socialization. Whereas the Chin ese remained in unusually persistent social isolation, the Japanese rapidly acculturated. With a few exceptions, the 81 Chinese tended to assume occupations that kept them in dependency; the Japanese, after a brief period as labor ers, began to pioneer independent businesses. McGrath (1983) indicated that the Asian-Americans were only about 1.5 percent of the population. Out of 40 Westinghouse Science Talent Search finalists, nine were born in Asia, three others were Asians. Ten percent of Harvard's freshman class is Asian-American. No more than 15 percent of California high school graduates are elig ible for the University of California system, 40 percent of the Asian-Americans qualify. Asian-Americans seem to excel in math. In California 68 percent of Japanese-born students scored over 600 on the SAT out of a possible 800; 66 percent of students born in Korea also scored over 600. Some factors attributing academic success by different authors have been genetic superiority, superior perfor mances on tests of block design, mazes and picture arrange ment (possibly related to complex ideograms of their alpha bets) nurture being more emphasized than nature, high Asian income levels, high regard for education as an avenue to recognition and success, pressure to work and respect for education, and the Confucian ethic (achieve ment is the only way to repay the debt to parents, of showing filial piety (McGrath, 1983, p. 52). Cultural discrimination through testing was dis- 82 cussed by Samuda (1971). Several studies support the fact that blacks score lower on tests of mental ability than do whites. The reasons suggested for this fact have been the subject of controversy for several years. The matter of bias and relevancy of tests results have become the cen tral issue. The Stanford-Binet Intelligence Scale "proved to be of doubtful validity in evaluating the intelligence of foreign born or Negro children or for comparing their intelligence with that of native white children" (p. 492). Inappropriate use of the tests and the consequence of such use is also at issue. In an attempt to answer some of the questions raised by these issues, much research has been done to examine the effects of the test environment on different groups, the effects of language differences, and the self-fullfilling prophesy. The author visualized two major trends in expansion and elaboration of objective tests ; One that seeks to retain the concepts of apti tude or fitness to perform in future situations from the results on a standardized set of behav ioral task; and the other, that emphasizes the purposes and goals of testing as essentially descriptive and prescriptive leading from an analysis of functional levels and cognitive styles to the prescription of learning experi ences matched to the individual needs of each student, (p. 499) Samuda indicated the emphasis on selection and prediction should be decreased in order to facilitate equal educa 83 tional opportunities rather than to preserve the elite (pp. 501-502). Haney, Michael, and Martois (1976) studied three ethnic groups (223 Caucasians, 73 Mexican-Americans, and 67 Negroes) to determine if there was any relationship between 15 predictor variables and 13 criterion variables as predictors of success. Mean scores were higher in every variable except age for the Caucasian sample. Differences between the other two ethnic samples were small, with Mexican-American being slightly higher. The single most valid predictor variable for all three groups was the Reading Vocabulary subtest of the California Achievement Test. Achievement Differences Between Males and Females Differences in the subject of mathematics is the areas of focus of much research in studying the differ ences between males and females. Hilton and Berglund (1974) reported a longitudinal study that focused on sex- typed interests as a possible cause of differences in mathematical achievement. The literature seems to agree that "girls usually do better in verbal and linguistic abilities, while boys generally do better in numerical and spatial aptitudes and tests of arithmetical reasoning" (p. 231), and the differences remain among college students 84 and adults. Problem-solving differences were thought to be due to the tendency of men to take a more analytical approach where women tend toward a global field approach. Attitudes toward problem solving was suggested as another reason for sex differences. The authors included another explanation for differences may be the differences in role expectations, which was termed sex-typed interests The study included a large sample (over 1,700) where the amount of training in math was held approxi mately constant. The results indicated that arithmetic means and standard deviations increase with age. Students who enrolled in academic programs in high school had high er scores than those enrolling in non-academic programs, and males and females were equal at grade five. However, there were statistically significant differences, with males scoring higher at grades seven, nine, and eleven for the academic group in Step Math (Sequential Test of Educa tional Progress). Differences in the non-academic group were only significant at grade eleven. The Background and Experience Questionnaire (BEQ), a special instrument for the Growth Study, was developed. The results indicated in general that males read more books and magazines on science, were more interested in math, felt math was more useful to them for their future, discussed science more with their peers and family, and that differences between 85 males and females were almost negligible at grade seven but increased with age. One final observation the authors made was that the data indicate a close relationship between perception and performance in math (p. 237). Astin (1975) studied sex differences in mathe matically and scientifically precocious sixth grade and junior high students. The purpose of the study was to identify those who had special talents in math and science and to provide programs to enhance their development. The results indicated that boys scored higher in math tests than did girls and the differences increased with age. The author ascertained differences in occupational inter ests : girls chose investigative, artistic, and social types of occupations and boys chose investigative, enter prising, and realistic (e.g., airplane mechanic, elec trician, etc.) occupations using Holland's (1965) Check list of Occupations. In the study not one girl admitted she did not like school, more girls than boys liked school, over time boys' liking for school decreased whereas more girls indi cated their liking increased. An interesting result re lated to math indicated that dislike for school had a positive relationship with SAT mean scores on mathematics independent of sex (p. 86). (The more precocious boys and girls were less pleased with school.) 86 Child rearing practices were studied by Astin and the results indicated no significant difference if the mother of the child was working or nonworking. There was a direct relationship between the parents* education and the mean score on the SAT-Math: the greater the parents education, the higher the mean score. Lastly, the study included the parents* perceptions of the precocity of the child, and the results indicated boys showed interest in math and science earlier than girls. Parents rated girls higher on conscientiousness, sociability, tenderness and sympathy. Both boys and girls were described as very likable children. A commentary on the above study was done by Anastasi (1975). The author agreed with the tendency of the project to move away from identifying outstanding performance by means of "IQ" and instead identify by par ticular areas of developed ability. Anastasi viewed pur poseful intervention (environmental manipulation) to be an appropriate means of incorporating the knowledge of the influence of heredity and environment on behavioral devel opment. However, the author cautioned the writer of the project to take great care in making statements pertaining to heredity and environment to the readers. Verbal apti tude has generally proved to be the best predictor of scholastic performance in most academic courses. Of inter- 87 est to Anastasi was the suggestion that psychologically mathematics was a "closed system," while verbal learning was more "open-ended" (p. 96). Verbal learning requires more external contacts than do mathematical abilities. Difficulty level of the items on the test of problem solv ing may be related to the subject * s response to the items from the standpoint of the individual’s experiential back ground. The author also commented on an earlier study by Astin which indicated that high mathematical aptitude was the best predictor of career plans in science, teaching, or other professions (p. 100). A final comment by the author brings up the issue of the Hawthorne effect and a self-fulfilling prophecy. "Positive reinforcement is a well-established principle of operant conditioning" (p. 102) . Summary of Literature Review Allied Health Professions Admission Test The AHPAT was designed for use in admissions of upper-division students into allied health professions beginning in 1972 by the Psychological Corporation, Pro fessional Examination Division. Following a trial adminis tration, standardization and validation of the test was done using follow-up data from the three schools involved in the initial administration of the test in 1974. Sever- 88 al studies reported in the review, and the conclusions drawn regarding the AHPAT were: 1. the test was not as good a predictor of suc cess as preprofessional GPA and ACT math scores (Schimp- fhauser & Broski (1976). 2. preprofessional GPA correlated consistently with professional GPA (Katzell, 1977). 3. AHPAT scores correlated significantly with preprofessional GPAs, such that the scores could provide (Katzell, 1977). 4. the AHPAT discriminatory ability in selection was questioned, however, GPA proved to be a good predictor (Laurencelle, Kay, & Edelsberg, 1979). 5. the AHPAT was significantly correlated with GPA and improved predictions of GPA when used in conjunc tion with other factors (Leiken & Cunningham 1980). 6. acceptance GPA ranked highest when compared to cumulative GPA, while AHPAT-reading comprehension ranked fourth. When the analysis was done to determine the best predictor of cumulative GPA, the best predictor was accep tance GPA followed by AHPAT-Biology score. When analyzed by correlating all variables with professional board exam, acceptance overall GPA ranked first, AHPAT-chemistry ranked second and AHPAT-verbal reasoning ranked third. The conclusion was that the AHPAT offered more help in 89 prediction of board examination scores than in predicting GPA; however, the AHPAT was retained as one method for evaluating success of prospective applicants (Wiesseman, 1976). 7. the AHPAT had little apparent value, and did not have strong applicability to physical therapy (Soder- burg, 1983). Professional Education- Admissions/Selections The use of specific tests for admissions, GPA, interviews, autobiographies, and letters of recommendation have been the most common methods of selection in pro fessional schools. The majority of the schools utilized GPA and interviews. Semi-structured to structured inter views were found to be most successful. GPA for prereq uisite science courses was found to be the most signif icant factor in selection by Landen (1977). The DAT was not found to be a good predictor. MCAT science scores were found to correlate highly with high pre-med grades (Landen, 1977). A high correlation was found between academic success and clinical success (Landen, 1977). A screening form, which included academic record, letters of recom mendation and biographical information, was found not to be significant predictor. However, prerequisite GPA and 90 overall GPA were found to be the best predictors (Green & Cox, 1978). Occupational Therapy The Occupational Therapy Career Inventory was highly correlated with clinical performance (Johnson, Arbes, & Thompson, 1974). ACT scores correlated sig nificantly with academic course performance and the occupational therapy registration examination. High corre lations also were found in specific core courses and the registration exam. Johnson et al. concluded that a multi factor method of selection was preferable. Allied Health/Medicine Student selection is a dilemma in other allied health professions and in medicine as well. Problems which face them are much the same as in Physical Therapy education. Some of the same issues are GPA, MCAT scores, computer-based selection, interviews, predictability of various tests, minority admissions, grade inflation, ram pant cheating, and specialists vs. generalists. Law A heuristic model was proposed by Watson, Anthony, and Crowder (1973) which utilized computer-based rankings utilizing processes simulating human thought and decision making. Weighted values were assigned to the variables and applicants were ranked. 91 Education Student admissions to postsecondary education has involved predictions of success as a means of selecting students for college/university level study. High school GPA was found to be the best single predictor of success by Giristi (1964), Others suggested adopting the British system. The influence of heredity, environment, age, and culture-free testing was discussed as each plays a role in admissions. Admissions decisions have been based upon: (1) multifactor predictive system, (2) interviews, (3) admissions tests, (4) cognitive style tests, (5) financial resources, (6) sex discrimination considerations, and (7) racial, political, or religious minority discrimination considerations as well as "reverse” discrimination. All were discussed in relation to their influences on admis sions committees. Culture and Socialization Cultural factors and socialization factors were briefly reviewed to determine which, if any, were the greatest influence on college achievement. The students* environment from early school years was discussed as a major influence on future patterns in school. Comparisons between some racial groups, discrimination through test ing, difference between the two sexes, role expectation differences, and math skill differences were mentioned as 92 factors which might influence achievement differences. In summary, the literature review was designed to point out the multifaceted problems of student admissions and selection for postsecondary education from a multi disciplinary standpoint. Many problems were the same and, interestingly, approaches to the solutions of the problem were very similar. 