EFFECT OF ACADEMIC MOTIVATION AND OTHER SELECTED CRITERIA ON ACHIEVEMENT OF FIRST AND SECOND SEMESTER SHORTHAND STUDENTS, PH. D. MICHIGAN STATE UNIVERSITY LEONARD JAMES VARAH 1966 THESIS This is to certify that the "I thesis entitled EFFECT OF ACADEMIC MOTIVATION AND OTHER SELECTED CRITERIA ON ACHIEVEMENT OF FIRST AND SECOND SEMESTER SHORTHAND STUDENTS presented by Leonard James Varah has been accepted towards fulfillment of the requirements for Eh...D.._degree influsinass and Distributive Education /;‘, /’} / as/e/ ,- / /4/;79fl<%/ gfi .,¢y4229%// Major professor 7 / 1/ , 7—) I /’ , -’ . MA 1‘. DatcAAk/gE/4/ iii 2/72 >7 0-169 I I I ‘VVT‘ L I B R A R Y Michigan State University ”VI "VII 39v .IIN‘Z? .; barb... .I'I u.IIl|I'.I..l.II’III.luIIIfIII\VI:I ABSTRACT EFFECT OF ACADEMIC MOTIVATION AND OTHER SELECTED CRITERIA ON ACHIEVEMENT OF FIRST AND SECOND SEMESTER SHORTHAND STUDENTS by Leonard J. Varah Purpose of the Study: The purpose of the study was to determine the predictive value, if any, of the Michigan M-Scales, a test of academic motivation, total score or subscores for predicting achieve- ment of eleventh grade girls in first and second semester of Gregg Shorthand when used individually or in combination with the total score or a subscore of an estimate of mental ability. This predictive value was then compared with the predictive value of academic grade point average, ninth grade English grades, and tenth grade English grades to determine the most accurate predictor of shorthand achievement. Procedures and Analysis of the Data: The information for the present study came from the first and second semester shorthand classes of eight senior high schools in Central and Southern Michigan. The pOpu- lation of the present study was eleventh grade girls in beginning Gregg Shorthand. The three types of data Leonard J. Varah concerning the population collected were: (1) an estimate of mental ability, (2) grades for ninth and tenth grade English and overall academic grade point average, and (3) a motivation score as measured by the Michigan M-Scales. Shorthand achievement was measured by teacher grades at the end of the first and second semesters and by a uniform shorthand examination at the end of the first and second semesters. The data was processed by the Control Data Corporation 3600 Computer through the use of the CORE program. The three statistical procedures that were used in the analysis of the data were: (1) estimate of reliability of the motivation score; (2) mean test of significance; and (3) correlation analysis. The Major Findings of the Study: The major findings of the study were: (1) Academic motivation as measured by the Michigan M—Scales is a factor in learning in first semester Gregg Shorthand but is not a factor in learning in second semester Gregg Shorthand. (2) The Michigan M-Scales when used in combination with an estimate of mental ability did significantly increase the precision of prediction by an estimate of mental ability in predicting the achievement of eleventh grade girls in first semester Gregg Shorthand. (3) The Word Rating List, a subtest of the Michigan M-Scales, was found to be a significant predictor and a Leonard J. Varah consistently significant predictor of shorthand achievement for both first and second semester of Gregg Shorthand. It was concluded that the academic self-concept of the student as measured by the Word Rating List is a factor in learning in first and second semesters of Gregg Shorthand. (4) The best single predictors of first semester Gregg Shorthand were: (a) grade point average; (b) ninth grade English grades; (0) tenth grade English grades; (d) estimate of mental ability. (5) The best single predictors of second semester Gregg Shorthand were: (a) Shorthand I teacher grades; (b) grade point average; (0) tenth grade English grades; and (d) mental ability. EFFECT OF ACADEMIC MOTIVATION AND OTHER SELECTED CRITERIA ON ACHIEVEMENT OF FIRST AND SECOND SEMESTER SHORTHAND STUDENTS by Leonard J; Varah A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY with Major in Business and Distributive Teacher Education 1966 TABLE OF CONTENTS Chapter I. INTRODUCTION Statement of the Problem Purpose of the Study Hypotheses Need for the Study Theory for Using the M-Scales Delimitations Limitation Definition of Terms II. REVIEW OF LITERATURE Review of Literature on Prognosis Review of Specific Studies Relating to Shorthand Prognosis Synthesis of Research Relating to Shorthand Prognosis Reported Research Using Mental Ability Reported Research Using English Grades Reported Research Using Overall Grade Point Average Summary of Shorthand Prognosis Review of Literature Review of Literature on Academic Motivation Review of a Specific Study Relating to Academic Motivation-—The Farquhar Study Implications of Related Literature for the Present Study III. PROCEDURES Collection of Data Sources of Information Nature of the Data Collected Processing of Data Preparation Processing ii I]: ID (D OOCDVKJJKDNI-J I—‘ K» M) b.) b.) mun» I—‘ I-' I-‘ I—’ O\ 0\ U1 4.1" NI—‘H fl 0\ F 44 46 46 1+6 47 51 51 52 Chapter Page Analysis of Data 52 Reliability of Estimate 52 Mean Tests of Significance 54 Correlation Analysis 55 Null Hypotheses 58 IV. FINDINGS 59 Reliability 59 Mean Differences Between Farquhar Findings and the Present Study 60 Test of Hypotheses 64 First Semester 64 Second Semester 76 First Semester Uniform Shorthand Examination 84 Second Semester Uniform Shorthand Examination 90 Summary of Findings 93 Summary of Findings of Hypotheses 98 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 100 Summary 100 Conclusions 105 Recommendations 112 BIBLIOGRAPHY 115 APPENDIX 126 fl?) iii Table II. III. IV. VI. VII. VIII. IX. XI. LIST OF TABLES (Appearing in the Appendix) Facsimile Letter to School Administrators Frequent Reasons for DrOpping Shorthand as Checked by Forty-five Students Partial Summary of Correlations Between Intelligence Test Results and Shorthand Achievement Reported in Previous Investigations Partial Summary of Correlations Between English Grades and Tests and Shorthand Achievement Reported in Previous Investigations Partial Summary of Correlations Between Academic Achievement and Shorthand Achievement Reported in Previous Investigations Size of Schools According to Student Enrollment, September, 1964 List of Courses Used in Determining Grade Point Average List of Courses Rejected in Determining Grade Point Average Summary of Kuder-Richardson #20 Formula for Estimate of Reliability for the M-Scales for the Eight School Sample of Female Shorthand Students Standard Deviations and F Tests for the Female Population of the Farquhar Study and for the POpulation of the Present Study Means and t-Tests for the Female POpulation of the Farquhar Study and for the POpulation of the Present Study iv Page 122 123 124 128 132 134 135 136 137 138 139 Table XII. XIII. XIV. XV. XVI. XVII. XVIII. Zero Order Correlations Between First Semester Shorthand Achievement and Predictor Variables for the Population of the Present Study Intercorrelation Matrix of M-Scales, Aptitude Measure, and Shorthand I Grades for the Population of the Present Study Tests of Significance of Increase in Precision of Prediction by Combining Each of the Sub-Scores and the Total Score of the M-Scales with the Total Score of the American Council on Education Psychological Examination in Estimating First Semester Shorthand Achievement for Eleventh Grade Girls Tests of Significance of Increase in Precision of Prediction by Combining Each of the Sub-Scores and the Total Score of the M-Scales with the Linguistic Sub-Score of the American Council on Education Psychological Examination in Estimating the First Semester Shorthand Achievement for Eleventh Grade Girls Tests of Significance of Increase in Precision of Prediction by Combining Each of the Sub-Scores and the Total Score of the M-Scales with the Quantitative Sub- Score of the American Council on Education Psychological Examination in Estimating the First Semester Shorthand Achievement for Eleventh Grade Girls Zero Order Correlations Between Second Semester Teacher Grades and the Predictor Variables for the Population of the Present Study Tests of Significance of Increase in Precision of Prediction by Combining Each of the Sub-Scores and Total Score of the M-Scales with the Total Score of the American Council on Education Psychological Examination in Estimating Second Semester Shorthand Achievement for Eleventh Grade Girls Page 140 141 142 143 144 145 146 Table Page XIX. Tests of Significance of Increase in Precision of Prediction by Combining Each of the Sub-Scores and Total Score of the M-Scales with the Linguistic Sub-Score of the American Council on Education Psychological Examination in Estimating Second Semester Shorthand Achievement for Eleventh Grade Girls 147 XX. Tests of Significance of Increase in Precision of Prediction by Combining Each of the Sub-Scores and Total Score of the M-Scales with the Quantitative Sub-Score of the American Council on Education Psychological Examination in Estimating Second Semester Shorthand Achievement for Eleventh Grade Girls 148 XXI. Correlations Between Predictor Variables and Uniform Shorthand I Examination, Uniform Shorthand II Examination, Shorthand I Teacher Grades, and Shorthand II Teacher Grades 149 XXII. Intercorrelations Between Uniform Shorthand I Examination, Uniform Shorthand II Examination, Shorthand I Teacher Grades, and Shorthand II Teacher Grades 151 XXIII. Correlations Between Predictor Variables of the Present Study and Uniform Shorthand 11 Examination, Shorthand 1 Teacher Grades, and Shorthand II Teacher Grades 152 XXIV. Intercorrelations Between Sub-Scores and Total Score of Uniform Shorthand II Examination, Shorthand I Teacher Grades, and Shorthand II Teacher Grades 153 XXV. Shorthand I Grades for Each School 154 XXVI. Shorthand II Grades for Each School 155 XXVII. Uniform Examination for First Semester Shorthand 156 XXVIII. Uniform Examination for Second Semester Shorthand 167 vi Table Page XXIX. EXplanation of Generalizability 178 vii ACKNOWLEDGEMENTS The researcher would like to express his sincere appreciation to the following people who made this study possible: To Dr. Stanley Hecker and Dr. Rollin Simonds for their guidance and advice. To Dr. W. W. Farquhar, author of the Michigan M—Scales, for permission to use the M-Soales in this research project, for his assistance with the statistical procedures, and for his advice throughout the research project. To the business teachers, school administrators, and shorthand students who requested to remain anonymous for providing the data for this research project. To Dr. Peter G. Haines for his advice and encouragement throughout the program and the research project. Especially to Dr. Helen H. Green, Committee Chairman and Project Director, for her encouragement, assistance, advice, and guidance throughout the program and research project. And to Marjorie for untold hours of encouragement and to James for waiting. viii CHAPTER I INTRODUCTION Why do some students with low mental ability do well in shorthand while some students with high mental ability do very poor work in shorthand? This is a question that short- hand teachers continue to ask. Researchers and scholars have studied this problem. Through research, the conclusion has been made that grade point average, English grades, and mental ability are the best predictors of shorthand achievement even though these predictors are not sufficiently accurate to make prediction reliable. As a result, researchers have hypothesized that motivation might possibly be a factor in the shorthand success; and if an instrument could be devised to accurately measure motivation, more accurate prediction of shorthand success might result through its use in combination with other predictors. Statement of the Problem: This study was an attempt to determine a partial answer to the question: ”What part does motivation play in the student's shorthand achievement?” By using an experimental test of academic motivation—-the Michigan M-Scales—-and an estimate of mental ability as measured by the American 1 Council on Education Psychological Examination for high school students, this study attempted to determine the value of academic motivation as measured by the Michigan M—Scales, in combination with the estimates of mental ability, in predicting achievement in first and second semester shorthand. This study was particularly concerned with the relation- ship of motivation to shorthand achievement, with identifying degrees of motivation, and of determining the extent to which measurement of motivation might add to precision in predicting shorthand achievement. Purpose of the Study: The purpose of this specific study was to determine the predictive value, if any, of the Michigan M-Scales total score or subscores for predicting achievement of eleventh grade girls in first and second semester of Gregg Shorthand when used individually or in combination with the total score or a subscore of an estimate of mental ability. This pre— dictive value was then compared with the predictive value of academic grade point average and ninth and tenth grade English grades to determine the most accurate predictor of shorthand achievement. Specific purposes of the investigation were: 1. To determine whether each predictor variable by itself significantly pre- dicted achievement of eleventh grade girls in first and second semester of Gregg Shorthand. 