93 CHAPTER III METHODOLOGY Background The purpose of this study was to determine if there was any relationship between scores on the Allied Health Professions Admission Text (AHPAT) and criteria selected from both pre-admission and post-admission data for students at California State University Northridge, Physical Therapy Curriculum. The data were collected using two methods: (1) examination of academic files between the years 1975 and 1978, and (2) a brief ques tionnaire sent to all students to obtain State Board scores for those whose California Board of Medical Exam iners Physical Therapy Licensing Examination Scores were not identified by student name. (Beginning with the July 6, 1979 examination, the Family Rights and Privacy Act precluded scores being reported by student name.) Data were collected from the record of each stu dent as follows: 1. sex 2. age (at application to professional program) 94 3. race 4. marital status 5. preprofessional major 6 . twelve preprofessional prerequisites a. Chemistry 1 b. Chemistry 2 c. Precalculus or d. College Algebra and/or e. Trigonometry f. Biology g. Anatomy h. Physiology i. Physics 1 j. Physics 2 k. Biostatistics 1. prerequisite grade point average 7. five didactic course grades a. Anatomy b . Physiology 0 . Biomechanics d. Neurology e. Psychology 8. overall professional program grade point average 9. success in internships (as measured by passed all internships or repeated one or two internships) 95 10. total score on Physical Therapy Licensing Examina tion plus three separate sections a. Part I--Basic Sciences b. Part II--Clinical Sciences c. Part III— Theory and Procedure 11. total score on Allied Health Professions Admis sions Test plus five separate sections a. Verbal Ability b. Quantitative Ability c. Biology d. Chemistry e. Reading Comprehension. Population All 194 students who completed the professional curriculum (program) in Physical Therapy at California State University Northridge (CSUN) between fall 1976 class and including the fall 1981 class were included in this research. As noted earlier, the AHPAT was first adminis tered in 1974 and was not required at CSUN as part of the selection criteria until the application period for the fall 1976 class (spring 1976). One hundred ninety four students represent 59 percent of 331 who entered the pro fessional physical therapy curriculum at California State University Northridge between fall 1968 at the programs inception and spring 1980. The spring 1980 class (Class 96 18) was the last group of students to have completed the program and taken the State Board Exam at the time the data were collected. Statistical Methods The Biomedical Data Package (BMDP 2D, BMDP 3D, and BMDP 6D) published by the Health Sciences Computing Facil ity, University of California, Los Angeles, 90024, (Copy righted 1977) Regents of California, were used in the analysis of the data. All of the data were analyzed using the Control Data NOS 750 computer at California State University, Northridge, California. Frequency distributions (BMDP 2D), simple linear regression (BMDP 6D), and student t tests (BMDP 3D) were applied to the data. Frequency distributions were constructed to in clude: sex, age, class, ethnic background, marital status, pre-physical therapy major, pre-physical therapy grades in chemistry (first and second semester), pre calculus, algebra, trigonometry, general biology, anatomy, physiology, physics (first and second semester) and bio- statistics, number of English courses completed, prereq uisite GPA, grades in the professional program in anatomy, physiology, biomechanics, neurology, and psychology, GPA at completion of professional program, success in intern ships, total scores and Part I, II and III of the 97 California State Board of Medical Examiners Exam, five scores on the AHPAT (verbal, quantitative, biology, chem istry, reading comprehension) and the total AHPAT score. BMDP 2D was applied to plot frequency distributions. The mean, median, percent of total number in the category and, where appropriate, the range and/or standard deviation were computed for each of the major factors, the profes sional curriculum prerequisites, selected professional curriculum courses and criteria, and scores of the Cali fornia State Board of Medical Examiners Licensure Exam ination . Simple linear regressions were applied using the BMDP 6D program. An alpha level of significance of .0005 for the simple linear regressions was applied to the data to test whether or not there was a correlation between prerequisite grades and the AHPAT, as well as age at which the test was taken and the AHPAT. "As a dividing line between value of P (probab ility of values being above or below the set value) that are small enough to cause us to reject the null hypothesis and those which will allow us to accept it. A signifi cance level is usually chosen in advance for the test" (Dunn & Clark, 1974). 98 It is possible to make the level of significance extremely small in order to reduce the possibility of rejecting the null hypothesis mistakenly. When several tests are made from the same data, they should be made with a significance level which tests whether differences exist among population means. The alpha .0005 level of significance was util ized; the reason for using this level was that multiple tests were performed on the same data, and the probability of finding statistical significance at levels above that is very high when running many tests (Dunn & Clark, 1974). The next step was to test whether or not the AHPAT and the Board Exam correlated with program grades, success on internships, and professional program GPA. The last step of the regression analysis tested whether or not the prerequisite GPA or age correlated with Board Exam scores. Correlations were done between full groupings : (1) sex; male or female, (2) race; Caucasian or other (because of small numbers of specific other races, (3) marital status; single or married, (4) majors; Health Science or other (because of small numbers of specific other majors), and (5) the number of English courses; minimum of one or more than one. The final statistical method applied to the data was the student t test (BMDP 3D) to determine if there was 99 any difference between means for GPAs of less than or equal to 3.09 vs. greater than 3.09, female vs. male, Caucasian vs. all others, and Caucasian vs. Oriental. The .05 level of significance was used in the analysis for the data in this statistical method. 100 CHAPTER IV ANALYSIS OF THE DATA The following information represents the data collected from 194 student records in the Physical Therapy Curriculum at California State University Northridge (CSUN). One hundred ninety four students represented 59 percent of the total 331 students who had completed the professional program and taken the licensure examination at the time the data were collected. Results of Frequency Distributions Frequency distribution were done to determine the means and standard deviations for each factor. The data were first analyzed individually using BMDP 2D (see Table 3). The major factors were: (1) sex, divided into male or female and combined, (2) age, (3) ethnic background, divided into Caucasian, Black, Mexican-American, and Oriental, (4) marital status, divided into married or single, and (5) preprofessional major, divided into Health Science or non-Health Science. Several professional curriculum prerequisites required at CSUN in which course content is related to the subject content of the AHPAT subscores are (1) chemistry— 101 Table 3 Frequency Distributions, Total Sample Categories n Mean Median % of Total Range Sex Male Female 67 127 35 65 Age 194 22.9 22 28.4 19-42 Class 10 13.5 13 9-18 Race Caucasian Black Mexican-Amer. Oriental 173 4 4 13 89.2 2.1 2.1 6.7 Marital Status Single Married 155 38 79.9 19.6 Pre-P.T. Major H.Sc. NHSc 148 46 76.3 23.6 Number of pEng. Course 1 Course 2 66 65 1.9 2 34.0 33.5 0-9 pGPA 192 3.3 3.2 2.60-4.00 GPA 194 3.5 3.5 2.24-4.00 Internship Passed All Repeated 1 189 5 1.0 1.0 1-2 102 1st semester (p Chem 1); (2) chemistry— 2nd semester (p Chem 2); (3) precalculus (p Math; (4) college algebra (p alg); (5) trigonometry (p Trig); (6) general biology (p Bio); (7) anatomy (p Anat; (8) physiology (p Physiol); (9) physics— 1st semester (p Physics 1); (10) physics— 2nd semester (p Physics 2); and (11) biostatistics (pBiostat). The distribution of grades received in the above named courses appear in Table 4. Students had a choice of either precalculus or two courses, college algebra and trigonometry. Courses in the professional curriculum most direct ly related by course content to the separate scores on the AHPAT are anatomy (Anat), physiology (Physiol), biomechan ics (Biomech), neurology (Neuro), and psychological as pects of disability (Psych). Distributions of grades in the above named courses appear in Table 5. The GPA for classes in the professional curriculum ranged from 2.24 to 4.00, with a mean of 3.46, and a median of 3.49. Success in internships was rated and 1 if the student was successful in all internships, or 2 if the student had to repeat one internship. No one had to re peat more than one internship in this sample. Those scoring 1 consisted of 97 percent of the total and those who scored 2 consisted of the remaining 3 percent. 103 Table 4 Distribution of Prerequisite Grades Grades' Class Values Counted Mean Medium 4 3 2 pChem 193 3.06 3 34.7 36.3 29.0 pChem 2 194 2.86 3 25.3 35.1 39.7 pMath^ 48 3.04 3 35.4 35.4 27.1 pAlg^ 138 3.12 3 40.6 32.6 25.4 pTrig^ 27 2.93 3 18.5 55.6 25.9 pBio 194 3.21 3 37.6 45.4 17.0 pAnat 194 3.5 4 56.7 37.6 5.7 pPhysiol 194 3.34 3 44.3 45.4 10.3 pPhysics 1 194 2.92 3 24.7 42.3 33.0 pPhysics 2^ 194 2.81 3 20.1 42.3 37.1 pBiostat 194 3.11 3 36.1 39.7 24.2 Percent in Category ’Some scores were missing where there was no record of the student(s) taking the course. Grades were recorded on a 4-point scale with 4=A, 3=B, 2=0, 1=D. Students were not eligible for submitting their application if they had less than a C with the exception of three cases where a grade was recorded as less than a C in pre-calculus, college algebra, and second semester physics. 104 Table 5 Distribution of Grades in Physical Therapy Curriculum Grades' Course Values Counted Mean Medium 4 3 2 Anat 194 3.56 4 62.4 32.0 5.7 Physiol 194 3.37 3 42.8 52.1 5.2 Biomech^ 193 3.44 4 50.3 43.0 6.7 Neuro^ 192 3.34 3 46.4 41 .7 12.0 Psych^ 192 3.48 3 49.0 50.0 1.0 Percent in category. 'values missing represent students who had equivalents to the course prior to admission into the professional program. 105 The licensure examination scores were not avail able for 64 of the 194 in the total sample. The values not included were the scores that were not received in responses to a request form mailed to those whose scores were not senLt to the school identified by student name. (See Appendix 1.) The California State Board of Medical Examiners Physical Therapy Licensure Examination consists of three parts reported as Total Score (for all parts), Part I. Basic Sciences, Part II, Clinical Sciences, and Part III, Theory and Procedure. The scores are reported in percentiles and in the State of California the examinee must score 75.00 or better in each part of the examination to pass that part. The distribution of scores available for 130 students, are illustrated in Table 6. Results of Simple Linear Regressions Simple linear regressions were done using BMDP-6D to see if there were any variables that could predict performance on the AHPAT at an alpha level of significance of .0005. The AHPAT consists of five separate scores and a total score; hereafter the scores are identified as follows: AHPAT-Verbal, AHPAT-V; AHPAT Quantitative, AHPAT-Q; AHPAT Biology, AHPAT-B; AHPAT Chemistry, AHPAT-C; AHPAT Reading Comprehension, AHPAT-RC; and AHPAT Total, AHPAT-T. 106 Table 6 Distribution of Scores on Physical Therapy Licensure Examination (n=130) Mean Median Range Standard Deviation Total Score 82.54 82.25 75.7-87.8 2.64 Part I 82.90 83.10 66,0-88.7 3.18 Part II 82.29 82.20 74.5-89.0 2.94 Part III 81.99 82.00 75.9-88.0 2.71 The major factors compared were pGPA, AHPAT-T, GPA, and total score on the licensing examination. The results were as follows: AHPAT-T GPA Lie. Exam pGPA .375* .554* .381* AHPAT-T — .265* .474* * significant at alpha .0005 level The data above indicate that all the major factors were significantly related. The hypotheses that follow analyze each of the components which make up each of the major factors by comparing the five sections of the AHPAT and the AHPAT Total score to each of the components of pre requisite requirements, five selected professional curric- 107 ulum courses and licensure examination results when in five subgroups. The subgroups were sex, age, ethnic back ground, marital status, and major. The first step (Stage 1) was to see if there were any correlations between preprofessional prerequisites or age and scores on the AHPAT when in groups. The groups were sex, age, ethnic background, marital status, and major. Hypothesis 1 Hypothesis one states that there is no relation ship between scores on the AHPAT and the individual’s performance in certain required preprofessional courses as measured by grades. Correlations between prerequisite grades and AHPAT scores showed the following significant relationships: (1) pPhysiol and AHPAT-B, (2) pPhysics and AHPAT-C, (3) pBiostat, and AHPAT-Q, AHPAT-C and AHPAT-T. There were no significant correlations between preprofes sional courses and AHPAT-V or AHPAT-RC. The results are shown in Table 7. In the case of pPhysiol, subject matter in the course should be related to the biology scores on the AHPAT. Problem-solving skills required in pPhysics 2 would be expected to be related to quantitative ability. Only 5 pre professional criteria out of 14 used for this comparison show any relationship to AHPAT scores for the 108 Table 7 Hypothesis One Correlations Between Preprofessional Criteria and AHPAT Scores in Overall Groups (n = 194) AHPAT Scores Criteria^^ V Q B C RC T pChem 1 (n=193) -.087 .178 .033 .213 .025 .096 pChem 2 .053 .162 .149 .192 .048 .135 pMath (n=48) .077 .430 .019 .263 .122 .275 pAlg (n=138) .118 .297 .143 .160 .176 .252 pTrig (n=27) .298 .028 .080 .437 .286 .324 pBio .062 .066 .122 .031 .023 .095 pAnat .001 .193 .152 .170 .170 .159 pPhysiol .161 .188 .328* .142 .142 .183 pPhysics 1 .015 .219 .152 .204 . 