2. To determine whether the predictor variables as a group significantly predicted achievement of eleventh grade girls in first and second semester of Gregg Shorthand. 3. To determine whether the total score of the Michigan M-Scales or each subscore when combined with an estimate of mental ability significantly increased the pre- cision of prediction of achieve- ment for eleventh grade girls in first and second semester Gregg Shorthand. Hypotheses: The research hypotheses tested in this study were: 1. Each predictor variable will predict achievement of eleventh grade girls in first and second semester of Gregg Shorthand. 2. The predictor variables as a group will predict achievement of eleventh grade girls in first and second semester of Gregg Shorthand. 3. The Michigan M-Scales when combined with an aptitude measure will increase the precision of prediction of achievement for eleventh grade girls in first and second semester of Gregg Shorthand. Need for the Study: An informal review of studies completed in shorthand prognosis indicated there is no known reliable predictor of success in beginning shorthand.1’2’3’u’5’6' Osborne in 1943 summarized all studies in shorthand prog— nosis when she said: . . . none of the correlations between the shorthand criterion and single tests or between the criterion and combinations of tests is high enough to make prediction valuable except in the negative sense.7 Leslie states: The problem in prognosis for shorthand (not for transcription) is to determine on the one hand the intensity or willingness and extent of coop- eration that may be expected in the learning of shorthand and on the other hand the general mental 1Ruth Irene Anderson, ”An Analysis and Classification of Research in Shorthand Transcription” (unpublished Doctoral dissertation, Indiana University, 1946). 2Edward E. Byers, ”Construction of Tests Predictive of Success in First Year Shorthand" (unpublished Doctoral dissertation, Boston University, 1959). 3Inez Frink, ”A Comprehensive Analysis and Synthesis of Research and Thought Pertaining to Shorthand and Tran— scription, 1946-1957” (unpublished Doctoral disseflmflion, Indiana University, 1961). “Billy T. Hutson, "Prognosis of Achievement in First- Year Gregg Shorthand Simplified" (unpublished Master's thesis, University of Tennessee, 1951). 5Agnes E. Osborne, The Relationship Between Certain Psychologipal Tests and Shorthand Achievement (New York: Bureau of Publication, Teachers College, Columbia Univer- sity, 1943). 6Dorothy Veon, The Relationship of Learning Factors Found in Certain Modern Foreign-Language Aptitude Tests to the Prediction of Shorthand Achievement in College (Stillwater, Oklahoma: Oklahoma Agricultural and Mechan— ical College; Delta Pi Epsilon Research Award Series, 1950). 7Osborne, op. cit., p. 53. ability and the work and study habits of the prospective pupil. It is also true, . . . , that there are degrees of success or failure. All other things being equal (such as intensity of desire and extent of coop- eration), the student with the greater degree of general ability and with the better study and work habits will achieve a higher degree of skill and success than the student with a lower general mental ability, and with less well-developed work habits. Studies9,lo have shown that a student with a high mental ability will tend to do better work in shorthand than a student with low mental ability; however, variations in achievement exist for unknown reasons. Researchers and scholars have theorized that motivation is a major factor in the academic achievements of an individual. High motivation to learn will result in a student with average mental ability achieving a better grade than that achieved by a student with high mental ability but no motivation. The Byers study of shorthand prognosis concluded that Motivation, interest, desire, and need are present in every student in varying degrees. A method of measuring these factors might prove to be the con- clusive answer to future prognosis in shorthand. Existing shorthand aptitude tests do not include measures of them. 8Louis A. Leslie, Methods of Teaching Gregg Shorthand (New York: Gregg Division, McGraw-Hill Book Company, 1953), p0 2870 9Veon, 0p. cit. loOsborne, op. cit. 11Byers, op. cit., p. 82. Hutson concluded in 1957 that . . . at the present time predicting individual success in shorthand or in other school subjects merely by the use of tests has not been demon- strated where real achievement is demanded. The inability to ascertain or control motivation appears to restrict the success of tests in prognosis. It is evident that motivation or the determination to succeed cannot be measured. Since motivation cannot be measured, little can be eXpected from the use of tests to predict success in shorthand for level of motivation appears to affect performance significantly when other conditions are equal. Researchers and scholars such as Atkinson and McClelland,13 Bindra,14 and Kagan15 have studied the problem of measuring motivation but have not been able to produce an instrument that is usable for the classroom teacher; however, an instrument, known as the Michigan M-Scales, was recently created by Dr. W. W. Farquhar16 and his associates of the Michigan State University College of Education faculty. 12Hutson, 0p. cit. 13David C. McClelland, et al., The Achievement Motive (New York: Appleton-Century-Crofts, Inc., 1953). 1LIDalbir Bindra, Motiyation:r A Systematic Reinter— pretation (New York: The Ronald Press Company, 1959). 15Jerome Kagan and Gerald S. Lesser, Contemporary Issues in Thematic Apperceptive Methods (Springfield, Illinois: Charles C. Thomas, Publisher, 1961). 16William W. Farquhar, Motivation Factors Related to Academic Achievement, A Final Report of Cooperative Research Project No. 846 entitled "A Comprehensive Study of the Motivational Factors Underlying Achievement of Eleventh Grade High School Students" supported by the U.S. Office of Edgcation in 000peration with Michigan State University, 19 3. 7 Their research effort was to identify the factors of academic motivation and produce an instrument that would measure academic motivation on the high school level. This instrument has proven useful in predicting grade point average. Theory for Using the M-Scales: The Michigan M-Scales have proven useful in predicting grade point average by means of measuring certain personal factors. These factors are emotionality, conformity, unreliability, compulsivity, fantasy, conservatism, social distance, emotional stability, long-term versus short-term involvement in relation to time, independence versus dependence in solving complex problems, unique accomplish— ment versus avoidance of education, long-term versus short- term involvement, and high versus low job interests. It was theorized that the Michigan M-Scales battery will aid in predicting achievement of eleventh grade girls in first and second semester Gregg Shorthand for the reason that several of the factors being measured by the M—Scales (emotional stability, responsibility, unreliability, long- term involvement, and unique accomplishment) are factors that research has identified as being present in signifi— cantly greater quantity in the successful shorthand student than in the unsuccessful shorthand student.17 17Georgia Faye Powell, "An Analysis of Shorthand DrOp- outs at Ottawa Township High School” (unpublished Master's thesis, Illinois State Normal University, 1961), p. 48. Delimitations: The scope of this study had the following delimitations: 1. Subjects: The study was limited to girls in the eleventh grade who were enrolled in first and second semes- ter of Gregg Shorthand in the eight high schools partici- pating in this study. All boys were excluded for the reason that only a small number of boys were enrolled in shorthand in the participating schools; and as the M-Scales were created for each sex, the small number of boys precluded any statistical analysis of the male population as measured by the M-Scales. 2. Schools: The participating schools were limited to the eight senior high schools located in seven cities and towns in central and southern Michigan. 3. ggpp: The study was conducted during the first and second semesters of the 1964-1965 school year. 4. Teachinngethod: The study did not include an analysis of the teaching methods, motivation devices used in the classroom, equipment for teaching, objectives of the courses for each teacher, standards of the course, standards for grades by the teachers, professional preparation of the teachers, or the teacher's personality for the reason of generalizability. (See Appendix, page 178, for explanation). 5. Subject Matter: The Gregg Shorthand systems being taught in the schools were utilized, but not analyzed, for this study. 6. Personal Data: The subject's age, race, the distance he traveled to school, other classes taken by him during time of this study, his study habits, school attend— ance, participation in extra—curricular activities, home and family membership, parent's occupation or occupations, parents' marital status, socio-economic factors, or hour of the day of the shorthand class were not analyzed as this information was considered beyond the scope of this study. Limitation: This study had the following limitation: Teacher Grades: The teacher grades were used as the criteria of achievement in shorthand in this study for two reasons: 1. The Michigan M-Scales were created for use with teacher grades from a number of different schools. In order to utilize the M-Scales, such teacher grades were essential to the study. 2. The teacher grades represented the achievement of an individual in a particular group or class. Teacher grades were the determinants of achievement or nonachieve- ment. Teacher judgments were valued in this case for the reason that the teachers were certified, experienced teachers of Gregg Shorthand and were able to judge degrees of achievement. 10 Definition of Terms: The following definitions of terms were used in this study: 1. Academic Motivation: Academic Motivation is a combination of forces which initiate, direct, and sustain behavior toward a scholarly goal.18 2. Shorthand Achievement: Shorthand Achievement includes a, b, and 0 below: a. Shorthand I Achievement is the grade given by the teacher to the specific pupil for first semester accom- plishment in shorthand. The grade may include whatever the teacher may consider important--speed of reading, knowledge of theory, speed of taking dictation, and general subjective evaluation that enters teacher grades. b. Shorthand II Achievement is the grade given by the teacher to the specific pupil for second semester accom- plishment in shorthand. The grade may include whatever the teacher may consider important-—speed of reading, knowledge of theory, speed of taking dictation, and general subjective evaluation that enters teacher grades. 0. Shorthand achievement and shorthand success are used interchangeably in this study. 3. Grade Point Average: Grade Point Average is a score computed by assigning five points to each grade of A, 18Farquhar, op. cit., p. 3. 11 four points to each grade of B, three points to each grade of C, two points to each grade of D, and one point to each grade of E or F received in each academic subject; i.e., each subject requiring homework taken during the ninth and tenth grades. These points were summed and divided by the number of academic courses taken during the ninth and tenth grades. The quotient thus derived was carried two placed beyond the decimal point in order to provide precision in assigning categories in the computation of the regression equation. 4. ACE Test: The American Council on Education Psychological Examination for high school students (ACE) is a test of mental ability which contains four tests-- same-opposite, completion, arithmetic reasoning, and number series.19 Results are subdivided into Linguistic and Quantitative subscores primarily aimed at deter- mining mental ability to do reasoning in a verbal setting.20 5. M—Scales: Four functioning measures (The Generalized Situational Choidelnventory, Preferred Job Characteristics Scale, Word Rating List, and Human Trait Inventory), which were designed to measure academic motivation, were combined and labeled the Michigan M-Scales.21 The scales were 19Burros 4th Yearbook of Mental Measurements (Highland Park, New Jersey: The Gyphon Press, 1949), p. 218. 2OBurros 5th Yearbook of Mentacheasurements (Highland Park, New Jersey: The Gyphon Press, 1959), p. 309. 21Farquhar, 0p. cit., p. 96. 12 authored by Dr. W. W. Farquhar and his associates of Michigan State University College of Education in January, 1963. 6. Gregg Diamond Jubilee Shorthand System: Gregg Diamond Jubilee Shorthand System is a revised edition, pub- lished in 1963, of the shorthand system created by the late John Robert Gregg. 7. Gregg Shorthand Simplified: Gregg Shorthand Simplified is a revised edition, published in 1950, of the shorthand system created by the late John Robert Gregg. 8. Predictor Variables: Predictor Variables are the factors that were introduced into the experimental situation.22 These variables consisted of (1) grade point average computed from ninth and tenth grades in academic courses, (2) English grades in the ninth and tenth grades, (3) an estimate of mental ability as measured by the American Council on Education Psychological Examination for high school students, (4) Linguistic subscore of the estimate of mental ability, (5) Quantitative subscore of the estimate of mental ability, and (6) motivation as measured by the Michigan M-Scales. Predictor variables and independent variables are used synonymously in this study. 9. Dependent Variable: "Dependent variable is the 22Allen L. Edwards, Statistical Methods for the Behavioral Sciences (New York: Holt, Rinehart and Winston, 1961) . pp- 8-9~ 13 variable for which we observe changes."23 The dependent variable was the achievement in Gregg Shorthand as measured by the grades assigned each individual by the teacher at the end of the first semester and at the end of the second semester of Gregg Shorthand. Teacher grades were validated by a uniform shorthand test created by the researcher. The tests were administered at the end of the first semester and at the end of the second semester. The first semester test was administered to shorthand classes in three partici- pating schools and the second semester test was administered to shorthand classes in all eight participating schools. CHAPTER II REVIEW OF LITERATURE A review of DISSERTATION ABSTRACTS, BUSINESS EDUCATION INDEX, and EDUCATION INDEX was made; also articles appearing in the professional journals and magazines concerning research studies were examined. From this review, a bibliography was constructed; and theses and dissertations that were not in the Michigan State University Microfilm Library or Main Library were requested on interlibrary loan. A number of universities have adopted the policy that no thesis, master's or doctora1,is now available on interlibrary loan although such studies were formerly available through interlibrary loan. Some universities do not microfilm dissertations. In spite of the unavailability of certain pieces of research, secondary sources such as the Anderson Study,1 the Frink Study,2 and the Byers Study3 provided excellent information to make a review of 1Ruth Irene Anderson, "An Analysis and Classification of Research in Shorthand Transcription" (unpublished Doctoral dissertation, Indiana University, 1946). 2Inez Frink, "A Comprehensive Analysis and Synthesis of Research and Thought Pertaining to Shorthand and Transcription 1946-1957” (unpublished Doctoral dissertation, Indiana University, 1961). 3Edward E. Byers, "Construction of Tests Predictive of Success in First Year Shorthand" (unpublished Doctoral dissertation, Boston University, 1959). 14 l5 literature in shorthand prognosis possible for this study. This chapter is divided into two sections: Section I is a review of literature on Shorthand Prognosis, and Section II is a review of literature on Academic Motivation. Section I is subdivided into two parts: Part I is a review of specific studies in shorthand concerned with prognosis; Part II is a synthesis of some of the research on shorthand prognosis. This synthesis is from primary and secondary sources and is divided into two parts: (1) research concerning mental ability as a predictive factor and (2) research concerning school grades as possible predictors. Tables are used to supplement this part of the review. In Section II, which deals with academic motivation, the Farquhar Studyu on academic motivation was especially helpful for the reason that it reviewed much of the literature on motivation. The Farquhar Study, therefore, will be reviewed in detail in Section II of this chapter. “William W. Farquhar, Motivation Factors Related to Academic Achievement, a Final Report of Cooperative Research Project No. 846 entitled ”A Comprehensive Study of Motivational Factors Underlying Achievement of Eleventh Grade High School Students” supported by the U.S. Office of Education in Cooperation with Michigan State University, 1963. 16 SECTION I Review of Literature on Shorthand Prognosis The review of literature on shorthand prognosis revealed a number of studies that dealt with shorthand prognosis; how- ever, three were significantly helpful in making possible a review of literature on shorthand prognosis. The first two studies were the Anderson Study,5 and the Frink Study6 which were particularly helpful for the reason that they were reviews of the literature and research con- cerning Gregg Shorthand. The Anderson Study,7 which will be reviewed in detail later in this chapter, reviewed shorthand literature and research from the beginning of.shorthand literature and research in the United States to 1946. The Frink Study,8 also to be reviewed in detail later in this chapter, reviewed shorthand literature and research from 1946 to 1957. The third study was the Byers Study9 which was helpful for the reason that it reviewed shorthand prognostic tests. The Byers Study10 will also be reviewed in detail later 5Anderson, 0p. cit. 6Frink, op. cit. 7Anderson, 0p. cit. 8Frink, op. cit. 9Byers, 0p. cit. loIbid. 17 in this chapter. Other dissertations and theses that were found helpful will be reviewed in lesser detail later in this chapter after the Anderson, Frink, and Byers Studies have been reviewed. Review of Specific Studies Relating to Shorthand Prognosis: The Anderson Study:11 The Anderson Study, completed in 1946, was an analysis and classification of all available literature and research in shorthand and transcription through 1945--a total of 298 research studies and articles in professional journals. The study was restricted to the topic of Gregg Shorthand since its introduction into the United States. One of the classifications in the Anderson Study was Prognosis. The study reviewed 44 research studies dealing with prognosis in shorthand and transcription. An abstract of each research project was made. The abstract contained a statement of the problem, purpose of the study, procedures, and findings. Following the abstract, Anderson commented on selected research projects. The comments point out conflicts in the reporting of the data, implications of the findings, or lack of information in the original report. llRuth Irene Anderson, "An Analysis and Classification of Research in Shorthand Transcription" (unpublished Doctoral dissertation, Indiana University, 1946). l8 Summarization of findings from the prognosis section of the Anderson Study are listed below: 1. Three investigators reported that the correlation between intelligence scores and short- hand marks was very low. Other persons reported that there was some relationship between the two factors. In an investigation by Dodson, the best pre- dictive measure used was the Terman Test of Mental Ability. The correlation between the scores on this test and marks in advanced shorthand was .71. 2. Phillips found that only two persons with intelligence quotients below 90 were doing passing work in transcription. None of the persons who failed or who drOpped shorthand had intelligence quotients below (pip) 120.* 3. Mandell concluded that a high I.Q. was indicative of probable success in shorthand, but a low I.Q. did not necessarily preclude success. 4. Several researchers have concluded that intelligence tests and shorthand marks did not measure the same abilities but that the intelli- gence quotient might be useful in a battery of predictive tests. 5. Billeter found that the selection of candidates for stenographic training in the army on the basis of intelligence and clerical aptitude reduced the number of failures 25 per cent. 6. In four studies, the Hoke Prognostic Test of Stenographic Ability was found to be of prac- tically no value. 7. In the investigations by Wood and Kessinger, the correlations between the scores on the Hoke Prognostic Test of Stenographic Ability and Shorthand marks were .464 and .559 respectively. *This researcher questioned the word "below." In the review of the Phillips thesis by Anderson, page 707-708, the word ”above" appeared; thus, the researcher assumes that the word "below" in the summary should have been "above.” 19 8. Whitley reported that the Hoke Prognostic Test of Stenographic Ability was the only test in the battery used in the study which differentiated in advance between those pupils who completed the shorthand course and those who drOpped shorthand. 9. Davis and Dodson reported correlations of .525 and .79 between the scores on the Turse Short- hand Aptitude Test and shorthand marks. Cruzan found the test of little value for the prediction of success or failure in shorthand. In a recent study by Phillips, it was found that 66 per cent of the students doing superior work in transcription made high scores on the Turse Shorthand Aptitude Test. 10. Tuckman concluded that the high correla- tion between the scores on the Turse Shorthand Aptitude Test and the American Council on Education Psychological Examination for high school students indicated that the use of both tests in the same battery might be questionable. 11. In five studies, correlations from .425 to .64 between marks in English and marks in short- hand were reported. 12. Four researchers found correlations between scores on English usage tests and shorthand marks ranging from .53 to .707. 13. In a study by Cruzan, it was reported that 84 per cent of the students who receive "A" in shorthand scored above the 50 percentile on the COOperative English Test while a large portion of pupils with scores below the 50 percentile on the Cooperative English Test failed in shorthand. 14. The correlations obtained between the average of previous scholastic averages and short- hand marks have ranged from .594 In) .70. Four investigators concluded that the previous scho- lastic average of the pupil was one of the best bases for prediction of success or failure in shorthand. 15. Correlations obtained between vocabulary test scores and shorthand marks ranged from .40 to .55. 20 16. In four studies, correlations from .55 to .759 between foreign language marks and shorthand marks have been reported. 17. Davis reported a low correlation between personality ratings and shorthand marks. Whitmore found personality ratings to be one of the best predictive measures of success in shorthand. 18. Other predictive measures which have been used with little or no success include speed of writing tests, memory tests, the Monroe Reading Comprehension Test, the Minnesota Vocational Test for Clerical Workers. Phillips reported that, of the students doing passing work in transcription, only 5 per cent were rated below average on the Detroit Clerical Aptitude Test. 19. Investigators have generally agreed that a battery of tests is more effective as a basis for the prediction of success or failure in shorthand than a single measure. The correlations between scores on batteries and marks in shorthand have ranged from .61 to .76.12 The Frink Study:13 The Frink Study, completed in 1961, was an analysis and synthesis of research and thought per- taining to shorthand transcription. The Frink Study resembled the Anderson Study in that it was a review of literature and research in shorthand. The period of time covered by the Frink Study was from January 1, 1946, where the Anderson Study stopped, through December 31, 1956. A total of 258 references in professional journals were used to supplement 117 research projects. An abstract was prepared for each. lzlEli-y PP- 733-737- 13Inez Frink, ”A Comprehensive Analysis and Synthesis of Research and Thought Pertaining to Shorthand and Transcription 1946-1957" (unpublished Doctoral dissertation, Indiana University, 1961). 21 Following selected abstracts, the author commented on each research project. The comments pointed out conflicts in the reporting of the data, implications of the findings, or lack of information in the original report. A summary of the findings of the Frink Study is listed below: 1. Low level ability of students, which may be interpreted as a need for guidance and selection-- despite the fact that research reveals that drOpouts and failures can be reduced by guidance and selection, apparently few schools practice either. 2. Need for improvement of instruction--there is some evidence that considerable improvement could be made in the quality of classroom instruction. Factors which seemingly influence the quality of instruction are confusion of objectives; failure to recognize the level of instruction at which certain techniques, practices, or procedures should be introduced or discontinued; need for greater empha- sis on shorthand theory and the related learning; and need for less emphasis on dictation speeds and adjustment techniques. 3. Need for increased instruction time-- although it is believed that additional instruction time is needed, studies of the relationship of increased instruction time to achievement in stenography are inconclusive. 4. Need for nonsymbol shorthan --there is some evidence that nonsymbol shorthand can be learned in less time than traditional shorthand and that it has a definite potential for vocational use.1 The Byers Study:15 The Byers Study, completed in 1958, was a research project designed to construct a predictive test 14Ibid., National Business Education Quarterly, XXXI 15Edward E. Byers, "Construction of Tests Predictive of Success in First-Year Shorthand” (unpublished Doctoral dissertation, Boston University, 1958). 