066 .183 pPhysics 2 .006 .251 .140 .253* .130 .218 pBiostat .021 .305* .111 .272* .205 .254* pGPA .174 .365* .284» .278* .210 .375* Age .254* -.245 .068 .172 .151 — .062 Note- n = 194 unless soeeified ^Preprofessional Criteria p values for n - 194 = .253, .299, n - 48 = .503, n - 27 n - 193 = 589. = .254, n - 138 = * _ = significant p. <.0005 109 overall groups. There were only 5 of the 66 possible factors that were related. Hvpothesis 2 Hypothesis 2 states that there is no relationship between scores on the AHPAT and the individual’s success in the didactic education portion of the professional physical therapy curriculum as measured by grades. The five courses in the professional curriculum used were: (1) Anatomy, (2) Physiology, (3) Biomechanics, (4) Neurology, and (5) Psychological Aspects of Disability. The compari son was the second step (Stage 2) in the analysis of the data: the total group and subgroups were analyzed. For the total group there were significant rela tionships between Physiol and AHPAT-Q, AHPAT-RC and AHPAT- T, between Psych and AHPAT-B and AHPAT-T, and between GPA and AHPAT-Q and AHPAH-T. The remaining results are shown in Table 8. The results for subgroups were as follows: 1. Grouped by sex, males— no significant relation ships; females— significant relationships between GPA and AHPAT-Q and AHPAT-C. The results are shown in Table 9. 2. Grouped by ethnic background, Caucasians— significant relationships between Physiol and AHPAT-Q, and between GPA and AHPAT-Q and AHPAT-RC; other--no signif icant relationships. The results are shown in Table 10. 110 Table 8 Correlations Between GPA in the Professional Program and AHPAT (n Scores for = 194) Total Group AHPAT Scores Criteria^ V Q B C RC T Anat -.119 .236 .110 .224 .200 .176 Physio -.028 .287 * .238 .181 .266* .260* Biomech -.125 .224 .162 .250 .183 .188 Neuro -.105 .169 .021 .140 .175 .108 Psych .105 .170 .254* .196 .212 .268* GPA —. 026 .291 * .174 .217 .296* .265* Note. D value for n-194 = .253 Professional program criteria * significant at 2. <.0005 - inverse relationship Ill Table 9 Hypothesis Two; Correlation Between Professional Program Courses and AHPAT Scores When Separated by Sex (Males=67, Females=127) AHPAT Scores Group Criteria^ V Q B C RC T Male Anat -.076 .146 .144 .112 .292 .182 Female Anat — .146 .304 .099 .297 .143 .176 Male Physiol -.004 .293 .244 .091 .289 .272 Female Physiol -.034 .276 .232 .216 .275 .254 Male Biomech — .160 .249 .239 .220 .314 .253 Female Biomech -.121 .232 .131 .297 .093 .158 Male Neuro -.242 .076 -. 028 .002 .189 .005 Female Neuro —. 063 .273 .064 .265 .127 .170 Male Psych —. 038 .257 .246 .186 .261 .269 Female Psych .172 .132 .266 .226 .173 .272 Male GPA -.007 .317 .154 .135 .317 .270 Female GPA -.053 .319* .202 .309* .256 .272 Note. 2 value for n-127=.308, n-67=*395 ^Professional curriculum criteria * significant at 2 <.0005 - inverse relationship 112 Table 10 Hypothesis Two: Correlations Between Professional Program Courses and AHPAT Scores When Separated by Ethnic Background (Caucasians=173, 0ther=21) AHPAT Scores Group Criteria V Q B C RC • T Caucasian Anat -.129 .255 .114 .225 .204 .184 Other Anat -.120 .105 .085 .202 .182 .116 Caucasian Physiol -.075 .306* .255 .177 .257 .257 Other Physiol .318 .123 .100 .196 .344 .280 Caucasian Biomech -.177 .198 .138 .230 .166 .150 Other Biomech .267 .485 .412 .417 .393 .497 Caucasian Neuro -.145 .154 .024 .123 .168 .087 Other Neuro .076 .246 - .027 .209 .253 .193 Caucasian Psych .083 .163 .204 .163 .199 .238 Other Psych .299 .231 .645 .437 .317 .489 Caucasian GPA -.074 .287* .169 .200 .282* .243 Other GPA .190 .282 .203 .281 .445 .358 Note. 2 values for n-173=.270, n-21=.653 Professional curriculum critera * significant at jq<.0005 - inverse relationship 113 3. Grouped by marital status, singles— no signif icant relationships; married— significant relationships between Anat and AHPAT-Q, between Physiol and AHPAT-T, and between GPA and AHPAT-Q, AHPAT-Q and AHPAT-T. The results are shown in Table 11. It should be noted that the married group consisted of 38 out of the total sample. 4. Grouped by major, there were significant rela tionships for health science majors between Physiol and AHPAT-Q, AHPAT-RC and AHPAT-T, and GPA and AHPAT-Q, AHPAT- RC and AHPAT-T. There were no'significant relationships for non-health science majors. The results are shown in Table 12. 5. Grouped by number of English courses taken, minimum number— no significant relationships ; more than minimum— one significant relationship between GPA and AHPAT-RC. (This was one of the few relationships with the AHPAT-RC in the entire statistical analysis.) The results are shown in Table 13. The relationships between the subject matter or academic skills required in physiology and psychology and the specific AHPAT scores to which they were significantly related is questionable. The hypothesis was not supported by the data based on the results for the subgroups and weakly supported by the results in the total groups. 114 Table 11 Hypothesis Two: Correlation Between Professional Program Courses and AHPAT Scores When Separated by Marital Status (Single=155, Married=38) AHPAT Scores Group Criteria V Q B C RC T Single Anat -.144 .157 .108 .169 .171 .125 Married Anat .005 .628* .123 .496 .306 .408 Single Physiol -.116 .229 .191 .103 .235 .180 Married Physiol .234 .448 .491 .478 .392 .548* Single Biomech -.152 .222 .175 .233 .188 .183 Married Biomech -.068 .214 .100 .309 .170 .190 Single Neuro -.085 .118 .020 .070 .170 .080 Married Neuro -.161 .401 .028 .394 .188 .209 Single Psych .060 .138 .209 .200 .173 .225 Married Psych .273 .314 .472 .133 .362 .429 Single GPA —. 063 .238 .154 .153 .267 .210 Married GPA .078 .547* .278 .506* .434 .494 Note. 2 values for n-155=.285, n-38=.505 a Professional curriculum criteria * significant at Pi<. 0005 - inverse relationship 115 Table 12 Hypothesis Two: Correlation Between Professional Program Courses and AHPAT Scores When Separated by Major (Health Science=148, Non-Health=45) AHPAT Scores Group Criteria V Q ' B C RC T H.Sc. Anat -.128 .281 .183 .235 .279 .239 NH.Sc. Anat -.036 .091 -.125 .188 -.106 -.017 H.Sc. Physiol -.094 .347* .281 .222 .313* .300 NH.Sc. Physiol .188 .108 .105 .056 .111 .154 H.Sc. Biomech -.180 .270 .206 .205 .226 .221 NH.Sc. Biomech .063 . 066 .011 .199 .010 .082 H.Sc. Neuro -.093 .288 .080 .192 .258 .208 NH.Sc. Neuro —. 06 2 -.157 - .182 -.005 -.079 - .142 H.Sc. Psych .067 .175 .217 .179 .205 .245 NH.Sc. Psych .185 .154 .381 .241 .233 .327 H.Sc. GPA -.083 .348* .218 .263 .353* .314* NH.Sc. GPA .195 .139 .030 .086 .128 .154 Note. D value for n—148=. 291. n-45=. 507 ^^Professional curriculum criteria * significant at ^<.0005 - inverse relationship 116 Table 13 Hypothesis Two: Correlations Between Professional Program Courses and AHPAT Scores When Separated by Number of English Courses (Minimum=66, More Than Minimum=109) AHPAT Scores Group Criteria® V Q B C RC T Minimum Anat .065 .258 .041 .173 .191 .200 >Minimura^ Anat -.199 .259 .116 .217 .207 .166 Minimum Physiol .087 .276 .278 .173 .320 .310 >Minimura Physiol .075 .281 .151 .153 .260 .219 Minimum Biomech -.237 .076 .059 .175 .007 .005 >Minimum Biomech -.045 .306 .195 .252 .315 .300 Minimum Neuro — .180 .111 .004 .064 .183 .037 >Minimum Neuro -.049 .185 .014 .122 .186 .134 Minimum Psych .042 .147 .040 .060 .068 .102 >Minimum Psych .107 .128 .284 .185 .238 .280 Minimum GPA -.048 .225 .110 .100 .166 .143 >Minimum GPA -.002 .299 .168 .210 .366* .304 Note. n values for n-109=.323, n-66=.398 Professional curriculum criteria ^Minimum = more than minimum * significant at ë. <.0005 - inverse relationship 117 Hypothesis 3 Hypothesis three states that there is no relation ship between the scores on the AHPAT and the individual’s success in the clinical education areas of the physical therapy curriculum as measured by whether or not the indiv idual successfully completed each internship period in the internship experiences. The frequency distribution indi cated there were only 5 out of 194 individuals who did not successfully complete all of the internship experiences, therefore the hypothesis was not tested due to the small sample. Evaluation of the internship experience was on a pass fail basis. The internships were the final semester of a four semester plus one summer session process. Rela tional skills were a major part of success on internships until recently. Summer session of 1983 marked the first official use of a new clinical experience evaluation instrument which is based on competencies. This new instrument should lend itself to more objectivity in the clinical evaluation process. Hypothesis 4 Hypothesis 4 states that there is no relationship between the scores on the AHPAT and the individual’s grade point average at the completion of the entire professional program. Comparisons between AHPAT scores and GPA were 118 made using the following subgroups; sex, ethnic back ground, marital status, major and number of English courses taken. Out of these tests the following were significantly related: The GPA in the professional pro gram was related at the alpha .0005 level of significance with AHPAT-Q, AHPAT-RC, and AHPAT-T consistently in the total groups. When in subgroups, AHPAT-Q and AHPAT-G were significantly related for females and marrieds, APAT-Q and AHPAT-RC for Caucasians, AHPAT-Q, AHPAT-RC and AHPAT-T for Health Science major and AHPAT-RC when grouped by number of English courses taken. The results are included in Tables 8—13# Hypothesis 5 Hypothesis 5 states that there is no relationship between scores on the AHPAT and the individual’s scores on the California State Board of Medical Examiner’s Physical Therapist Licensure Examination. Part I of the licensure examination is the basic sciences. Part II is the clinical sciences. Part III is theory and procedure, and BD-TOT is an average of all three sections. Out of these tests, the following were significantly related. AHPAT-B correlated significantly with all three sections of the licensure examination plus BD-TOT. AHPAT-C score was significantly related to the score on Part I and the total score of the licensure 119 examination, and AHPAT-RC was significantly related to BD-TOT. Finally, the total score on the AHPAT was signif icantly correlated with all three sections of the licen sure examination plus BD-TOT. The results are shown in Table 14. All factors were not related. The highest correlations should obtain between the variables which seem to have the most in common; there fore, it is difficult to explain the correlation between the AHPAT-B and Part II on the licensure examination. One explanation might be considered, that is the ability of some individuals to take an examination and score very well simply by having excellent test-taking skills. Any examination of the AHPAT type represents a relatively short period of time spent on performance ver sus the time spent on performance in the approximately three years plus spent on prerequisite coursework or the scores in one or more courses in a specific subject area (Plapp & Psathas & Caputo, 1965). Hypothesis 6 Hypothesis 6 states that there is no relationship between the scores on the AHPAT and the preprofessional criteria when grouped by sex. Out of these tests the following were significantly related: males between pMath and AHPAT-V and AHPAT-Q; females, between pAlg and AHPAT- Q, and between pGPA and AHPAT-Q. The results are shown in 120 Table 14 Hypothesis Five: Stage 2— Combined Groups, Licensure Examination (n=130) AHPAT Scores Total Part I Part II Part III Verbal Ability .202 .170 .193 .171 Quantitative Ability .273 .224 .240 .232 Biology .453* .391* .366* .418* Chemistry .392* .370* .282 .304 Reading Comprehension .319* .253 .284 .300 AHPAT .474* .403* .399* .413* Note. Ë value for n-130-.305 *significant at £<.0005 121 Table 15. All other relationships were not significant. Differences in correlations between males and females were noted only in the area of quantitative abil ity. It should be noted that pMath (precalculus) is a higher level of mathematics than algebra or trigonometry. Forty-eight students out of 194 completed pMath (11 males, 37 females), 138 students completed pAlg (54 males, 84 females) and 27 completed pTrig (11 males, 16 females). Nineteen students completed more than the required amount of mathematics. All students are required to take pBio- stat (Biostatistics) which provides more mathematical skills prior to taking the AHPAT; therefore, correlations would be expected to be high in quantitative ability. When i tests were applied to determine if there was a difference in performance between males and females, males performed better in AHPAT-C and BD I, females per formed better in pChem 2, pPhysic 1, pGPA, Neuro, GPA, and AHPAT-RC. (See Table 16.) Hypothesis 7 Hypothesis 7 states that there is no relationship between the scores on the AHPAT and the age of the individ ual at the time the examination was taken. Age was com pared to scores on the AHPAT in the following subgroups. Sex, ethnic background, marital status, and major. Out of these tests, the following were significantly related: Table 15 Hypothesis Six: Correlations Between Preprofessional Criteria and AHPAT Scores When Grouped by Sex (Male=69, Female=127) 122 AHPAT Group Criteria^ V Q B RC Males pChem 1 -.209 .224 .065 .237 .005 .092 Females pChem 1 -.040 .169 .022 .231 .015 .100 Males pChem 2 -.284 .121 .131 .117 -.114 —.007 Females pChem 2 .024 .234 .178 .287 .080 .211 Males pMath (n=11) .434* .861* -.337 .059 -.159 .323 Females pMath (n=37) .008 .298 .084 .313 .211 .262 Males pAlg (n=54) .001 .135 .132 .137 —.046 .108 Females pAlg (n=84) .172 .414* .152 .181 .298 .323 Males pTrig (n=11) .442 .335 .041 .679 .510 .494 Females pTrig (n=l6) .207 -.249 .086 .321 .201 .202 Males pBio .001 .143 .026 .003 .011 .060 Females pBio .077 .049 .187 .082 -.005 .115 Males pAnat .105 .145 —.004 -.022 .128 .104 Females pAnat -.057 .240 .237 .283 .012 .189 Males pPhysiol .112 .138 .185 -.039 -.019 .114 Females pPhysiol .181 .221 .396 .225 .281 .375 Males pPhysics 1 -.039 .192 -.009 .142 —.126 .046 Females pPhysics 1 .011 .304 .246 .299 .094 .253 123 Table 15— continued AHPAT Group Criteria^ V Q B C RC T Males pPhysics 2 -.087 .233 .055 .297 .127 .185 Females pPhysics 2 .039 .281 .188 .259 .117 .237 Males pBiostat -.127 .299 - .068 .182 .134 .127 Females pBiostat .076 .332* .205 .341* .225 .317 Males pGPA (n=67) .089 .362 .180 .218 .113 .282 Females pGPA (n=126) .196 .423* .354 .365 .219 .431 Note, p values for n-127=.308; n--84=.331; n=l6=.742; n-•126=.309 > Prerequisite courses * significant at £ <.0005 - inverse relationship 124 Table 16 Hypothesis Six: Differences Between Male and Female Performance Male ( SD) Female ( SD) Ê Value' pChem 1 2.97(±.83) 3.10(*.78) .283 pChem 2 2.54(+.77) 2.97(+.79) .006» pMath 3.09(+.83) 3.03(1.87) .827 pAlg 3.09(±.83) 3.14(1.85) .732 pTrig 3.00(±.63) 2.87(1.72) .638 pBio 3.07(±.75) 3.28(1.69) .069 pAnat 3.45(±.61) 3.54(1.60) .299 pPhysiol 3.33( +.64) 3.35(1.67) .854 pPhysics 1 2.67(+.66) 3.05(1.78) .001» pPhysics 2 2.7 5-(± .75) 2.86(1.75) .323 pBiostat 3.03C +.76) 3.17(1.77) .242 pEng 2.13C +L.70) 1.79(11.34) .158 pGPA 3.17(+.32) 3.30(+.34) .008» Anat 3.52(+.61) 3.59(1.59) .458 Physiol 3.41(+.61) 3.35(1.57) .480 Biomech 3.36C+.64) 3.48(+.60) .218 Neuro 3.17(±.72) 3.43(1.65) .018» Psych 3.43C +.56) 3.50(+.50) .384 GPA 3.38C+.38) 3.50(1.35) .035» BD T 83.02(+2.63) 82.28(12.62) .133 BD I 83.69C+2.56) 82.49(13.40) .026* BD II 82.41(+3.11) 82.23(12.86) .756 BD III 82.36(+2.74 81.80(12.68) .263 pooled when variances were not significant; separated when variances were different or significant, ^significant at £<.05 125 AMPAT-V correlated with age at the alpha level of .0005 for the total group. The results are shown in Table 17. All other factors were not related. For subgroups, the following were significantly related; (1) age and AHPAT-V for females, (2) age and AHPAT-V for singles. The results are shown in Table 17. Hypothesis 8 Hypothesis 8 states that there is no relationship between the scores on the AHPAT and the preprofessional criteria when grouped by ethnic backgroùnds of the individ ual. Out of these tests, the following were significantly related at the alpha level of .0005; Caucasians between pMath and AHPAT-Q. There were no significant relation ships for others (combined Black, Mexican-Americans, Orientals, and others). The results are shown in Table 18. When t. tests were applied to the data to determine if there was a difference between Caucasian and others, Caucasians performed better in pChem 1, pBio and pGPA. (See Table 19.) Hypothesis 9 Hypothesis 9 states that there is no relationship between the scores on the AHPAT and preprofessional cri teria when grouped by the marital status of the individ ual. When the grouping was by marital status (single or 126 Table 17 Hypothesis Seven: Correlations Between Categories of the Time of Application and AHPAT Scores When Grouped by Age AHPAT Group Categories V Q B C RC T Age Male (n=67) .125 -.390 -.045 -.192 -.007 —. 166 Female (n=128) .330* -.211 .108 -.211 -.152 -.021 Caucasian (n=173) .248 -.240 .034 -.151 -.157 .064 Other (n=21) .393 .262 .170 -.262 .124 -.019 Married (n=38) .096 -.531 .156 -.416 -.193 -.224 Single (n=155) .315* -.190 .043 —.063 -.135 —.001 H.Sc.Major (n=148).197 —.296 .060 -.225 -.216 -.135 Non-H.Sc. Major (n=45) .355 -.097 .202 .010 .103 .165 Combined Groups (n=194) .254* -.245 .068 -.172 -.151 —.062 ^Separate and combined groups 4e values for n-194=.253; n-173=.270; n-155=.285; n-148=,291; n-126=.309; n-67=.395; n-45=.507; n-38=.505; n-21=.653. * significant at £ <.0005 - inverse relationship Table 18 Correlations Between Preprofessional Criteria and AHPAT Scores When Grouped by Ethnic Background 127 Groupé Criteria^ AHPAT V Q B C RC T Caucasian pChem 1 -.096 .185 .033 .229 .065 .115 Other pChem 1 -.225 .034 .002 .039 —.261 -.112 Caucasian pChem 2 -.081 .161 .155 .216 .031 .132 Other pChem 2 .103 .158 .090 —.020 .199 .130 Caucasian pMath (n=43) .092 .496 .007 .336 .182 .358 Other pMath (n=5) —#628 -.211 -.547 -.460 -.588 -.575 Caucasian pAlg (n=122) .112 .275 .129 .134 .146 .226 Other pAlg (n=l6) .306 .591 .262 .445 .503 .565 Caucasian pTrig (n=23) .173 -.064 .187 .511 .283 .346 Other pTrig (n=4) .603 -.044 -.543 -.043 —.336 -.120 Caucasian pBiostat .067 .083 .122 .051 .058 .121 Other pBiostat -.283 —. 186 .105 -.265 —.236 -.224 Caucasian pAnat .027 .204 .183 .191 .084 .195 Other pAnat -.252 .094 —.069 .025 -.069 -.073 Caucasian pPhysiol .134 .194 .309 .112 .187 .275 Other pPhysiol .339 .286 .497 .336 .166 .403 Caucasian pPhysics 1 .009 .236 .164 .228 .078 .198 Other pPhysics 1 .113 .070 .058 .027 -.007 .065 128 Table 18— continued AHPAT Groupé Criteria^ V Q B C RC T Caucasian pPhysics 2 -.005 .246 .122 .280 .137 .223 Other pPhysics 2 .122 .307 .284 .083 .081 .214 Caucasian pBiostat .029 .299 .082 .239 .207 .243 Other pBiostat -.075 .341 .318 .479 .197 .314 Caucasian pGPA (n=171) .162 .372 .280 .274 .227 .385 Other pGPA (n=22) .091 .278 .318 ,248 .116 .260 .Caucasian n=172; Other n=21, unless specified Preprofessional criteria £ values for n-172=.271; n-122=.312; n-171=.272; n-43=.486; n-23=.730; n-4=>.991; n-16=.742 129 Table 19 Hypothesis Eight: Differences in Performance Between Caucasians and Others Caucasians Others £Value pChem 1 pChem 2 pMath 3.09( 2.87( 3.07( .79) .80) .86) 2.71 ( 2.71( 2.80C .78) .72) .84) .044» .354 .526 pAlg pTrig pBio 3.09( 3.00( 3.26( .87) .67) .70) 3.31 ( 2.50C 2.76( .60) .58) .62) .219 .186 .002» pAnat pPhysiol pPhysics 1 3.52( 3.35C 2.94( .60) .66) .75) 3.43C 3.24C 2.76( .68) .63) .83) .558 .438 .368 pPhysics 2 pBiostat pEng 2.82( 3.13( 1.92( .76) .77) 1.48) 2.86( 3.05C 1.86C .73) .80) 1.46) .805 .671 .856 pGPA Anat Physiol 3.28( 3.57( 3.38( .33) .58) .58) 3.10C 3.48C 3.33( .32) .75) .58) .025» .554 .721 Biomech Neuro Psych 3.45C 3.57( 3.48( .63) .69) .52) 3.29C 3.14C 3.48C .46) .65) .51 ) .144 .150 .978 GPA AHPAT-V AHPAT-Q 3.48C 52.87( 64.66( .36) 25.75) 25.66) 3.32C 41.24( 60.8K .36) 27.36) 24.08) .069 .076 .493 AHPAT-B AHPAT-C AHPAT-RC AHPAT-T 67.02( 65.04C 59.33C 309.02( 23.19) 22.55) 25.07) 81.43) 65.33( 59.52( 60.29( 287.05( 25.70) 27.92) 28.67) 105.84) .776 .392 .885 .368 pooled when variances were not significant; separated when variances were different or significant. *significant at £<.05 130 married) the only significant relationships for singles were between pMath and AHPAT-Q and between pPhysiol and AHPAT. For marrieds, there were significant relationships between pPhysiol and AHPAT-B and AHPAT-T and between pBio stat and AHPAT-Q and AHPAT-C. The remaining results are shown in Table 20. All other factors were not related. When t tests were applied to the data to determine if there was a difference between singles and marrieds, none was found. See Table 21. Hypothesis 10 Hypothesis 10 states that there is no relationship between scores on the AHPAT and preprofessional criteria when grouped by the preprofessional major of the individ ual. When the majors were combined there were significant relationships at the alpha level of .0005: between pPhys iol and AHPAT-B and AHPAT-T; between pPhysics 2 and AHPAT- C, between pBiostat and AHPAT-Q, AHPAT-C, and AHPAT-T, and between pGPA and AHPAT-Q, AHPAT-B, AHPAT-C, and AHPAT-T. The remaining results are shown in Table 22. Out of these tests the above mentioned were significantly related, all other factors were not related. When the majors were separated and compared to AHPAT separate and total scores, the following were signif icantly related: for Health Service major, pPhysiol with AHPAT-B, pBiostat with AHPAT-Q, AHPAT-C, AHPAT-T, and pGPA 131 Table 20 Correlations Between Preprofessional Courses and AHPAT Scores When Grouped by Marital Status AHPAT Groups Criteria- V Q B C RC T Single pChem 1 —«098 .168 .033 .200 .055 .099 Married pChem 1 -.093 .200 .040 .211 -.097 .050 Single pChem 2 -.059 .162 .123 .182 .100 .140 Married pChem 2 -.009 .178 .259 .248 -.146 .128 Single pMath (n=41) .085 * .434 .044 .273 .223 .325 Married pMath (n=7) .073 .416 -.177 .276 -.552 —.016 Single pAlg (n=105) .197 .298 .145 .076 .218 .268 Married pAlg (n=32) -.066 .303 .149 .294 .038 .179 Single pTrig (n=23) .267 .059 .157 .411 .259 .335 Married pTrig (n=3) -.052 -.974 -.115 .500 .866 -.229 Single pBio .045 .105 .110 .160 -.003 .094 Married pBio .164 -.117 .238 .021 .150 .139 Single pAnat -.008 .153 .200 .110 .068 .164 Married pAnat .036 .354 —.003 .252 .056 .170 Single pPhysiol .128 .119 .285 .214 .127 .230 Married pPhysiol .326 .492 .507 .301 .365 .537 Single pPhysics 1 -.054 .227 .118 .258 .088 .165 Married pPhysics 1 .205 .171 .307 .122 -.014 .221 Table 20— continued 132 AHPAT Groupé Criteria^ V Q B C RC T Single pPhysics 2 -.076 .280 .159 .268 .125 .212 Married pPhysics 2 .218 .129 .073 .170 .148 .218 Single pBiostat —.026 .251 .128 .188 .176 .201 Married pBiostat .170 .535 .046 .538 .308 .424 Single pGPA .136 .343 .269 .258 .188 .343 Married pGPA .340 .480 .363 .352 .281 .503 ,Single n=155, married n=38, except were specified Preprofessional criteria p values for n-155=.285; n-105=.326; n-41=.546; n-32=.556; “ n-23=.710; n-7=.951; n-3=>.991 Table 21 133 Hypothesis Nine : Differences in Performance Between Singles and Marrieds Singles Marrieds aValue* pChem 1 3.06 +.80) 3.08 +.78) .886 pChera 2 2.87 +.79) 2.79 ±.84) .591 pMath 3.07 +.88) 2.86 ±.69) .481 pAlg 3.17 +.81) 3.00 ±.92) .347 pTrig 2.91 +.67) 3.33 ±.58) .338 pBio 3.20 ±.73) 3.21 ±.62) .928 pAnat 3.51 +.56) 3.50 ±.76) .942 pPhysiol 3.36 ±.63) 3.25 ±.75) .297 pPhysics 1 2.92 +.75) 2.95 ±.80) .828 pPhysics 2 2.81 +.73) 2.87 ±.84) .711 pBiostat 3.14 +.76) 3.05 ±.80) .538 pEng 1.91 +1 .47) 1.92 ±1.55) .967 pGPA 3.27 + .34) 3.21 ±.34) .403 Anat 3.59 + .60) 3.47 ±.60) .303 Physiol 3.37 + .58) 3.42 ±.60) .623 Biomech 3.43 + .64) 3.47 ±.56) .665 Neuro 3.37 +.69) 3.26 ±.69) .411 Psych 3.49 +.53) 3.45 ±.50) .644 GPA 3.45 +.37) 3.50 ±.34) .518 BD T 82.64 +2.63) 82.29 ±2.70) .534 BD I 83.08 ±3.33) 82.41 ±2.64) .257 BD II 82.28 ±2.86) 82.53 ±3.10) .689 BD III 82.08 ±2.67) 81.73 ±2.90) .566 AHPAT-V 50.24 ±25.0) 57.92 ±29.8) .148 AHPAT-Q 63.93 ±23.2) 65.90 ±21.3) .618 AHPAT-B 66.79 ±23.9) 66.95 ±21.8) .970 AHPAT-C 65.13 ±22.9) 62.95 ±23.4) .607 AHPAT-RC 59.87 ±25.1) 57.76 ±27.1) .665 AHPAT-T 306.29 ±83.2) 310.50 ±89.8) .794 ‘ Spooled when variances were not significant Table 22 Hypothesis Ten: Correlations Between Preprofess AHPAT Scores When Combine (n=194) ional Course and d by Major 134 Criteria^ V Q B AHPAT C RC T pChem 1 -.087 .178 .033 .213 .025 .096 pChem 2 -.053 .162 . 149 .192 .048 .135 pMath (n=48) .077 .430 .019 .263 .122 .275 pAlg (n=138) .118 .297 .143 .160 .176 .252 pTrig (n=27) .298 .028 .080 .437 .286 .324 pBio .062 .066 .122 .031 .023 .095 pAnat —. 001 .193 .152 .170 .063 .159 Physiol .161 .188 .328 .142 .183 .292 pPhysics 1 .015 .219 .152 .204 .066 .183 pPhysics 2 .006 .251 .140 .253 .130 .218 pBiostat .021 .305 . 1 1 1 .272 .205 .254 pGPA .174 .365 .284 .278 .210 .375 fPreprofessional criteria p values for n-194=.253, n—138=. 300, n-48=.503 9 n-27=.596 135 with AHPAT-Q and AHPAT-T. The remaining results are shown in Table 23. When t tests were applied to see if there was a difference between Health Service Major and Non-Health Service major, there was no significant difference. Additional Questions to be Answered Only two out of seven bachelors degree and/or certificate programs in California require the AHPAT: Loraa Linda University at Loma Linda, California, and Cali fornia State University Northridge, Northridge, Califor nia. In each of the schools the AHPAT is used as part of the selection criteria among several others. At CSUN the maximum point value assigned to the AHPAT is five, with a range of zero for those who do not complete the examina tion before the application deadline and the selection process to five for those who score in the highest quin- tile of applicants at all testing centers taking the exam ination on each specific examination date. The AHPAT rank of 5 points maximum is part of over 200 possible points in the point system utilized for selection at CSUN* Mean scores on the AHPAT for CSUN students are as follows; (1) AHPAT-V-51.6, (2) AHPAT-Q-64.2, (3) AHPAT-B, 66.8, (4) AHPAT-C-64.4, (5) AHPAT-RC-59.4, and (6) AHPAT- T-306.6. Table 23 Correlations Between Preprofessional Course and AHPAT Scores When Separated by Major (n=148)® 136 Group Criteria^ AHPAT V Q B C RC T H.So. pChera 1 (n=147) .066 .219 .072 .247 .068 .150 NH.Sc. pChem 1 -.105 .078 -.093 .125 -.080 -.034 H.Sc. pChem 2 —.030 .183 .192 .206 .089 .175 NH.Sc. pChem 2 -.071 .103 -.009 .154 —. 101 .026 H.Sc. pMath (n=34) .092 .484 .130 .370 .147 .363 NH.Sc. pMath (n=14) .039 .224 -.518 —.083 .002 —.062 H.Sc. pAlg (n=106) .125 .297 .112 .169 .227 .262 NH.Sc. pAlg (n=32) .176 .288 .203 .178 .063 .255 H.Sc. pTrig (n=23) .345 .146 .095 .516 .426 .484 NH.Sc. pTrig (n=4) .115 -.195 .943 .676 —.016 .427 H.Sc. pBio .049 .056 .139 .069 .057 .115 NH.Sc. pBio .165 .106 .048 -.088 —.089 .051 H.Sc. pAnat -.015 .210 .189 .193 .089 .188 NH.Sc. pAnat .078 .146 .034 .104 -.017 .038 H.Sc. pPhysiol .131 .194 .291 .118 .169 .267 NH.Sc. pPhysiol .378 .185 .440 .242 .270 .418 H.Sc. pPhysics 1 -.024 .236 .122 .189 .065 .166 NH.Sc. pPhysics 1 .254 .190 .210 .271 .098 .278 Table 23— continued 137 Group Criteria^ AHPAT V Q B C RC T H.Sc. pPhysics 2 —.009 .272 .179 .291 .142 .241 NH.Sc. pPhysics 2 .042 .180 .022 .191 .078 .143 H.Sc. pBiostat .038 .372 .224 .318 .273 .351 NH.Sc. pBiostat -.141 .084 -.216 .116 -.078 -.077 H.Sc. pGPA .188 .362 .280 .282 .251 .394 NH.Sc. pGPA .247 .406 .272 .281 .076 .357 .Health Science major n=148; non-Health Science major n=46 .Preprofessional criteria p values for n-148=.291, n-106=.325, n-46=.471, n-32=,556, n-34=.537, n-23=.710, n-14=.780, n-4=>.991 138 Additional Data Analysis: Non-Hypothesized Findings The data collected in the research yielded a wealth of additional facts from which to gain interesting information that could be useful in the selection process. Simple regressions were applied to the data in Stage 1 (preprofessional criteria) as Stage 2 (professional cri teria) to determine if there were any relationships between the prerequisite grades, GPA or age and profes sional curriculum grades, GPA or Licensure Examination scores in total groups. For subgrouping by sex, there were no significant correlations between prerequisite grades, pGPA, or age and the professional program grades, GPA for males. For females there were numerous signifi cant correlations (see Table 24). Prerequisite anatomy and pGPA were consistently significantly related for females, followed by pBiostat, to professional criteria. Out of these tests, the following were significantly related for combined sexes: pAnat with BD-T, BD-II, pPhysics with BD-I, and pGPA with BD-T, and BD-II. The remaining results are shown in Table 24. For subgrouping by ethnic background, all were designated as "Other" yersus Caucasians since there was only a small number of Blacks, Mexican-Americans, and Orientals. There were no significant correlations for the 139 M CM m OO on t — t — o CM H VO in CM in in CM VO in OO M O o CM CM CM o • • • « • • • t- in OO on o in o VO in M =r CT) VO OO z j - in in OO on zr T5 M CM CM on CM CM x-~ O • oë m m C T » St on on t — H o o > t> - CM OO O ZJ- CM in S o O CM on m T5 1 1 C D C •H E - * CM in in on C T v on zr o Q C T » =r =T on C T » VO c n VO in C T v OQ O CM O r - CM on CM CM O o t — OO t — in o C T » zr C O H c n =t CM vO in in VO CM OO OO X M CM m in CM in on on on CM on <D • CO >» ( 0JH m in CO OO VO o on T~ in O ) ^ s T3 CM VO VO VO zr t — OO k . o m o in on zr CM CM T3 pH ( 0 « VO OO on in CM O L. A L T > t — o o z j - o on cn zr on L. L. CQ c n CM CM in Z f zr CM CM on O 3 • C \ J i ^ ( D CM rH < CO on OO on zr VO t — r~i ( 0 C L , =t on in t — V — in CM in CJ\ ( D C O CM on on on on on on zr ( 0 bO O • E - * C O *H 4 - > C O o 00 C O L. L T > on t — o on on T — CM ,0) 3 L T > C T » on OO o C T v t- VO CM • C i H C D CM =r Z f CM CM CM zr C O O Z • • * > L. cu m 1 —1 S =r CM VO in CM OO VO CM zr o t — VO CM t — OO t — CM 0> •H CM CM O CM on on CM on on bO CQ C O •H CO C O OO OO on S- in o c o v VO i n on >» o OO VO i n VO on X3 O on CM on on on on on CM on Ou • • • • • • • • • • 4^ o o C T v C T v z j - OO C T v t — on O ) C O OO VO in in Zf on on CM o c CM CM in CM on CM in • • • • • • • ' /-X C D in OO 4-) VO OO •H II II II CM C O c c C •H 1 —1 1 — I t o t o +4 3 3 CM o o o (0 O ' O •rH •r H •r H 4-> C D •H E E X! bO 4-) t o ( 0 C O U U C D C D -P bO •H C O O C D L. x : X: 0 3 1 -4 L. C X X X •r H 3 o C _ ) E - « Ou IX IX 0_ O Ou Ou Ou Ou Ou Ou O , Ou o . Ou 1 4 0 1-4 cr» m 1-4 t — n 1-4 CM o 1 o OO z r M t — t ' •al 1-4 e n T- II o c du 1 t u CM CT» 1-4 O C7V « a 2 : m «— S • o 1 (D T 3 C O <D •iH E l X C X Q OO n t — *H B OQ m t — • -G O Il B O 1 n o t r - O o n t - 1 1-4 o OO C bO 1-4 ir> % — • H SU 1 o n m (X M 1-4 CM CM X •O OO V— "O * #. L o n CM <D II m co • H X — O o 1 4-1 CM V— CQ • H 1 II E-« o OO O C C Q OO CM <D OQ = r CM O , .* t u C O t ' 1 e n oü c m «a; z r z r <D • 2 : Ou LO o X II O LO z r 5 OO T3 X <D X 1 c O e u c *H S^ r - O <D X 3 m o o s » .* CM > » 'T- CM i H X II X z r CM c CM c m < x o o n m o • CM t u t— co il CJv •a X t - o n c m * « a II < Ü m VO 'T - o o II 3 c z r z r X V - OO <D S t u C c 2 : co o X <D SU e x <D o , Il 2 : <0 SU 3 O M t u o J 2 : eu E-« < x O CU « 4 (0 X 0 * 0 141 other subgroup between prerequisite grades, pGPA, or age and professional program grades or GPA, Out of these tests for the Caucasian subgroup, the following were sig nificantly related: pChem 1 with Biomech, pChem 2 with Anat, Physiol, Biomech, and GPA, pAnat with all the pro fessional courses, plus BD II, pPhysiol with BD-T, pPhysics 1 with GPA, pPhysics 2 with Biomech and GPA, pBiostat with Anat, Neuro and GPA, pGPA with all pro fessional courses, professional course plus BD-T and BD II, and age with Anat, Biomech, Neuro and GPA. Prerequ isite anatomy and pGPA were again, as in the subgrouping by sex, significantly related for Caucasians to profes sional criteria. The other factors were not related, the results are shown on Table 25. When the ethnic backgrounds were combined the following were significantly related: pAnat with BDT, BD I and BD II, pPhysio with BD I and pGPA with BDT and BD II. The remaining factors are not significantly related (see Table 25). The following were significantly related for the Health Science group at the alpha .0005 level: pChem 2 with Anat and GPA, pAlg with Neuro, pAnat with all other factors, pPhysiol with BD I and pGPA with BD-I. Again, pAnat and pGPA are consistently significantly related to 142 M VO CO p 1 - 4 OJ =T VO M O OJ OJ OJ bO 1 C m m tn VO VO C - - M •p4 < y \ to ■ = r =r C - - •3 M C O O OJ m O C O Du %4 E o =r OJ in t > - M C O in o t > - =r OJ S 43 o m C O 73 73 1 1 C D C C O •H E- =r < 3 \ o\ m T3 Q o OJ O oo CO C C Q OJ o ( \ j m 3 • * O O 1 bO CO in tn OJ VO VO VO M O C T \ L T > t — C - - O 1 - 4 O OJ m 0 3 • • CQ O C OM m o =r oo oo m r 4 S 73 m LO o o- m C 3 L , o o m Æ i - l C O • • 4 - > 3 O 1 ^ i 1 1 x 3 O oq c- CO • r - l E - * c — » — r j - « — OJ p >» t . Q LO t — o % — OJ C X 3 ^ t - OQ O II OJ II o OJ OJ m 3 • c • a • C \ J T3 O I < D O.f-4 < OO VO to ro r - 4 3 ( 0 D u m o t > - ■ = r OO tn OJ xa O C O OJ m OJ o =r 0 3 L . O E - I 9 3 O 1 C O U ( \ j o C 3 \ oo oo OJ ( \ j m 3 m C-- CO X C-- VO ( # - * C D OJ ( \ j O OJ X o m C D O S bO L . C O Du C D 4^ E =r OJ CO o\ oo o m 00 O E-- CO tn m cn VO =r • f 4 OJ OJ o o o m OQ C O > •H C O C-- b- C 3 \ OJ oo o\ >» in C-- o\ cr* m X! OJ OJ m C D CU bO C O 4^ 43 LO =r C 3 \ vO m CM CO C O CO a\ tn o oo o C OJ o OJ o =r c • • • • • • C D OJ X —\ 4^ ro OJ ro •H =r OJ C O II II II •H r-4 c c c 3 3 OJ a O C D •p4 E E X: bO 43 s u 0 3 0 ) 43 bO •H o C Q E u 3 S 6 s A •r4 o q 5 D U O cx cx cx CX CX CX CX 143 M m oo on M OJ =t on VO oo M OJ o OJ O • • • • OJ on LO LO on M i > - on LO LO OJ 73 M OJ OJ m O Du 1 OJ oo on LO oo c — M oo LO o\ VO o s m o O OJ OJ C — 73 1 on ( D C I I •H E - « OO o o c r v c — 43 G c r \ OJ on VO VO e oq OJ OJ on o 1 o 1 c VO o\ VO LO o o\ M o LO on LO on on LO M OJ OJ on o • • on 1 ( ûM LO OJ c r \ O OJ o8 73 oo o\ oo S - ro o o OJ ( Q 1 iS 1 c E- m OJ o- i > - o G xr OJ xr o on o OJ CQ OJ OJ on o • on 1 I I < * ! VO =T o\ c — OJ OU oo on LO OJ O OJ OJ on on zf on 1 c O m o o\ VO o\ 3 OV zf OJ =t o\ c r v ( U OJ OJ on on OJ on o z • c — on < D I l I I e oo l > - LO o ov c r v c z a - o in lO Zf on c r v OJ •H OJ on OJ on OJ OQ >> 1 iH C •H C m GO o on o • » >» LO OJ Zf VO LO O o 43 OJ OJ OJ OJ on OJ e o c — on OU G OJ c — o • • •H I I I I 73 43 OJ OV oo VO oo o 43 on • * 0 ) 0 3 OV LO LO LO I r OJ c a c — on 3 C OJ OJ on on on rH T — CM a • S 3 0 > 1 1 •H G c c 4 - ï u C o G O < D O o VO o 43 <M on 1 •p4 OJ c LO 1 M < Q ( 0 • m •H 1 —1 1 —1 ( 0 W 43 •H 0 > I l OJ 3 3 G O G 0 3 w 3 on CT O •H • r 4 • r 4 43 C a rH z j - < D m •H W C O w ( 0 O ( 0 1 1 —1 L . >» >, >> O •< 3 > c x> < L > L . 43 43 Si •H Ou ( D C Q 1 ( 0 3 Ou Ou ou oq O bO o ûj. E - t Ou O CL CL CL Ou CL « = C ( Q 43 144 professional criteria. The remaining results are shown in Table 26. There were only two significant relationships for non-health science majors (n=46); pPhysics 2 and Biomech (.479 level, ^ value= .455), pGPA and GPA (.529 level, p. valuer .455). When the majors were combined the following were significantly related: pPhysiol with BD-T and BD I and pGPA and BD-T, BD I, and BD II. The remain ing results are shown in Table 26. When grouped by marital status, there were numer ous significant relationships for singles (Table 27); however, only two for the marrieds: pPhysiol with Physiol in the professional program (.674 level, ^ valuer.306) and with GPA (.555 level, ^ valuerg.06). When combined there were significant relationships between pAnat and BD-T, I, II and pGPA and BD-T and BD II (Table 27). Pre requisite anatomy and pGPA were again fairly consistently significantly related to professional criteria. The final group was by the number of English courses taken, comparing prerequisite courses to profes sional courses and either a minimum of one or more than one English courses. Significant relationships were evident between prerequisite courses and professional curriculum grades and GPA as shown in Table 28. 145 (0 G O C O S >v g X iH 73 3 < D O CL•H 3 G O G vo G 3 C M O O 0 ) i rH 1 —1 C M C Q X S c o ( D O t - * b O• r H C Q Ï O 43 C O OQ G C m O C O G > C L . %- Q ) bO C Q 43 tn M M M 1 lu 73 M O du oü M s: 73 0) C •H f r * 43 G S OQ O c _ > M M (0 M 73 G I § § 2 I I m I I s w § r 4 1 - 4 §•§ < D *H G G ( D G G 3 a , o CM on OO c r v on VO LO CM m O O CM • # • • • C — L T l CO on zr on vO OO zr CM CM • • • • on on zr CM o c r v t — CM OO o O CM i I CM m LO on c r v c r v zr O on c r v o CM O CM • • • • • c r v c r v L T » c r v o o zr on LO zr CM CM CM CM CM • • r é on on C 3 V OO oo VO m o o on • • i • • CTv on VO m t — on vo LO o CM O CM on • • • • • c r v on on on m on CM on on o • • • r t — o CM on o CM on vo CM CM CM o on o • • • 1 t — o o CM zr zr o o o o on m vo CM CM o CM o • • 1 • • t — t — vo L T > t — on I T » zr O o CM CM CM • • • • tn o OV O CM t — o CM on O CM o • • • • vo zr O on on CM I I I I I I C c c s - / CM B B 43 b O 93 C D 43 b O •H o ê A C 3 , CL o < CL CL vo s en % St vo OO Rr zr v o s , CM zr m o o m zr o crv o m m vo o n zr m m % on oo on zr on zr CM z r on ov on on on on on zr zr % ï 146 H t- O CM c r * 1 - 4 CM vo L T t OO c — t - i CM O CM in L T » O VO LO O o M L T » OO on O on 73 1 - 4 CM CM CM on O • I I Du o on oü zr on on L T l CM 1 - 4 zr CM c — zr o 1 S on on c • 73 0 > o C on • r 4E - t zr O C- t — O Q c r v VO L T » L T » oo em CM CM on I I o ' • on o 73 CM C D 1 CM O zr t- CM •H C M o on t- t- M CM CM CM on • r H • » O t — Q ) on Ou L T » G O1 - 4 c r v o CM L T » zr G O • 73 c r v c r v VO O c r v I I G CM o CM CM t o zr C O G O on o 0 > 1 m r - 4 C e - t c r v t- on on oo c Q zr CM zr zr oo 3 DQ CM CM CM on t- • oo LO zr zr « a j CM VO CM on CM I l I I (G C- L T » on vo L T » c LO O CM CM on on zr * * » 1 >> C O r - 4 G O on zr CM C 3 O oo on on O LO Q ) CM CM CM on on CM Z w « — on G O • ( D G r- I l S on i > - c r v r- •O I I LO o oo c r v CM c— on c g C o •H on CM on s V- OQ • • » 1 G >> C • p4 G r - 4 C O CM CM on vo C C >> L T l CM t — oo G G r- 43 CM CM CM CM on • r H c r v Ou • G G O o CM 00 G on • G I l 73 43 OV zr m c r v 43 *0 I I oo < D C Q O on t — L T t c r v 43 m c zr 3 C CM CM CM on on t - 4 s ^ «7 C C O W 1 •H G G G c 43 □3 G G , C c -o G O 0 ) G G c g O O 43 G r - l s ( t u 1 •H CM <M G 1 G O CO 73 W vo •H r - 4 r - 4 C O G O 43 C O G G CM 3 3 O o o C O 43 43 C 3 C T O •H •r4 •r4 43 r - 4 43 r H r - 4 < D Q ) •H C O w G O G O 3 r - 4 a G 1 —1 G G >v >v >» O « a ! G O G S > O Q ) G 43 43 43 •H ou G G G ^ , C O G 3 eu eu Ou CQ C3 « ne 0 CL I E-t Ou O CL CL CL CL a G 43 G 73 147 M G T \ on oo c — OV o zr 1 4 3 M 1 4 3 1 4 3 OJ zr vo yp OJ VO M O o OJ OJ OJ • • • • • • • • • Ov OJ oo vO VO ov 1 4 3 1 4 3 M OJ =r V 4 3 vo 1 4 3 vo b- VO 73 M OJ O OJ on OJ O • •, lu oÿ OJ zr zr OJ on vo on VO M ov E- b- b- 1 4 3 S O o on on 73 1 1 G C •H E - t oo 1 4 3 c r v OJ 1 4 3 G •Q Q OO zr zr OJ oo C 3 V b- O 1 4 3 W CQ o OJ o OJ on on 3 43 o G 43 c r v vo vo OJ oo OJ CM 1 4 3 00 M OJ zr on ou on zr M on OJ OJ O OJ on CM CM 1 —l • • G 1 43 •H G M 1 4 3 on t — on 1 4 3 o T— zr G g 73 on on OJ 1 4 3 b— b— b— G 3 G o OJ o OJ on CM S r4 G * 3 O 1 ^ 1 1 >* O OQ 43 *H E - t on ov T- C T V on on OO 1 4 3 zr G G c r v c r v T- C T V b- b- vo VO O zr 73 G OQ O l i on H o OJ OJ zr CM CM c — G 3 • c • c • OJ CXO G O i - l « a j in O 1 4 3 1 4 3 OJ zr OJ c r v 1 —1 G G eu on 1 4 3 C T V OJ b- in zr 43 Q C o OJ on OJ o zr on G 1 O E - * 1 *H OJ W p G G OJ on zr zr OJ zr 1 4 3 c r v zr G G 3 b- zr OJ vo zr b- vo oo b- bû ( m G OJ o o on CM G O S • • • 43 G 1 00 Dl, G S OJ OJ oo b- zr oo OJ CO oo O vo zr VO vO zr on on t — b- G •H OJ on o o o on CM > OQ • • • • • • • • r 4 G on ov oo OJ vo b- vO G >. ov oo VO C 3 V VO on 1 4 3 on C T V bû 43 o OJ OJ O O OJ CM G eu * * to 43 in OJ OV on zr 1 4 3 c r v c r v G b- 1 4 3 O OJ zr on on c r v C on O O zr CM < * ! • • • • • • • • oo G OJ 43 b- •H I I b- G c I I b- I I •H rH C I I c r - 4 G 3 3 OJ C O O O* Ü N - r •ri •rH G •H B B 43 bO 43 G G G G G G 43 bO •H o G >» >> G G 43 43 G •H G •H C 43 43 G 3 O O Z E- CQ eu Ou Dl,O ex ex ex ex ex ex ex ex ex 148 M OV OV OO CM 1 —1 zr b- m 1 —l ou O CM o • • • zr CO zr on b- M ou CO b- zr 73 M CM on oo o| • • tu 1 i l =8 b- CM b- M vo on o O v 1 S on c 73 1 G 143 C CM •H E-t zr b- CM OO b - X Q z r z r oo z r B CQ CM on T — I I O b- o 1 vo on r - OV C M C 3 V VO 1 — 1 m zr O « 1 — 1 1 1 — 1 143 G M z r oo vo b- 73 CM I I G CM on b- G b- S 1 1 C E - t CM vo ov Q L T V OO CQ on zr o VO CM 1 on < a j in oo oo I I Ou on on b - C 3 V ov O m m zr m cbt 1 1 c O G ov vo o o 3 CO ov on L T V G CM CM on on o 2 on 1 G • I l E CM zr LT» vo >> oo O CO vo zr z r r-H C M •H on CM on on C ^ CQ O 1 r £3 •rH G G vo LO oo C - >» m o o b- O vO 43 CM CM on zr •H 73 O eu • 43 G on 1 G r H • r-H (M I I 73 43 CO on zr on G r H OV G G vO zr ov oo G O CM D C CM on on CM G G T— C « 3 g O CL 1 •H r O G C 43 C CX G G O G 3 G p O 43 O G Lh 1 •rH CM G 1 -4 1 G O C G b- • i H•H G 43 3 G C\1 3 3 O G G 3 O * O •rH 43 1 —1 OV 1 —1 G G •H G G bÛCM G 1—1 G G >» o « = * ! G C t- > O G G 43 •r H Ou bû •r H II 1 G G 3 Ou CQ C3 G e n C CX E - t Ou O Ou CL Ou Ou G o o Table 28 Preprofessional Courses vs. Professional Curriculum (Stage 1 vs. Stage 2) When Grouped by Number of English Courses Taken 149 Courses a Anat Physiol Biomech Neuro GPA pChem 1 min .247 .244 .499* .303 .313 pChem 1 more .091 .045 .165 .136 .124 pChem 2 min .291 .350 .348 .350 .379 pChem 2 more .268 .180 .275 .154 .238 pMath (25) min N/S .267 —.061 N/S .118 pMath (18) more .081 .266 .348 .266 .329 pAlg (38) min .099 .149 .215 .209 .162 pAlg (85) more .204 .208 .147 .287 .214 pTrig (5) min .612 .612 .167 -.167 .547 pTrig (22) more N/S .275 .216 .109 .085 pBio rain .178 .236 .099 .127 .209 pBio more .061 .060 .033 .041 .083 pAnat min .247 .403* .307 .296 .450* pAnat more .492* .315 .332* .392* .438* pPhysiol min .278 .419* .153 .344 .329 pPhysiol more .200 .166 .201 .139 .271 pPhysics 1 min .443* .385 .391 .448* .475* pPhysics 1 more .108 .101 .082 .048 .106 pPhysics 2 min .266 .408* .422* .273 .370 pPhysics 2 more .207 .117 .337* .205 .285 pBiostat min .264 .203 .165 .288 .296 pBiostat more .415* .319 .268 .336* .372* pGPA min .379 .505* .344 .376 .545* pGPA more .361* .255 .326* .311 .396* Note, s . values for 109=.323, 85=.331, 66=.398, 38=.505, 25=.618, 22=.652, 18=.708, 5=.991 min=66, more=109, unless specified ^significant at ^<.0005 -inverse relationship 150 The final statistical method applied to the data was the student t tests (BMDP-3D) to determine if there were differences in student performance between means for the following: (1) GPA of < 3.09 (low) vs GPA of > 3.09 (high) (See Table 29), (2) female vs male (See Table 16), (3) Caucasian vs Others (See Table 19), and (4) Caucasian vs. Oriental. The results of the t test for differences between males vs. females indicated six areas where females were significantly different from males at the ^.05 level of significance. These areas were (1) pChem 2 .006, (2) pPhysics 1 .001, (3) pGPA .008, (4) Neuro .018, (5) GPA .035, and (6) AHPAT-RC .018. Males were significantly different from females in two areas; (1) AHPAT-C .012 and (2) BD I .026. Differences between Caucasians and others indi cated that Caucasians were significantly different from others at £ .05 level of significance in three areas: (1) pChem 1 .044, (2) pBio .002, and (3) pGPA .025. When Caucasians were compared to Orientals, there were no sig nificant differences in any areas. 