22 for success in first—year shorthand. As part of the research for the construction of the test, Byers analyzed previous tests that were designed to predict success in shorthand. He analyzed six different published tests. The tests analyzed by Byers were the Hoke Prognostic Test, the Bennett Stenographic Aptitude Test, the Turse Shorthand Aptitude Test, the Educational Research Corporation Stenographic Aptitude Test, the Detroit Clerical Aptitude Examination, and the Vocational Aptitude Test for Shorthand Students. Byers' review of the tests is, in part, as follows: The Hoke Prognostic Test was created to provide the teacher with pupil data in order to tailor the shorthand course to the student. The test consisted of seven subtests which are as follows: (1) motor reaction; (2) speed of writing; (3) quality of writing; (4) speed of reading; (5) memory; (6) spelling; and (7) symbols. The test was not success- ful in predicting success or failure of shorthand students; thus, it was declared ozt of print in 1948 by the Gregg Publishing Company.1 The Bennett Stenographic Aptitude Test was created to predict ability to acquire the skills of shorthand and typewriting. The test consisted of two subtests--transcription and spelling. Empirical studies have reported that the test is of little value for the reason that more information is needed to predict achievement in the areas of shorthand and typewriting.17 The Turse Shorthand Aptitude Test was an attempt to find a valid method to eliminate unquali- fied pupils from shorthand courses before they made an effort to undertake work for which they were not suited. The test consisted of seven subtests which are as follows: (1) stroking; (2) spelling; (3) 16Ibid., pp. 10-12. l71bid., p. 20. 23 phonetic association; (4) symbol transcription; (5) word discrimination; (6) dictation; and (7) word sense. Recent research findings are quite con- sistent on the Turse Test with correlagions being reported consistently of .60 and .65.1 The Educational Research Corporation Steno- graphic Aptitude Test was constructed by Walter L. Deemer, Jr., to measure the student's ability to learn to write shorthand and to transcribe it. The test consisted of five subtests which are as follows: (1) speed of writing; (2) word discrimi- nation; (3) phonetic spelling; (4) vocabulary; and (5) sentence dictation. It is evident that second-year shorthand accomplishment measures predominate in the deter- mination of the criterion measure for each testee. Since the functional factors, as measured by the subtests, include those prevailing for students of advanced shorthand, this choice of composite is probably justified. It is interesting to note that in this respect both Deemer and Turse are in com- plete agreement--both attempt to predict success in transcription rather than success in first-year shorthand.l The Detroit Clerical Aptitude Examination was designed to identify pupils who have suitable abilities for high school business courses. The examination consisted of eight subtests which are as follows: (1) rate and quality of handwriting; (2) rate and accuracy in checking; (3) knowledge of simple arithmetic; (4) motor speed and accuracy; (5) knowledge of simple commercial terms; (6) visual imagery; (7) rate and accuracy in classifi- cation; and (8) alphabetical filing. Their value as predictors of success in shorthand and type- writing is questionable.20 The Vocational Aptitude Tests for Shorthand Students consisted of seven subtests designed to measure (1) ability to distinguish between vowel sounds and to indicate them by a particular 18Ibid., pp. 24—26. 19Ibid., pp. 3u-36. 20 Ibid., pp. 38-40. 24 notation; (2) substitution powers of the candidates; (3) retention powers and ability of the testee to form geometrical characters quickly and accurately; (4) ability to write a passage of ordinary dictation at a rapid rate and with correct spelling. The tests were devised for use in the military and reported studies showed good results.2 The Test of Shorthand Aptitude created in the Byers dissertation consisted of five subtests which were as follows: (1) phonetic perception; (2) retention ability; (3) observation aptitude; (4) pattern from parts; and (5) hand dexterity. The subtests correlated with the criterion measure as follows: .36; .44; .18; .28; and .68, respectively. The multiple correlation was reported as .62 for high school shorthand classes. Results of the Byers investigation indicate that the test results have a signifi- cant relationship with first-year shorthand achievement; how- ever, difficulty of scoring reduces the value of the test. In addition to the Anderson, Frink, and Byers Studies, the following studies were of interest in the area of prognosis. The Pauk Study:22 Pauk made use of the American Council on Education Psychological Examination which produces the Linguistic scores. He compared the correlations found from the Linguistic score and the Turse Test. Pauk found that 21Ibid., pp. 43-44. 22Walter Pauk, "What's the Best Way to Predict Success in Shorthand?" Business Education World, LXIII (April, 1963), pp- 7-8: 34. 25 "The L-Score of the ACE predicted shorthand success as well as the total Turse Test (.63 versus .63)." The Sanders Study:23 Sanders reported in a study of college students in 1961 that " a significant relationship was found between the ACE percentile rank and achievement in college shorthand grades." The Veon Study:2LI Veon reported in a study of 299 elementary shorthand students at George Washington University in 1948 that the completion section of the ACE correlated .4208 with a complete shorthand test. Other correlations to the complete shorthand test reported by the Veon Study were: Same-Opposite--.4923; Verbal Analogies--.5685. The corre- lation between the subtest, verbal analogies, and shorthand reading, .7129; between the subtest, verbal analogies, and shorthand transcription, 6313. Multiple correlation between the Linguistic sections of the ACE Test and complete short- hand test was not reported. The correlations found in the Veon Study were high but 23Ce1ene Honeycutt Sanders, "A Study of the Relationship Between Certain Redford College Students' ACE Scores, Years of Shorthand in High School, and Achievement in Shorthand" National Business Education Quarterly, XXXI (October, 1962). 2“Dorothy Veon, The Relationship of Learning Factors Found in Certain Modern Foreign-Language Aptitude Tests to the Prediction of Shorthand Achievement in Collegp. Delta Pi Epsilon Research Award Series. Stillwater, Oklahoma: Oklahoma Agricultural and Mechanical College, 1950. 26 still not high enough to make the ACE Tests useful indi- vidually or collectively in predicting shorthand achievement. The Powell Study:25 The Powell Study, in 1961, of shorthand enrollments in Ottawa Township High School, Illinois, attempted to determine if in a two-year study significant differences did exist between shorthand dropouts- failures and continuants. The report analyzed national test scores, personality factor ratings by teachers, English grades, foreign language grades, typewriting grades, short- hand grades, attendance, part-time work activities of the students, educational and vocational plans, reasons students gave for dropping shorthand, and teacher Opinions of why students dropped shorthand. The study concluded that several differences do exist between shorthand dropouts and continuants as evidenced by the testing program of a particular school. The scores received by the continuants were higher than those of the drOpouts in each case. The areas of difference in the test- ing program are as follows: a) Of all the tests, the Science Research Asso- ciates Reading Record was the most signifi- cantly different. The drOpouts-failures received a significantly lower score here than did the continuants. 25Georgia Faye Powell, "An Analysis of Shorthand Dropouts at Ottawa Township High School” (unpublished Master's thesis, Illinois State Normal University, 1961). b) d) e) f) 27 The total score from the Reading Record has the greatest degree of statistical significance of any of the test scores used in the study with the dropouts—discontinuants receiving the lower scores. Other Reading Record test scores of marked statistical significance were the sentence meaning score and the vocabulary score. The national test score from the Science Research Associates Primary Mental Abilities Tests having the greatest statistical significance for short— hand success was the verbal meaning score. The correctness in writing score on the Iowa Tests of Educational DevelOpment was the one found to have the greatest statistical signifi- cance from that group of tests. Eight of the fourteen dropout means were below the national means, while twelve of the four- teen co tinuant means were above the national means.2 Thus, it is evident from such findings that the national testing program of this particular school reveals certain differences in existence between continuants and failures and drOpouts. The Powell Study confirms previous research findings which illustrate the relationship between verbal abilities and shorthand achievement. A second conclusion of the Powell Study was: . . Students continuing shorthand tend to rate higher than the dropouts on all personality factors used in the study. The three factors of greatest significappe were industry, initiative, and respon- sibility. 26 Ibid., p. 78. 27Ibid., p. 79. 28 Teachers were asked to rate the student on a five-point scale. A comparison of the dropouts and continuants showed a difference in means for Industry, Initiative, and Responsibi- lity of .550, .500, and .500 respectively; however, Serious- ness of Purpose and Emotional Stability each showed differences in means of .450.28 The continuants received the favorable rating in each comparison. A third conclusion of the Powell Study was: The average grades for the continuants for the courses analyzed were: English, 2.39; foreign language, 2.39; typewriting, 2.56; shorthand, 2.74. The average grades for the dropouts for the courses analyzed were: English, 1.94; foreign language, 1.87; typewriting, 1.63; shorthand, 1.42.29 It is apparent that grades tend to be higher for the continuants than for the discontinuants. A list of frequent reasons for drOpping shorthand as checked by forty-five students was compiled in the study and is reproduced in Table II on page 123 of the appendix. This compilation illustrates student reasons for discontinuing shorthand class. Analysis of the list shows that classwork-—the students lack of effort and/or desire to learn the subject matter and quality of performance--is the most frequently given reason 28Ibid., p. 48. 291bid., pp. 79-80. 29 for dropping shorthand. The other reasons given could be grouped into a classification of a change in vocational plans. The Osborne Study:30 The purpose of the study, com- pleted in 1943, was to determine the value of certain psychological tests in predicting success of secondary school students in the study of shorthand. The psychological tests administered for the project were the Otis Self-Administering Test of Mental Ability, Iowa Silent Reading Test-~New Edition, I. E. R. Clerical Test Form C-l, Revised Minnesota Paper Formboard, Shorthand Learning Test--Semester I, and Gates Visual Perception Test. The Carmichael Shorthand Learning Test was used for the criterion of shorthand achievement. Correlations were computed among all variables. The study concluded that none of the correlations between the score on the shorthand criterion and scores on single tests or between the score on the criterion and scores on combinations of tests was high enough to make prediction valuable. 30Agnes E. Osborne, "The Relationship Between Certain Psychological Tests and Shorthand Achievement"(New York: Bureau of Publications, Teachers College, Columbia University, 1943). 30 The Dodson Study:31 The study, completed in 1943, was an attempt to determine what means are available for pre- dicting success in the study of shorthand and typewriting and to evaluate these means in the light of experiments conducted. Coefficients of correlation were computed between the variables which were marks in high school subjects, mental ability, shorthand and typewriting grades, and the grades received in shorthand and typewriting in college. Eighty subjects in the study came from three high schools. The Hutson Study:32 The study, completed in 1951, determined the relationship between the Germane Vocabulary Test, a teacher-prepared spelling test, the Otis Quick- Scoring Mental Ability Test, the Lee-Thorpe Occupational Interest Test, the Educational Research Corporation Stenographic Aptitude Test, and achievement in first year of high school instruction in Gregg Shorthand. Beginning shorthand classes in five eastern Tennessee high schools were used for the study. Semester achievement tests and final achievement records were correlated with the predictors. 31Mary H. Dodson, "A Study of Typewriting and Shorthand Prognosis” (unpublished Master's thesis, University of Kentucky, 1943). 32Billy T. Hutson, ”Prognosis of Achievement in First- Year Gregg Shorthand Simplified" (unpublished Master's thesis, University of Tennessee, 1951). 