151 Table 29 Differences Between Students* Performance with Low (<3.09) and High 03.09) Prerequisite GPAs Course Prerequisite GPA LowT" (±SD) High "X (±SD) EValue® Anatomy 3.34(1.70) 3.69(1 .51) .001 Physiology 3.16(1.57) 3.48(1.56) .0001 Biomechanics 3.25(1.57) 3.52(1.63) .003 Neurology 3.14(1.67) 3.40(1.66) .004 Psychology 3.33(±.51) 3.55(1.51) .01 GPA 3.27(1.32) 3.56(1.35) .0001 Internship 1.03(1.17) 1.02(1.15) NS*) Board T 81.65(12.74) 83.24(12.31) (.001 ) I 81.82(13.52) 83.74(12.64) (.001) II 81.36(13.09) 83.02(12.60) (.002) III 81.48(12.79) 82.42(12.57) (.057) Pooled when variances were not significant; separated when they were different or significant. *NS=not significant 152 CHAPTER V SUMMARY, CONCLUSIONS, RECOMMENDATIONS Summary and Conclusions As health care grows increasingly more complex and, in the process becomes more fragmented and special ized, life for the health care professional becomes more intellectually and personally demanding. Health care professionals include many separate professions each having its own area of expertise and scope of practice. All of them, however, also have much in common. One com monality is the problem of selection and admission of candidates into the professional education program. In fact, most of the professional education programs are impacted; thus the problem of selection of those best suited for admission into the profession becomes an impor tant issue to the educators involved in the program. Educators must concern themselves with the fair ness, humaneness, and accountability of their selection processes, particularly in view of the predictions of declining college-age population in the next two decades (Dietrich, 1981). Some criteria commonly used in the 153 admission process include: (1) biographical information, (2) GPA, (3) interviews, (4) letters of recommendation, and (5) admissions tests. The primary purpose of this investigation was to examine one segment of the physical therapy selection process in detail and secondarily to investigate other segments of selection processes in physical therapy and other professions to determine whether there are any reliable predictors of success in a profession and, if any, what they might be. The Allied Health Professions Admission Test (AHPAT) was examined to investigate the relationship among the AHPAT and several preprofessional criteria, five courses in the professional physical therapy curriculum, and the scores on the Licensure Examination, and to deter mine the relationships among all of the criteria and sex, age, ethnic background, marital status, and preprofes sional major. Based on the conflicting results reported in the literature regarding the AHPAT, the investigation was undertaken to attempt to resolve the issues raised, to consider alternatives, and to make recommendations regard ing the use of the AHPAT and/or other selection criteria for the most reliable and fairest procedure for selection in physical therapy education. The investigation was conducted at California 154 State University Northridge, Health Science Department, Physical Therapy Curriculum. The sample included 194 student members of nine separate classes matriculated into the curriculum between 1976 and 1981. Student files, examination reports, and returned letters of request (Appendix A) were used to obtain data for this study. The letters of request were used only for those students whose Licensure Examination scores were not available on examin ation reports. Strict confidentiality was observed in preparing the data for the statistical treatment used in this investigation. Data were analyzed using frequency distributions, simple linear regressions, and student t. test techniques. The .0005 alpha level of significance was applied to the simple linear regression technique due to the multiple tests that were performed on the same data in that portion of the analysis. The .05 level of significance was used for the student t tests. The hypotheses were tested to determine the relationship between the AHPAT and selected prepro fessional courses, preprofessional GPA, selected pro fessional courses, clinical education, professional GPA, and scores on the Licensure Examination of the individual student when grouped by sex, age, ethnic background, mari- 155 tal status, preprofessional major, and number of English courses taken. Hypothesis One There were only three out of eleven preprofes sional courses that were significantly related to the AHPAT. Based on these findings, the individual prepro fessional courses do not appear to be a reliable predictor of scores on the AHPAT, Hypothesis Two In the overall groups, two of the five selected professional courses were related to the quantitative, biology, and reading comprehension scores on the AHPAT, plus the total score. When the groups were separated by sex, ethnic background, the single category in marital status, and by the number of English courses taken, there were no significant relationships. There were two signif icant relationships for marrieds: (1) between Anat and AHPAT-Q, (2) between physical and AHPAT-T. Based on the tests run the AHPAT does not appear to be a good predictor of performance in the professional curriculum. The results are shown in Table 11. Hypothesis Three Because the relationship between clinical educa tion and the AHPAT was not tested due to the small number of individuals in the total sample who did not success 156 fully complete all internships, no conclusions can be drawn regarding the relationship. In light of the fact that a new evaluation of clinical performance instrument is now being utilized,.it appears that there is now a more objective means of evaluating the student intern. There is no section on the AHPAT that measures clinical skills; therefore, one would question if there was a relationship between clinical education and the AHPAT. Hypothesis Four The professional curriculum GPA showed a rela tionship to the AHPAT when the groups were combined. It is of interest to note that the AHPAT scores that were related were not in the basic sciences, but rather with quantitative and reading comprehension along with total score. It appears that the specific sections of the AHPAT mentioned previously may be a measure of general educa tional skills as measured by the overall professional curriculum GPA, Hypothesis Five When comparing the AHPAT and the Physical Therapy Licensure Examination, the tests show the AHPAT to be related to the Licensure Examination consistently in the basic science area and in all three parts of the Licensure Examination plus total score with the total score on the 157 AHPAT, One might suggest that test-taking skills may be a factor in these results. Hypothesis.Six When tests were applied to the data using sex as the variable, there were varied significantly related factors between preprofessional courses and AHPAT scores. Males did not perform as well in the quantitative area as the review of literature would suggest. Female scores in the quantitative area were related to preprofessional algebra, plus the professional GPA was related to several section of the AHPAT, It is interesting that relation ships were significant in areas not considered to be strong academic areas for females, except for reading comprehension which the literature suggests is a strong area. One might consider that individuals applying to physical therapy professional programs may not necessarily fall into the typical sex role norms due to the fact that prerequisites mainly fall into scientific and mathematical subject areas, which is not the case in non-medically related professions. Hypothesis Seven There appears to be no significant relationship between age and scores on the AHPAT except in the area of 158 verbal ability which may be a function of language experi ence (See Table 17. Hypothesis.Eight Ethnic backgrounds is not a significant factor in the prediction of success according to the results of this investigation, however, it must be noted that 89.2 percent of the total sample were Caucasians. Student t tests applied to the data to compare Caucasians and others showed only three significantly different factors, with all other factors not related. There were no significant differences between Caucasians and Orientals when student tests were applied to the data. The results indicate that there are no significant relationships between ethnic background and scores on the AHPAT. Prerequisite requirements may be a factor in equal izing the groups. Hypothesis Nine When comparisons were made between marital status and scores on the AHPAT, the most significant relationship was GPA. Grade point average seems to be the most reli able predictor of success in other studies reviewed in the literature. Hypothesis Ten As with marital status and the AHPAT, preprofes sional major compared to AHPAT indicates that grade point 159 average is significantly related to scores on the AHPAT. Questions to be Answered Two questions were considered in this research; (1) how the scores of the AHPAT were utilized at CSÜN, and (2) what value was placed on the scores in the selection process at CSUN. Selection was based on a point system at CSUN using the following criteria: (1) prerequisite GPA, (2) personal interviews, (3) work experience, (4) letters of recommendation, (5) application status, (6) biographical data, and (7) AHPAT scores. The maximum number of points assigned to the AHPAT score was five which represented 2.17 percent of the total possible points. Major criteria were prerequisite grades and personal interviews. In the past, reading comprehension and verbal ability were regarded as being particularly indicative of future perfor mance in the professional curriculum; however, the results of this research did not support this. Nonhvpothesized Findings Regression Analysis of Preprofes sional and Professional Criteria In order to explore further into the complex task of student selection, the investigation included in the analysis simple linear regressions to compare pre professional criteria versus professional criteria and 160 licensure examination performance to determine the rela tionships among the criteria. The independent variables were the following major groupings: (1) sex, (2) ethnic background, (3) marital status, (4) preprofessional major, and (5) number of English courses completed at college level prior to the AHPAT. The regression analysis technique applied to the data when grouping by sex indicated no significant corre lations between preprofessional and professional criteria for males; however, there were several correlations for females (Table 9). When grouped by ethnic background, there was no relationship between preprofessional and professional criteria for the other group (Black, Mexican-American, Oriental). Several criteria were related for Caucasians (Table 10). The regression analysis applied to marital status grouping indicated little relationship for marrieds, only pPhysiol with Physio and GPA. There were several corre lations for singles (Table 11). The analysis of preprofessional criteria versus professional criteria for non-health science majors indi cated little relationship among criteria; second semester Physics (pPhysics 2) was related to Biomechanics (Biomech) and pGPA was related to GPA. The statistical treatment 161 indicated several correlations for health science majors (Table 12). Prerequisite GPA and prerequisite anatomy were the two most consistent variables related to professional curriculum criteria and licensure examination scores Part I, Part II, and BD-TOT. There were no significant corre lations with Part III of the licensure examination. Part I of the licensure examination was basic sciences which included anatomy, kinesiology, physical sciences, pathology, physiology, psychology, and growth and development; Part II was clinical sciences which included internal medicine, neurology, orthopedics, sur gery, pediatrics, geriatrics, psychiatry, and community health ; Part III, theory and procedure, included electro therapy, infrared, ultraviolet and ultrasonics, bandaging and aseptic techniques, prosthetics and orthotics, hydro therapy, massage, therapeutic exercise and functional training, evaluation-principles and procedures, profes sional ethnics, attitudes, liability, and administration (American Physical Therapy Association, 1983). Prerequisite anatomy was the most consistent pre dictor for the licensure examination when the combined groupings were analyzed. Prerequisite GPA was consis tently related to professional curriculum GPA and licens ure examination scores. The latter finding was consistent 162 with the findings of Landen (1977); Laurencelle, Kay, and Edelsberg (1979), Schimpfhauser and Broski (1976); and Wiesseman (n.d.). Prerequisite first semester chemistry (pChem 1), PChem 2, pPhysic 1, and pPhysic 2, and pBiostat were related to certain professional criteria when in separate groups (Table 15, 18, 20, and 23). The t test analysis of GPAs (Table 29) suggested that perhaps the students with lower GPAs (<3.09) per formed equally as well in the clinical setting (intern ships) as did the students with higher GPAs 03.09). The t tests also indicated that the females in the sample performed better than males in 6 of 22 preprofessional and professional criteria (pChem 2, pPhysic 1, pGPA, Neuro, GPA, and AHPAT-RC) (Table 16). Higher performance for females in reading comprehension was consistent with the findings of Hilton and Berglund (1974); however, there was no evidence to support the authors* findings regarding better performance of males in mathematical skills. Recommendations Based on the results of the investigation, this researcher recommends the following: 1. The Allied Health Professions Admission Test should be discontinued as a tool for selection of physical therapy students in programs presently requiring the exam 163 ination. This recommendation was specifically made to the Physical Therapy Curriculum selection committee at Cali fornia State University Northridge and was accepted July, 1983. 2. The possible discontinuation of the AHPAT indicates a need for the development of an admissions examination that would be more appropriate to physical therapy and one which would include the affective domain, not presently included in the AHPAT, along with appro priate biological and physical sciences and social and communicative skills. A pilot study would be necessary to validate the examination. Such a task has already been initiated by Soderburg at the University of Iowa (1983). Efforts should be made toward development of a culture- free, non-biased examination. 3. Replication of this investigation should be accomplished, using only upper-division prerequisites as preprofessional course criteria which would sample cours- ework generally taken close to the application deadline and perhaps more indicative of current academic abilities. Added to the professional criteria could be the results of the Blue Max (a new competency based clinical performance evaluation tool being partially used for the first time at CSUN in summer school 1983 Clerkship experiences, to pre pare for its complete use in fall 1983). The Blue Max 164 would provide more quantifiable rating of clinical perfor mance . 4. Replication of this investigation should be done comparing all of the prerequisite and professional coursework to the AHPAT rather than to selected courses. 5. Further investigation is needed into the inter view as a selection tool, appropriate methods of training interviewers, and a reliable method of scoring the inter view such as that suggested by Shepard (1980). 6. Multifactored computer-based analysis of candidates applications, such as suggested by Robinson, Braaten, and Bailey (1979) and Weiss (1970), should be investigated. It appears there is much valuable infor mation that could be gleaned from an application blank that many admission and selection committees overlook. 7. Replication of this investigation including applicants not selected for the professional program and comparing the unsuccessful applicants to the successful applicants should be investigated ; however, only pre professional criteria could be compared. 155 BIBLIOGRAPHY 166 BIBLIOGRAPHY Aiken, R. (1971). Psychological and educational testing, Boston; Allyn and Bacon. American Physical Therapy Association. (1976). Clinical education in physical therapy; Present status/ future needs (DHEW, PHS, HRA, BHM, DAHP Contract No. 1-AH-4nil2). American Physical Therapy Association. (1983). Careers in Physical therapy. Alexandria, Virginia. (a) American Physical Therapy Association. ( 1983). Inf or- mation for candidate physical therapy licensure examination. Alexandria, Virginia. (b) Anastasi, A. (1975, Spring). Commentary on precocity project. Journal of Special Education, â(1), 93-103. Astin, H. S, (1975, Spring). Sex differences in mathe matical and scientific precocity. Journal of Special Education, S(1), 79-91. Baird, L. L. (1976). Biographical and educational correlation of graduate and professional school admissions test scores. Educational Psychology 26, 415-420. Bradshaw, R. 0. F. (1975, September 8). Use of the interview in the selection of candidates_for Physical therapy training. Unpublished manuscript. University of Toronto, Faculty of Medicine, Depart ment of Rehabilitation Medicine, Toronto. Broski, D. C., Schimpfhauser, F. T., & Cook. S. (1977, Fall). The AHPAT; Does it predict success? Journal of Allied Health, pp. 21-25. Burgess, M. M., Calkins, V., & Richards, J. M. (1972), Summer). The standard interview: A selection device. College and University ^ 47, 400-407. Carline, J. D., Cullen, T. J., Scott, C. S., Shanon, N. F., & Schaad, D. (1983, January). Predicting per formance during clinical years from the new medical 167 college admissions test. Journal of Medical Education, 58, 18-25. Cattell, R. B., & Butcher, H. J. (1968). The prediction of achievement and creativity. Indianapolis, NY; Bobbs-Merrill. Chaisson, G. M. (1976, Spring). Student selection: Logic or lottery. Journal of Allied Health, pp. 7-16. The Chronicle of Higher Education. Fact file, employment outlook for 1990, October 21, 1981. Crossland, F. E. (1965, March). Politics and policies in college admissions. Phi Delta Kappan, &6(7), 299-302. Crossland, F. E. (1976, September). Equality, equity, equilibrium, policy toward access to higher educa tion . Paper presented at March, 1976, I8th Annual Conference of Council of Higher Education in Ameri can Republics, New York, NY. Dietrich, M. C. (1981, November). Putting objectivity in the allied health student selection process. Journal of Allied Health, pp. 226-239. Dunn, 0. J, & Clark, V. A. (1974). Applied statistics! Analysis of variance and regression. New York; John Wiley & Sons. El-Din, D. (1977; Winter). Admissions in a school of allied health: An organization model. Journal of Allied Health, pp. 16-22. Giristi, J. P. (1964, Winter). High school average as a predictor of college success: A survey of the literature. College and University, 39(3), 253-258. Gordon, M. E., Arvey, R. D., Daffron, W. C. Jr., & Umberger, D. L. (1974). Racial differences in the impact of mathmetics training at a manpower develop ment program. Journal of Applied Psychology, 22(3), 253-258. 168 Gray, J. L., & Knief, L, M. (1974-1976, Fall-Summer)• The relationship between cognitive style and school achievement. Journal of Experimental Education, pp. 43-44; 67-71. Green, V. L., & Cox, R. C. (1978). Development of an instrument in physical therapy admission. Unpub lished special interest paper presented at National Conference American Physical Therapy Association, Las Vegas, NV, University of Pittsburg, School of Allied Health Related Professions, Department of Physical Therapy, Pennsylvania. Grove, W. J. The problem of selection. In G. E. Miller (Ed.). (1976). Medical education and contemporary world (DHEW Publication No. NIH 77-1232). Chicago, II.: Author Haney, R., Michael, W . B., & Martois, J. (1976). The prediction of success of three ethnic groups in the academic components of a nursing-training program at a large metropolitan hospital. Educational and Psychological Measurement, 26, 421-431. Hills, J. R., & Klock, J. A. (1966, Winter). "Total" vs "Academic" high school average in college grade prediction. College and University, 41 , 231-232. Hilton, T. L., & Berglund, G. W. (1974, January). Sex differences in mathematics achievement— longitudinal study. Journal of Educational Research, 67(5), 231-237. Johnson, R. W., Arbes, B. H., & Thompson, G. (1974, November-December). Selection of occupational therapy student. The American Journal of Occupa tional Therapy, 28(10), 597-601. Katzell, M. E. (1977, Fall). The allied health profession admission test. Journal of Allied Health, 6(4), 14-20. Kerr, C. (1978, Winter). Solving the selective admis sions dilemma, a two state model for higher educa tion. Phi Kappa,Phi, 28(1), 4-6. The Kiplinger Washington Letter. The Kiplinger Washington Editors, 1729H. Street N.W., Washington, D.C. 20006, Washington, August 13, 1982. 169 Landen, B. R. (1977). Selection of students for Physical therapy education. Unpublished doctoral dissertation, Georgia State University, Atlanta, Georgia. Laurencelle, P., Kay, B. A., & Edelsberg, S. P. (1979). A study of the applicant pool for the programs in physical therapy, class of _1977-78, Northwestern University Medical School. Northwestern Univer sity, Medical School, Department of Health Science and Art. Leiken, A. M., & Cunningham, B. M. (1980, May). The predictive ability of the allied health profession admission test. Journal of Allied Health, pp. 132-138. Lucci, J. A. (1974). An approach to the selection of students. The American Journal of Occupational Therapy, 28(3), 91-93. Lyman, S. M. (1971). Contrasts in the community organi zation of Chinese and Japanese in north America. In N. R. Yetman & C. H. Steele (Eds.), Majority and minority (pp. 285-296). Boston, MA: Allyn & Bacon. McClothlin, W. J. (I960). Patterns of professional education. New Yorki G. B. Putnam’s Sons. McGrath, E. (1983, March 28). Reported by P. Delaney, Confucian work ethic/Asian-born students head for the head of the class. Time, p. 52. Miller, G. E. (Ed.). (1976, September 13-14). Medical education and the contemporary world (DHEW Publi cation No. NIH 77-1232). Chicago, IL.: Author Plapp, J. M., Psathas, G., & Caputo, D. V. (1965). Intellective prediction of success in nursing school. Educational and Psychological Measurement, 25(2), 565-577. The Psychological Corporation, Professional Examination Division. (1974, December). Report on standardiza tion and preliminary validation of the allied health professions admission test. New York, NY; Author. 170 The Psychological Corporation, Professional Examination Division. (1974-1975). Allied health professions admission test: A guide to interpretation of test reports (Report No. FED 7987). New York, NY: Author. (a) The Psychological Corporation, Professional Examinations Division. (1974-1975). Allied health professions admission test : The first year, 1974-1975. New York, NY: Author. (b) The Psychological Corporation, Professional Examinations Division. (1975-1976. Allied health professions admissions test: Annual report, 1975-1976. New York, NY: Author. The Psychological Corporation, Professional Examinations Division. (1976, October). Allied health profes sions admission test: Reliability and validity of the AHPAT- New York, NY; Author. The Psychological Corporation, Professional Examinations Division. (1976-1977). Allied health professions admission test : Annual report, 1976-1977. New York, NY; Author. (a) The Psychological Corporation, Professional Examinations Division. (1976-1977). Report on 1976-1977 applicants to physical therapy programs. New York, NY; Author. (b) The Psychological Corporation, Professional Examinations Division. (1977-1978). Allied health professions admission test: Annual report, 1977-1978. New York, NY; Author. (a) The Psychological Corporation, Professional Examinations Division. (1977-1978). Allied health professions admission test : Report on 1977-1978 applicants to Physical therapy programs. New York, NY: Author, (b) The Psychological Corporation. (1978-1979). Allied health professions admission test: 1978-1979 announcement. New York, NY: Author. The Psychological Corporation, Professional Examinations Division. (1979-1980). Allied health professions 171 admission test: Report on 1979-1980 applicants to Physical therapy programs- New York, NY: Author. The Psychological Corporation, Professional Examinations Division. (1980-1981). Allied health professions admission test: Report on 1980-1981 applicants to physical therapy programs. New York, NY: Author. Robinson, T. C., Braaten, J. N., & Bailey, R. D. (1979, May). An automated admissions system for allied health schools. Journal of Allied Health, pp. 90-95. Samuda, R. J. (1971). Cultural discrimination through testing. In N. R. Yetman & & C. H. Steele (Eds.). Majority and minority (pp. 490-505). Boston, MA: Allyn & Bacon. Schimpfhauser, F. T., & Broski, D. C. (1976, Winter). Predicting academic success in allied health cur riculum. Journal of Allied Health, 2, 35-46/ Schwartz, A. J. (1975). The schools and socialization. New York, NY: Harper & Row. Seymour, R. J., McDougall, R. V., Wadsworth, C. T., & Sanders, B. R. (1982, May). Use of the interview in admissions. Journal of Allied Health, 11, 88-95. Shepard, K. F. (1980, May). Use of small group inter views for selection into allied health educational program. Journal of Allied Health, 2(2), 85-94. Soderburg, G. (1983, March 4). Results of a question naire to determine the feasibility of developing and administering a physical therapy admissions test. Unpublished manuscript. Sternburg, R. J. (1973, Spring). Cost-benefit analysis of the Yale admissions office interview. College and University, 48, 154-164. Stevens, E. I. (1973). Grading system and student mobility. Educational record, 24(4), 308-312. Tratter, M. J., & Fordyce, W. E. (1975). A process for physical therapy student selection— The Q technique. Physical Therapy, 22, 151-156. 172 Watson, J. J., Anthony, T. F., & Crowder, W. S. (1973, Spring). A heuristic model for law school admissions screening. College and University. 48, 195-204. Weinstein, P., & Milgron, P. (1977). The applicant’s perception of the admission interview. Journal of Dental Education, iLl(3), 149-152. Weiss, K. P. (1970, Winter). A multi-factor admissions predictive system. College and University, 45, 203-210. Westen, R. J., & Lenning, A. T. (1973, Fall). Predic tion at a highly selective institution after cor rections have been made for selections: ACT versus SAT. College and University, 48, 68-76. Wiesseman, J. (Undated). Does the AHPAT add enough predictive ability to the college GPA to justify its use? Unpublished manuscript. Williams, J. H., Pengav, R. E., Sachs, L., Melica, B., & Comiscioni, J. S. (1977). A computer assisted admissions process. Journal of Medical Education, 22, 384-389. Willingham, W. W. (1965, Spring). The application blank as a predictive instrument. College and University. 40, 271-281. Wingard, J., & Williamson, J. (1973). Grades as predic tion of physician’s career performance: An evalua tion literature review. Journal of Medical Education, 48, 311-322. Yetman, N. R., & Steele, C. H, (1971). Majority and minority (2nd ed.). Boston, MA: Allyn and Bacon. APPENDIX 174 APPENDIX Department of Health Science California State University, Northridge 18111 Nordhoff St. Northridge, CA 91330 May 27, 1 Dear The time has finally arrived for collecting data for my dissertation. In order for the data to be complete, I need the identification number from your State Board exam scores. The Examining Committee stopped reporting scores with names identified as of the June 6, 1979, exam. If you recall, my dissertation is based on comparing the scores on the Allied Health Professions Admissions test which you took as part of your selection into our program with a number of other items, one of which is the scores on the Board exams. All of the data is coded and no one is identified in the study to protect your privacy. If you would be willing to send your identification number of your individual scores on the exam, I would appreciate it very much. If you do not know it, you can request your identification number from; Board of Medical Quality Assurance Physical Therapy Examining Committee 1430 Howe Avenue Sacramento, CA 95825 Telephone: (916) 920-6373 I will look forward to hearing from you. I have enclosed a stamped envelope and detachable form (below) for your convenience. Thank you very much. Very truly yours. Deona Lilly Name _____________ _ Identification Number ________________ State Board Scores Parti ______Partll Fartlll _____ ___ Average _______ Date Taken ___________________</s>