31 The study concluded that none of the correlations were considered to be of value in prognosis. The Goodenow Study:33 The purpose of the study, com— pleted in 1948, was to review the thought and practices as eXpressed in the literature and of the findings and con- clusions of research pertaining to stenographic prognosis. The data was collected from literature pertaining to prognosis which included books, professional articles, theses, and prognostic tests. The findings were divided into two parts--present status of stenographic prognosis and the future of steno- graphic prognosis. Synthesis of Research Relating to Shorthand Prognosis: Reported Research Using Mental Ability: The literature showed that general mental ability was frequently used in studies of shorthand prognosis. A partial summary of studies reporting correlations of mental ability to shorthand achieve- ment is compiled in Table III, Appendix, page 124. The table, compiled by this researcher from the studies reviewed, shows correlations reported, year, level, and criteria used for determining shorthand achievement. 33Jean E. Goodenow, "A Review of Professional Literature Relating to Stenographic Prognosis" (unpublished Master's thesis, State University of Iowa, 1948). 32 The range of correlations between mental ability and shorthand achievement is from a .218 reported by the Sherman Study to a .75 reported by the Dodson Study. These corre- lations cannot be compared for the reason that a variety of mental ability tests was used and a variety of criteria for shorthand achievement was used. The median correlation in Table III, Appendix, page 124, is .36 reported by the Beal Study of 1929. The researchers reported in the original studies that intelligence tended to predict failure better than it predicted average or superior achievement.34 Frink pointed out: . . . There was general agreement that intelli- gence is not a good predictor of success in beginning shorthand although it is as closely related to shorthand as to most academic subjects.3 Even though there is great variance in the use of intelligence as a predictor of shorthand achievement, there is consensus that intelligence criterion forecasts shorthand achievement as efficiently as it forecasts achievement in other academic subjects. Reported Research Using_English Grades: English grades and English tests of various types were frequently used criteria for predicting achievement in shorthand. The investi- gations involving such criteria have produced a variety of 3”Frink, 0p. cit., p. 34. 35Ibid. 33 findings. A review of the literature indicates that English grades were defined in different ways. Some investigations defined English grades as all courses in English taken prior to enrolling in shorthand while others used a specific course or year. This difference in definition of English grades was in part responsible for the variation in findings. Table IV, Appendix, page 128, is a partial summary of findings of studies as reported by researchers of studies that used English grades and/or tests as one of the variables. This table, also compiled by the researcher from the studies reviewed, shows correlations reported, year, level, and criteria used for determining shorthand achievement. The 14 studies reported 23 correlations with English grades and/or tests. The studies reported four correla- tions of .70 or higher; five correlations of .60 to .69; six correlations of .50 to .59; three correlations of .40 to .49; two correlations of .30 to .39; two correlations of .20 to .29; and one correlation of .01 was reported by the Mandell study. The .01 correlation was the relationship between grammar grades and shorthand grades. The correlations listed in Table IV, Appendix, page 128, were .50 or higher in 15 of the 23 correlations. It is apparent from this review of studies using English grades and/or tests that achievement in English is one of the better factors in predicting shorthand achievement. 34 Reported Research Using Overall Grade Point Average: Grade Point Average has been studied for its potential in predicting shorthand achievement. Table V, Appendix, page 132, is a partial summary of findings reported by research projects that used grade point average as one of the variables in predicting shorthand achievement. The table shows the correlations of six research studies, the year each study was completed, the level of the investi- gation,and the criteria used to predict shorthand achieve- ment. The lowest correlation reported was .381 in the Hutson Study. The highest correlation was found in the Kortendick Study which reported a correlation of .7947 between grade point average and first semester shorthand grades. Of the 10 correlations reported in the studies of Table V, Appendix, page 132, five were .70 or higher. Even though there is disagreement among researchers as to the degree of value of grade point average in predicting achievement in shorthand, it is apparent from the preceding table that grade point average is of value in forecasting shorthand grades. Summarypof Shorthand Prognosis: A review of the literature on predicting achievement in shorthand indicates that there is a definite need for a means of forecasting shorthand achievement. The high rate . .‘lq - l. 1‘. s an- ‘ ' -'~—- ‘3‘. a "n: y A- r. 35 of failure is a waste of time, effort, and money. Researchers have not been able to produce a means of pre- dicting achievement even though a large number of investi— gations have been done in shorthand prognosis. Some of the Inedictors of shorthand ability that have been studied are summarized below: 1. General Mental Ability tests are commonly used criteria for predicting shorthand achievement. Research shows, however, that mental ability does no better in pre- dicting shorthand achievement than in predicting achieve— ment in other academic subjects. 2. Special Aptitude Tests have been only partially Successful as predictors of shorthand success. The Turse shorthand Aptitude Test is the most successful; however, research indicates there are other tests (the ACE Ling- uistic score, for example) which will do as well as the TuI‘se and require less time to administer. 3. English Grades have frequently been used in studies IRD'Predict shorthand success. The studies have concluded Inuit English grades are of value in predicting shorthand achievement. 4. Grade Point Average has frequently been used by researchers who have agreed that it is of value in pre— dieting shorthand achievement, but they have disagreed as 36 to the degree of value. SECTION II Review of Literature on Academic Motivation Motivation has been studied by a number of researchers and scholars36’37’38’39 as is evident by a review of literature and indices. None of the discoveries has pro- duced a usable measure of motivation--one that could be utilized by a classroom teacher and interpreted by the classroom teacher--until the objective test created by Dr. W. W. Farquhar and his associatesLIO in 1963. Review of a Specific Study Relating to Academic Motivation: *The Farquhar Study: Farquhar and associatesul polarized the McClelland TheoryLI2 which was that achievement motivation is composed of (l) long—term involvement, 36Dalbir Bindra, Motivation: A Systematic Reinterpreta- tion (New York: The Ronald Press Company, 1959). 37Jerome Kagan, and Gerald S. Lesser, Contemporary Issues in Thematic Apperceptive Methods (Springfield, Illinois: Charles C. Thomas, Publisher, 1961). 38K. B. Madsen, Theories of Motivation (COpenhagen, Denmark: Munksgaard, 1959). 39David C. McClelland, et al. The Achievement Motive (New York: Appleton-Century-Crofts, Inc., 19537: quarquhar, 0p. cit. ”libid. 42 McClelland, op. cit. 37 (2) unique accomplishment, and (3) competition with a standard of excellence. Their assumption was that there existed a continuum of achievement motivation. At one pole of the continuum would be found (1) short-term involvement, (2) common accomplishment, and (3) competition with a minimal standard of excellence. Along the continuum they assumed that highly motivated students would be found at the tOp with less motivated students falling somewhere below that level but above failure. By utilizing an aptitude base, Farquhar and associates identified those students who exceeded an aptitude based eXpectation of academic performance as overachievers and those students falling below expectation as underachievers. By dividing the students into male and female overachievers and under- achievers and placing them on a regression diagram, the population was divided into subgroups of achievers as illustrated in Figure 1 (page 38). Farquhar and associates theorized that significant and identifiable differences could be found in student responses and attitudes of over- and underachievers. After identifying the over- and underachiever, six general areas of investigation were identified and researched. The areas were (1) Word Rating List--"Looking- glass-self;" (2) Preferred Teacher Characteristics Scale-- measure of non-intellectual personality factors important to academic success; (3) Human Trait Inventory--measure to 38 FIGURE I REGRESSION EQUATION DIAGRAM SHOWNNG OVER- AND UNDERACHIEVERSa overachievers Achievement Measure underachievers Aptitude Measure aFarquhar, op. cit., p. 12. 39 differentiate between discrepant achievers; (4) Perceived Parental Attitudes Inventory--measure of student views of his parents' child rearing practices; (5) Generalized Situational Choice Inventory--measurement of subject's need for long-term involvement, unique accomplishment, and competition with a standard of excellence; (6) Preferred Job Characteristics Scale-—measurement of extremes in occupational motivation. After factor analysis and validating and cross—vali- dating the test items, 136 questions remained in the female form. The Preferred Teacher Characteristics Scale and Perceived Parental Attitudes Inventory did not withstand the factor analysis and cross validation process; therefore, these two measures were dropped from the project. The Word Rating List:LI3 The Word Rating List measures the student's perception of how her teachers perceive her. Four factors met the requirement of the research. The factors are (1) academician; (2) emotionality; (3) con- formity; and (4) unreliability. The academician factors are items in which the highly motivated girl believes that the teacher views her as "bright, effective, goal oriented, and amenable to learning'ILILI 43Farquhar, op. cit., p. 135. qubid., p. 138. 40 The emotionality factors are items in which the "under- achieving student sees herself as distracted, impatient, bold, and resistant."LL5 The conformity factors are items in which the "achieving student sees herself meeting the expectancies of the school role."LI6 The unreliability factors are items in which the ”underachieving student sees herself as having a tendency 1+7 not to meet the expectancies of the school." The Human Trait Inventory:48 The Human Trait Inventory measures personal characteristics of the individual. Five female factors met the requirements of the research. The factors are (1) academic compulsivity; (2) fantasy; (3) con- servatism; (4) social distance; and (5) emotional instability. Academic compulsivity "describes high motivated dimen- sions related to high organizational needs which are fulfilled in school activities.“+9 I The second factor was labeled fantasy and is "related to withdrawing into a non-academic dream world which lacks stress uSIbid. 46Ibid. ”71bid. ”81bid., p. 137. ”9Ibid., pp. 146-147. 41 norms."5O This factor describes the low motivated person. The third factor, conservatism, was interpreted as measuring high motivation. The factor ”is related to a strain towards consistency in conforming to established norms, Whether they are her own or others."51 The fourth factor, social distance, was interpreted as measuring low motivation. The factor "is related to lack of desire for or understanding of social interaction beyond a superficial level.”52 Generalized Situational Choice Inventoryx53 The Generalized Situational Choice Inventory measures what the individual would like to do in a given situation. Five female factors met the requirements of the research. The factors are (1) long—term versus short-term educational involvement; (2) high versus low task commitment; (3) unique versus common accomplishment; (4) long-term versus short-term involvement; and (5) independence versus dependence in problem solving. The long-term versus short—term educational involvement factor consists of items describing highly motivated girls 5O;pgg., pp. 147-148. 5112l§° 5212;2- 53;pgg., p. 146. 42 desirous of "further education with a delay in immediate rewards as Opposed to the low motivated female whose responses were typified by desire for immediate reward with little extended involvement in education."5LI The high versus low task commitment factor was the second factor in the analysis. It is described as follows: The high motivated individual chooses the responses directed towards high effort with intellectual pro- ductivity, the low motivated individual chooses the responses which demand little ippediate effort but might produce monetary returns. The third factor was labeled unique versus common accomplishment. It is described as follows: The high motivated female chooses responses con- cerned with doing outstanding tasks, the low motivaged female wants to do things like everyone else.5 The fourth factor was labeled long-term versus short- term involvement. It is described as follows: The high motivated female chooses responses indi- cating a tolerance for waiting for rewards, the low motivated female wants immediate and certain rewards. The fifth factor was labeled independence versus dependence in problem solving. It is described as follows: The high motivated female chooses responses which 5”Ibid., p. 152. 55Ibi 561bid., p. 158. p. 57Ibid. 43 indicate a reliance on self for solving problems; the low motivated female looks to other's support.5 Preferred Job Characteristics Scale:59 The Preferred Job Characteristics Scale measures vocational choices after the completion of education. Three female factors met the requirements of the research. The three factors were (1) unique accomplishment versus avoidance of education; (2) long-term versus short—term job involvement; and (3) high versus low job interests. The first factor was labeled unique accomplishment versus avoidance of education. It is described as follows: The high motivated individual solves problems no one else can thereby gains respect at job tasks to which she is committed. The low motivated individual avoids personal decisions and ties wgth a job lacking demand of college education. 0 The second factor was labeled long—term versus short- term job involvement. It is described as follows: The high motivated individual wants a job requiring extended commitment. The low motivated individual wants little personal involvement with the job, but wants the security of knowing that job will be there for her when she wants it. 1 The third factor was labeled high versus low job interests. It is described as follows: 581bid. 59Ibid., p. 156. 60Ibid., p. 168, 61Ibid. 44 The high motivated individual wants job which absorbs interests. The low motivated individual wants job with few demands, put one that will be available when it is needed. 2 Multiple regression equations were constructed to determine the grade point average from the four subtests, total test score.and the Differential Aptitude Test of Verbal Reasoning which served as an aptitude base for the study. The multiple regression equation produced a corre- lation of .64 and .63 for females after cross validation. The study concluded that: 1. High and low motivated eleventh grade students do respond with significant differences to a number of items designed to measure reflected self-concept personality traits, and preference for certain types of occupational characteristics. 2. Scores based on the valid items (labeled the M-Scales) estimate grade point average better than chance. 3. When the scores of either the sub- or total test are added to the DAT-VR the precision of estimating the grade point criterion is signifi- cantly increased. 3 From these conclusions, an objective test was devised to measure factors related to academic achievement. Implications of Related Literature for the Present Study: The literature on prognosis demonstrates that English grades, grade point average, and mental ability, in order, Ibid. 63Ibid., p. 179. 45 are the three most reliable predictors of shorthand achieve- ment. Unfortunately, the correlations are not sufficient to warrant any one item or combination of items to be used as predictors of achievement. The literature on drOpouts and failures reaffirms the findings of the studies on prognosis. The reports on failures and dropouts show that the persons who drop out or fail shorthand differ significantly in several ways from those who are successful. The personality factors of significant differences are industry, initiative, and responsibility. Previous grades and scores on standard tests show there are significant differences with the achievers receiving more favorable scores each time. The Farquhar Study shows that some of the variance between aptitude and actual achievement as indicated by grade point average can be measured to increase the pre- dictability of achievement. The factors of the sub-tests of the Farquhar Study include nearly all the elements of differences found between shorthand achievers and failures. The postulate is made that the Michigan M-Scales when used in combination with an estimate of mental ability will increase the precision of prediction of achievement of eleventh grade girls in first semester and second semester of Gregg Shorthand. CHAPTER III PROCEDURES This chapter describes the procedures for the study and is divided into three parts: Part I contains the procedures used to collect the data; Part 11 contains the procedures used to process the data; and Part III contains the procedures used to statistically analyze the data. Part I Collection of Data Sources of Information: Schools: The information for this study came from two parochial high schools and six public senior high schools located in seven different cities and towns in central and southern Michigan. The participating schools had student enrollments ranging from 452 students to 1,619 students. These enrollment figures are presented in Table VI, Appendix, page 134. The school administrators were contacted in June and July, 1964, to ascertain their willingness to participate in the study; the administrators of each school were contacted in person by the researcher, then each was sent a letter (Appendix, page 122) explaining the experiment and 46 47 identifying the data that were needed. After permission was granted by the administrators, the shorthand teachers were contacted in August and September, 1964, to ascertain their willingness to participate in the project. If the teacher expressed willingness to participate in the study, arrange- ments were made to administer the tests to his students. Students: The subjects for this study were enrolled in beginning Gregg Shorthand classes in the eight partici- pating schools. The sample consisted of eleventh grade girls only who were present for the administration of the two tests, who had a complete transcript in the school office, and who completed the first and/or second semesters of Gregg Shorthand. No other criteria were used for selecting the sample pOpulation. This produced a sample of 177 subjects for the first semester and 171 subjects for the second semester. Nature of the Data Collected: Three types of data were gathered concerning each student: (1) an estimate of mental ability, (2) grades for ninth and tenth grade English and an overall academic grade point average, and (3) a motivation score. Estimate of Mental Ability: The instrument used to measure mental ability was the American Council on Education Psychological Examination for high school students, 48 1953 edition.1 The test contained four sections: sections one and two produced a linguistic score, and sections three and four produced a quantitative score. The total score, the sum of the four sections, represented the estimate of mental ability. The researcher administered the test between September 18, 1964, and October 22, 1964. Each section of the test was timed with a stOp watch according to the published directions. The subjects indicated their responses on special scoring sheets. Scoring of the examination was done by hand. The researcher first examined each answer sheet for multiple responses to a single item; the company—prepared key was then used to determine the number of correct items for each test. Each sheet was scored twice by the researcher and then double checked by a second person. Grades: To carry out the purposes of this investi- gation, it was necessary to obtain grades from the schools' permanent student records. The grades collected were the ninth grade English grades, tenth grade English grades, and grades from other academic subjects for the ninth and tenth grades. To provide a composite English grade for each student, an average was derived from the four semester grades each 1Burros, loc. cit. 49 student had in English. Grade Point Average was determined from all academic courses each student had taken during the ninth and tenth grades. A list of the academic courses included in the grade point average is found in Table VII, Appendix, page 135, a list of the non—academic subjects not included in the grade point average is found in Table VIII, Appendix, page 136. Assignment of a point value was made to each letter grade. A value of 5 points was assigned to a grade of A, 4 points to B, 3 points to C, 2 points to D, and 1 point to E and F. The point values were summed and divided by the number of academic courses taken during the ninth and tenth grades. The number of academic courses ranged from 16 to 21. Shorthand grades were collected from the teachers at the end of the first and second semester. Motivation Score: The motivation score was obtained from the student's responses to the Michigan M-Scales, an experimental objective test of academic motivation devised by Dr. W. W. Farquhar and his associates. The scale is composed of four sub—tests: (l) Generalized Situational Choice Inventory (GSCI), (2) Preferred Job Characteristics Scale (PJCS), (3) Word Rating List (WRL), and (4) Human Trait Inventory (HTI). An abridged description of 5O Farquhar's tests as given by Kipfmueller2 follows. The Generalized Situational Choice Inventory was constructed to describe the academic motivation situation. Students are required to make a forced choice between two types of situations, one which depicts a high and one which depicts a low academic motivation situation. A high score on this scale indicates an individual who has a high need for academic achievement. A low score indicates an individual who chooses activities disassociated from the school's program. The Preferred Job Characteristics Scale was designed to differentiate between the job aspirations of high-low motivated students. Students are required to make a forced choice between two types of jobs, one which depicts high and one which depicts low job aspiration. Students who score high on the scale tend to want jobs where their individuality is recognized, where their talents and skills are used, and where Opportunities exist for change and advancement. Students who score low on this scale tend to prefer jobs where entry is easy, require— ments are low, time restrictions are absent, and where one is "discovered" rather than worked into a position. The Word Rating List was designed to measure the academic self—concept of the student. Students are asked to rate themselves on a series of descriptive phrases and words describing high and low motivated and achieving students. A high score indicates an individual with academic self-concept oriented towards the school environment. An individual who scores low on this scale has a fairly clear-cut picture that he is not academic. The Human Trait Inventory consists of items from past personality tests that have been found to differentiate between high and low achieving and motivated students. Students are asked to rate how they feel about these statements. Individuals who score high on this scale tend to have personalities 2Mark K. Kipfmueller, ”The Predictibility (pig) and Factored Dimensions of the M-Scales for Eleventh Grade Parochial School Students” (unpublished Doctoral disserta— tion, Michigan State University, 1963), pp. 19-21. 51 more similar to highly motivated students and those scoring low tend to have personalities similar to low motivated students. The researcher read the test directions to the students. After allowing time for questions from the students, the subjects were allowed to mark their responses at their own speed on a scoring sheet with machine scoring pencils. The test was administered between September 19, 1964, and November 10, 1964. Scoring of the M-Scales was done by the Office of Evaluation Services on the campus of Michigan State University. Part II Processing of Data Preparation: After the tests were scored and grades were collected, the data were placed on worksheets in preparation for key- punching. The worksheet was taken to the Computer Center on Michigan State University Campus where the data was placed on punch cards and verified. The cards were processed through the line printer which transferred the information on the cards to printout paper. From this printout, the data on the cards were checked against the original data to determine accuracy of the cards. No errors were found. 52 Processing: The data was processed by the Control Data Corporation 3600 Computer on Michigan State University Campus. The CORE3 program was used and provided the necessary correlations for this study. Part III Analysis of Data* Three statistical procedures were used in the analysis of the data: (1) estimate of reliability of motivation score; (2) mean test of significance; and (3) correlation analysis. Reliability Estimate: - The Farquhar Study used Hoyt's Analysis of Variance to obtain an estimate of reliability to measure internal con- sistency of the M-Scales. Since this study attempted to follow the Farquhar Study as closely as possible, Hoyt's procedure for reliability estimate was in the original design for the present study; however, as the data for the 3D. F. Kiel, and W. L. Ruble, ”Calculation of More Than One Regression From a Set of Data (CORE)" Michigan State University Agricultural Experiment Station #6, September 30, 1963. *Help with the statistical design and the use of the computer program was secured from Dr. W. W. Farquhar, Miss Jean Dyer,and Mr. Bruce Rogers. 53 present study were analyzed, it was found that the responses of the subjects to the M-Scales were scored in a right or wrong category. Lindquist points out that The analysis of variance approach, however, appears useful for obtaining reliability estimates from items or trials which are scored with a range 0 scores, and not merely as 'passed' or 'failed'. For the reason that the subjects' responses were in passed or failed form, and for the reason that the researcher found it more economically feasible, the decision was made to substitute Kuder-Richardson #20 formula for the Hoyt's Analysis of Variance procedure to obtain an estimate of inter- nal consistency of the M-Scales. Lindquist points out that It has been indicated that the result obtained by Hoyt's procedure is identical with that from Kuder- Richardson #20, so nothing new has been added so far as analysis of items of a test is concerned. The Kuder-Richardson #206 formula is 2 _ rtt = ( ) (St piqi) (n " 1) ( Stz ) where rtt = reliability of the total test n = number of items in the test st2 = variance of the total test pi = proportion passing item 1 4 E. F. Lindquist, Design and Analysis of Experiments in .2sxchOlQal_and_Educaii2r.1Boston: Houshton Mifflin. 1953). 591. 5Ibid. 6Ibid. 54 Mean Tests of Significance: The Student "t" was used to test the differences in uncorrelated means between the responses of the subjects in the present study and the responses of the subjects in the Farquhar Study to the Michigan M—Scales, and the difference in uncorrelated means between the grade point average of the subjects in the present study and the grade point average of the subjects in the Farquhar Study. It was assumed that the means were uncorrelated for the following reasons: there were no common factors between the population samples; and the samples were independent. The formula7 used is as follows: t = (Kl—X2) (s- _ - X1 X2 where t i the t ratio with n1 plus n2 minus 2 degrees of freedom (177 + 261 - 2 = 436). X1 = the mean of group 1 (population of the present study). X2 = the mean of group 2 (female population of the Farquhar Study). 3. - = the standard error of difference. X1"3‘2 In order to perform the t test, it was necessary to test the equality of the variance between the population in the present study and the female pOpulationcfi‘the Farquhar Study. The F (Fisher F) test was used to test the 7Edwards, op. cit., p. 254. \ . ud..&. ‘ a 'l J 55 equality of variance of the four sub-tests and the total M-Scales as well as the grade point average for the two pOpulations. The formula8 used is as follows: 2 F = 81 822 2 . . where S1 = larger variance of the two pOpulations $22 = smaller variance of the two populations F the F ratio with n1 degrees of freedom for the numerator and n2 degrees of freedom for the denominator (261, 177). Correlation Analysis: A zero order and multiple correlation analysis was performed to predict shorthand grades from raw scores of the (1) General Situational Choice Inventory (GSCI), (2) Preferred Job Characteristics Scale (PJCS), (3) Word Rating List (WRL), (4) Human Trait Inventory (HTI), (5) Total M-Scales Score, (6) American Council on Education Psychological Examination (ACE), (7) English grades, and (8) Overall Grade Point Average (GPA). The zero order and multiple correlation equations were solved using the general linear hypothesis which means that "if the Y values are plotted against the X values in a graph, the resulting trend of the plotted points can be 8Ibid., p. 272. 56 represented by a straight line."9 The Michigan State University Computer, CDC 3600, with CORE program was used in the analysis. The zero order correlations were tested to determine whether they were significantly different from zero. The tabled values of r were used with confidence limits of .05.10 The .01 level of confidence was also examined. The degrees of freedom used was determined by N - 2 where N = number of paired observations.11 In the present study, N equals 177 which produced 175 degrees of freedom for the zero order test. The multiple correlations were of two types. The first type multiple correlation was computed by using all the predictor variables. The second multiple was computed by combining a subscore of the M-Scales with the estimate of mental ability or with a subscore of the estimate of mental ability. The first type multiple was tested to determine the significant difference from zero. The test used was an F test12 which is as follows: 91bid., p. 116. 10Ibid., p. 502. 11Ibid., p. 303. 12W. L. Hays, Statistics for Psychologists (New York: Holt, Rinehart and Winston, 1963), p. 573. 57 F = ( fl) (N - K) (1 B) (K - l) where N = number of subjects (177) K = number of predictor variables (11) F = ratio with N — K (177 - 11 = 166) degrees of freedom for the numerator and K - 1 (11 - l = 10) degrees of freedom for the denominator. The assumption was made that a multivariate normal pOpulation had been sampled. The second type multiple was tested to determine the increase in multiple R resulting from an increase in the number of variables. An increase in multiple R is equiva- 13 lent to an increase in predictive value or precision. It was tested by using an F test advocated by Guilford. The formulaln is as follows: F = (B12 - B2 2) (N - m1 - 1) (1 - R12); (m1 - m2) where R1 = multiple R with larger number of independent variables R2 = multiple R with one or more variables omitted m1 = larger number of the independent variables (2) m2 = smaller number of independent variables (1) F =. the F ratio with m1 - m2 (2 - 1) degrees of freedom for the numerator and N — m1 - l (177 - 2 - 1: 174) degrees of freedom for the denominator. 13J. P. Guilford, Fundamental Statistics in Psychology and Education (New York: McGraw-Hill Book Company, Inc., 1956), p. #00. 14Ibid. 58 Null Hypotheses: The following hypotheses are restatements of the hypotheses* in null form: Ho 1. Each predictor variable will not significantly predict achievement of eleventh grade girls in first and second semesters of Gregg Shorthand. The predictor variables as a group will not significantly predict achievement of eleventh grade girls in first and second semesters of Gregg Shorthand. The Michigan M-Scales when combined with an estimate of mental ability will not increase the precision of prediction of achievement for eleventh grade girls in first and second semesters of Gregg Shorthand. *See page 3 for statement of research hypotheses. CHAPTER IV FINDINGS This chapter reports the findings of the investigation. The findings are divided into three parts. Part I contains the findings regarding the reliability test of the motivation score; Part II contains the findings regarding the comparison of the data in the present study to that of corresponding data from the Farquhar Study; and Part III is divided into two sections: Section I is a test of hypotheses for first semester findings, and Section II is a test of hypotheses for second semester findings. Part I Reliability The reliability estimates for all subscores of the Michigan M-Scales are reproduced in Table IX, Appendix, page 137. Kuder-Richardson #20 formula produced an estimate of reliability of the subscores of the Michigan M-Scales for the 177 subjects in the first semester analysis. The esti- mates of reliability for the subscores were .73 for the General Situational Choice Inventory (GSCI) and for the Human Trait Inventory (HTI), and were .89 for the Preferred Job Characteristics Scale (PJCS) and for the Word Rating 59 'l.'o l \.,.., 1.4‘ In: or- --.\ 60 List (WRL). A comparison of the reliability estimates for the present study with the reliability estimates of the Farquhar Study reveals that the responses of the population of the present study were more reliable on one test, the Preferred Job Characteristics Scale (PJCS) (.89 vs. .83), and less reliable on three tests, the General Situational Choice Inventory (GSCI) (.73 vs. .90); the Word Rating List (WRL) (.89 vs. .93); and the Human Trait Inventory (HTI) (.73 VS- .93). Part II Mean Differences Between Farquhar Findings and the Present m The mean scores of the female pOpulation of the Farquhar Study on the sub-tests of the Michigan M-Scales and on the total score of the Michigan M-Scales were higher than the mean score of the female pOpulation of the present study on the same tests. However, the mean grade point average of the female pOpulation of the present study was higher than the mean grade point average of the female pOpulation of the Farquhar Study. In order to determine whether or not these differences were significant, the data were treated statistically. Standard deviations and F (Fisher F) ratios for the com- parison of the female population of the Farquhar Study to the 61 female population in the present study are presented in Table X, Appendix, page 138. The F ratios ranged from 1.01 for the Word Rating List (WRL) to 1.32 for the General Situational Choice Inventory (GSCI). For the reason that all the obtained F values were less than the tabled value, (1.431) at the .01 confidence level with 200 degrees of freedom in the numerator and 150 degrees of freedom in the denominator, the assumption was made that no significant differences existed in the variances of the subscores and the total score on the Michigan M-Scales and the grade point average between the female pOpulation of the Farquhar Study and the pOpulation of the present study. This statis- tical analysis of the distribution of the scores indicated that the scores on the Michigan M-Scales and grade point average were distributed in a manner sufficiently similar to warrant the researcher's safely assuming that the dis- tribution of scores and grades were similar in the two populations. This statistical analysis was necessary in order to make the t tests to determine differences, if any, in the mean scores of the Michigan M—Scales and grade point average of the female population of the Farquhar Study and the population of the present study. Mean scores and t ratios with assumed equal variances for the female pOpulation of the Farquhar Study and the 1Edwards, op. cit., pp. 504-507. 62 population of the present study are presented in Table XI, Appendix, page 139. By using a series of two-tailed tests at the .01 confidence level, the following significant differences were found between the means of the female pOpulation of the Farquhar Study and the means of the population of the present study: (1) There was a significant difference between the mean score of the female pOpulation of the Farquhar Study and the mean score of the pOpulation of the present study on the total score of the Michigan M—Scales (t = -4.0530; tabled value, 2.5882 at the .01 confidence level). (2) There was a significant difference between the mean score of the female pOpulation of the Farquhar Study and the mean score of the pOpulation of the present study on the Preferred Job Characteristics Scale (PJCS) (t = -6.l943; tabled value, 2.5883 at the .01 confidence level). By using a series of two-tailed tests at the .05 confidence level, the following significant differences were found between the means of the female population of the Farquhar Study and the means of the pOpulation of the present study: 2Ibid.,p. 501, 3Ibid. 63 (1) There was a significant difference between the mean score of the female population of the Farquhar Study and the mgan_figgre of the population of the present study on Grade Point Average (GPA) (t = 2.4878; tabled value, 1.966"I at the .05 confidence level). (2) There was a significant difference between the mean score of the female population of the Farquhar Study and the mean score of the population of the present study on the Word Rating List (WRL) (t = -2.3957; tabled value, 1.9665 at the .05 confidence level). (3) There was a significant difference between the mean score of the female population of the Farquhar Study and the mean score of the pOpulation of the present study on the General Situational Choice Inventory (GSCI) (t = -2.0704; tabled value, 1.9666 at the .05 confidence level). There was no significant difference between.the pppp ppppp of the female population of the Farquhar Study and the mean score of the population of the present study at either the .05 or at the .01 confidence levels on the Human Trait Inventory (HTI) (t = -.2564; tabled value, 1.9667 64 at the .05 confidence level). The distribution of the responses of the population of the present study to the Michigan M-Scales was similar to the distribution of the responses of the pOpulation of the Farquhar Study to the Michigan M-Scales; however, the mean scores of the total score and subscores of the Michigan M-Scales of the population of the present study were signifi— cantly lower than the mean scores of the total score and subscores of the Michigan M-Scales of the pOpulation of the Farquhar Study, which indicates that, according to the re- sults of this test, the pOpulation of the present study was not as academically motivated as the population of the Farquhar Study. However, the mean score of the Grade Point Average of the population of the present study was signifi- cantly higher than the mean score of the Grade Point Average of the female population of the Farquhar Study which indicates that the population of the present study had a better aca- demic record than the female pOpulation of the Farquhar Study. Part III Test of Hypotheses Section I--First Semester: The first null hypothesis test was: Each predictor variable will not significantly predict achievement of eleventh grade girls in first semester of Gregg Shorthand (r = 0). 65 The zero order correlations between the predictor variables and shorthand grades were used to test this hypothesis and are summarized in Table XII, Appendix, page 140. The Table ”Value of the Correlation Coefficient for Different Levels of Significance"8 was used to determine significance of the zero order correlations. The tabled values of r for 175 degrees of freedom at .05 confidence level are .148 and at the .01 confidence level, .193. All but two of the zero order correlations between first semester shorthand grades and the predictor variables were found to be significant at the .01 confidence level since all were found to be equal to or greater than .193. The correlation between General Situational Choice Inventory (GSCI) and the shorthand grades and the correlation between the Preferred Job Characteristics Scale (PJCS) and shorthand grades were the only zero order correlations that were not significant at the .01 confidence level. The correlation between General Situational Choice Inventory (GSCI) and shorthand grades was significant at the .05 confidence level (r = .1627; tabled value, .1439 at the .05 confidence level) although it had not been significant at the .01 confidence level. The correlation between Preferred Job Characteristics Scale (PJCS) 8Ibid.,p. 502. 91bid. 66 and shorthand grades was not significant at the .05 confidence level (r = .0592; tabled value, .14310 at the .05 confidence level). Table XIII, Appendix, page 141, presents a summary of the inter-correlations of the Michigan M-Scales, American Council on Education Psychological Examination for high school students, and first semester shorthand grades. The table shows that the Word RatingpList (WRL) correlated with the American Council on Education Psychological Examination (.2290) higher than the total Michigan M-Scales correlated with the American Council on Education Psycholggical Exami- nation (.1702). Correlations also show that the relation- ship between shorthand achievement and the Word Rating List (WRL) (.2280), and the Human Trait Inventory (HTI) (.2235), and the total score of the Michigan M-Scales (.2219) was significantly different from zero at the .01 confidence level (tabled value of r is .19311 at .01 confidence level with 175 degrees of freedom). The correlationsbetween subtests and total score of the Michigan M-Scales and American Council on Education Psychological Examination are relatively low (.10 or less) with the exception of the Word Rating List (WRL) which correlated .2290 with American Council on Education loIbid. 11Ibid. 67 Psychological Examination. The correlation between the Human Trait Inventory (HTI) and American Council on Education Psychological Examination is .0943; however, the correlation between Human Trait Inventory (HTI) and shorthand grades was .2235 which is slightly higher than the correlation between Michigan M—Scales total score and shorthand grades (.2219). The Word Rating List (WRL) was found to correlate significantly different from zero with shorthand achievement (.2280) and was slightly higher than the correlation between Human Trait Inventory (HTI) and shorthand grades (.2235). The correlation between the Word Rating List (WRL) and the Human Trait Inventory (HTI) was .5535; between the Human Trait Inventory (HTI) and the General Situational Choice Inventory (GSCI) was .5112; and between the General Situational Choice Inventory (GSCI) and the Preferred Job Characteristics Scale (PJCS) was .6226. These correlations indicate that there is a considerable amount of commonalty between the subtests and total score of the Michigan M-Scales but that there is a relatively small amount of commonalty between the American Council on Education Psychological Examination and the Michigan M-Scales. The correlations indicate that the relationship between the predictor variables and shorthand grades was significantly different from zero; thus, the null hypothesis was rejected except for the part that pertained to the Preferred Job 68 Characteristics Scale for which the null hypothesis was accepted. The second null hypothesis tested was: The predictor variables as a group will not significantly predict achievement of eleventh grade girls in first semester Gregg Shorthand (R = O). The multiple correlation computed from grade point average, tenth grade English grades, ninth grade English grades, American Council on Education Psychological Exami- nation subscores, and subscores of the Michigan M-Scales was .6984. For the reason that the obtained value from the pppp, (16.8098)12 was greater than the tabled value (F100,7= 5.70 at the .01 confidence levell3), the assumption was made that the multiple correlation obtained in the present study was significantly greater than zero; thus, the null hypoth- esis was rejected. The third null hypothesis tested was: The Michigan M-Scales subscores or total score when combined with an estimate of mental ability will not significantly increase the precision of prediction of achievement for eleventh grade girls in first semester Gregg Shorthand. Tables XIV, XV, and XVI, Appendix, pages 142, 143, and 144, present a summary of the zero order correlations, multiple order correlations, and results of the F tests made to statistically test the third hypothesis. 2 12Formula used: F = (* R 1 IN -K-) Hays 22;_2l2° (1-Rz)(K-1) ’ 13Edwards, op. cit.,pp. 504-507. 69 To test this hypothesis, a multiple correlation was computed by combining the total score or a subscore of the Michigan M-Scales with the total score or a subscore of the American Council on Education Psychological Examination. The computed multiple correlation between the two inde- pendent variables and the dependent variable was then statis- tically compared, by means of an F test,1LI with the zero order correlation between the total score or a subscore of the American Council on Education Psychological Examination and first semester shorthand grades to determine the signifi- cance of increase in predictive value. The F values obtained by the statistical tests were compared with the tabled values of F for l and 174 degrees of freedom at the .01 and .05 confidence levels. Obtained F values greater than 6.785,15 the tabled value of F for l and 174 degrees of freedom at the .01 confidence level, were considered significant at the .01 confidence level. Obtained F values greater than 3.90, the tabled value of F for l and 174 degrees of freedom at the .05 confidence level, but less than 6.785,16 the tabled value of F for 1 and 174 degrees of freedom at the .01 confidence level, were considered significant at the .05 confidence level. (R - R ) (N - m - 1) 1“Formula used: F = l g 1 (l - B1 ) (m1 - m2) 15Ibid. 16 Ibid. o .1.- F4: 1. n ”\h 11% .n .\\ 70 Obtained F values less than 3.90,17 the tabled value of F for l and 174 degrees of freedom at the .05 confidence level, were considered not significant. F Tests at the .01 Confidence Level: By using a series of F tests at the .01 confidence level, the following significant increases in predictive value were found when the total score or a subscore of the Michigan M-Scales was used in combination with the total score or a subscore of the American Council on Education Psychological Examination to predict achievement of eleventh grade girls in first semester Gregg Shorthand (Tables XIV, XV, and XVI, Appendix, pages 142, 143, and 144). (1) There was a significant increase in predictive value when the Human Trait Inventory (HTI) was combined with the American Council on Education Psychological Examination (ACE) (r .4190, R = .4621; F = 8.1856, Table ACE = ACE + HTI XIV, Appendix, page 142, tabled value, 6.785 with 1 and 175 degrees of freedom at .01 confidence leve118); thusL this part of the null hypothesis was rejected at the .01 confi- dence level. (2) There was a significant increase in predictive value when the total score of the Michigan M-Scales was combined with the American Council on Education Psychological 171bid. 181bid. cl.- 71 (rACE = RACE + total M—Scales ’ .4945; F = 17.3826, Table XIV, Appendix, page 142, tabled Examination (ACE) .4190, value, 6.785 at .01 confidence 1eve119); thusL thisgpart of the null hypothesis was rejected at the .01 confidence level. (3) There was a significant increase in predictive value when the Human Trait Inventory (HTI) was combined with the Linguistic score, a subscore of the American Council on Education Psychological Examination, (L—ACE) .3454, R = .4033; F = 7.6944, Table (TL—ACE = L-ACE + HTI XV, Appendix, page 143, tabled value, 6.785 at .01 confi- dence levelzo); thus, thisypart of the null hypothesis was rejected at the .01 confidence level. (4) There was a significant increase in predictive value when the total score of the Michigan M-Scales was combined with the Linguistic score, a subscore of the American Council on Education Psycholggical Examination, (L-ACE) (r .3454, R .4091; . L-ACE = L-ACE + total M-Scales = F = 10.0361, Table XV, Appendix, page 143, tabled value, 6.785 at .01 confidence leve121); thus, this part of the null hypothesis was rejected at the .01 confidence level. (5) There was a significant increase in predictive value when.the'mtal score of the Michigan M-Scales was l9Ibid. 20Ibid. 21Ibid. 72 combined with the Quantitative score; a subscore of the American Council on Education Psychological Examination, (Q-ACE) (r .4032, R Q-ACE = Q-ACE + total M—Scales = '4517‘ F = 9.0736, Table XVI, Appendix, page 144, tabled value, 6.785 at .01 confidence 1eve122); thus, thisypart of the null hypothesis was rejected at the .01 confidence level. F tests at the .05 Confidence Level: By using a series of F tests at the .05 confidence level, the following significant increases in predictive value were found when the total score or a subscore of the Michigan M—Scales was used in combination with the total score or a subscore of the American Council on Education Psychological Examination to predict achievement of eleventh grade girls in first semester Gregg Shorthand (Tables XIV, XV, and XVI, Appendix, pages 142, 143, and 144). (1) There was a significant increase inypredictive value when the Word Rating List (WRL) was combined with the American Council on Education Psychological Examination, (ACE) (rACE = .4190, R .4443; F = 4.7261, Table ACE + WRL = XIV, Appendix, page 144, tabled value,3.90 at .05 confidence leve123); thus, this part of the null hypothesis was rejected at the .05 confidence level. 221bid. 23Ibid. 73 (2) There was a significant increase inypredictive value when the Word Rating List (WRL) was combined with the Linguistic score, a subscore of the American Council on Education Psycholggical Examination,(L-ACE) (r .3454, L-ACE = R = 4.8954, Table XV, Appendix, page L—ACE + WRL = '3787‘ F 143, tabled value, 3.90 at .05 confidence levelzu); thus, this part of the null hypothesis was rejected at the .05 confidence level. (3) There was a significant increase in predictive value when the Word Rating List (WRL) was combined with the Quantitative score, a subscore of the American Council on Education Psychological Examination,(Q-ACE) (rQ—ACE = .4032, RQ-ACE + WRL = .4403; F = 6.7562, Table XVI, Appendix, page 144, tabled value, 3.90 at .05 confidence leve125); thus, this part of the null hypothesis was rejected at the .05 confidence level even though this F value was close to the significant value at the .01 confidence level (6.785). (4) There was a significant increase in predictive value when the Human Trait Inventory (HTI) was combined with the Quantitative scoreL a subscore of the American Council on Education Psychological Examination,(Q-ACE) (rQ-ACE = “4032’ BQ-ACE + HTI = XVI, Appendix, page 144, tabled value, 3.90 at .05 confidence .4312; F = 5.0013, Table Ibid. 251bid. 74 leve126); thus, this part of the null hypothesis was rejected at the .05 confidence level. By using a series of F tests, the foDowing increases in predictive value were found to be insignificant at either the .01 or .05 confidence level: (1) There was no significant increase in predictive value when the General Situational Choice Inventory (GSCI) was combined with the American Council on Education Psychological Examination (ACE) (rACE = .4190, RACE + GSCI = .4398; F = 3.8398, Table XIV, Appendix, page 142, tabled values, 6.785 at the .01 confidence level and 3.90 at the .05 confidence leve127); thus, this part of the null hypoth- esis was accepted at the .01 and .05 confidence levels. (2) There was no significant increase in predictive value when the Preferred Job Characteristics Scale (PJCS) was combined with the American Council on Education Psycholggical Examination (ACE) (rACE = .4190, RACE + PJCS = .4266; F = 1.3613, Table XIV, Appendix, page 142, tabled values, 6.785 at .01 confidence level and 3.90 at .05 confi- dence leve128); thus, this part of the null hypothesis was accepted at the .01 and .05 confidence levels. (3) There was no significant increase ingpredictive value when the General Situational Choice Inventory (GSCI) 26Ibid. 271bid. 28Ibid. 75 was combined with the Linguistic Score, a subscore of the American Council on Education Psychological Examination, (L-ACE) (r .3454, R = .3701; F = 3.6627, L—ACE = L-ACE + GSCI Table XV, Appendix, page 143, tabled values, 6.785 at .01 confidence level and 3.90 at .05 confidence leve129