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A study of Japanese students at the University of Southern California, 1946-1980: vocational impact of American academic experience on Japanese students after returning to Japan

PDF

The relationship between cultural-value orientation and achievement in three learning environments: Lecture, small-group, and self-study

PDF

The role of microbes in the formation of the tubeworm fouling community in Fish Harbor, Los Angeles Harbor, California

PDF

The interaction of learners' spatial ability aptitudes with filmic schematic operations

PDF

An analysis of traditional birth attendant (TBA) programs in selected third world countries

PDF

The 'Widmar' (1981) decision and the status of religious liberty on the public college and university campus: Antecedents, effects, and prospects

PDF

The effects of violating selected item-writing principles on the difficulty and reliability of multiple-choice tests for health profession students

PDF

Resocialization of older adults to work roles through worklife autobiography

PDF

Out of many...: A social history of the homosexual rights movement as originated and continued in Los Angeles, California

PDF

An analysis of the need and opportunity for the continuing liberal education of doctors of medicine

PDF

Chemical and immunological studies on a heterophile antigen extracted from bovine erythrocyte stroma

PDF

Studies on the aerobic oxidation of fatty acids by bacteria

PDF

The innervation of the teeth and periodontium of the rat

PDF

Studies of frog retinal ganglion cell responses to temporal, spatial and chromatic stimuli

PDF

Self-perceived job satisfaction and dissatisfaction of urban elementary classroom teachers

PDF

Endocrine aspects of aggression and dominance in champanzees of the Kibale Forest

PDF

Nude mouse grown human nephroblastomas

PDF

Cation localization in growth hormone and prolactin cells of the rat pars distalis

PDF

Numerical analysis of ecological survey data

PDF

Studies of two naturally occurring compounds which effect release of acetylcholine from synaptosomes

Asset Metadata

Creator Lilly-Masuda, Deona M. (author)

Core Title The Allied Health Professions Admission Test : its role in selection for physical therapy programs

School Graduate School

Degree Doctor of Philosophy

Degree Program Education

Degree Conferral Date 1984-05

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag education,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c30-273586

Unique identifier UC11227398

Identifier DP24996.pdf (filename),usctheses-c30-273586 (legacy record id)

Legacy Identifier DP24996.pdf

Dmrecord 273586

Document Type Dissertation

Rights Lilly-Masuda, Deona M.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA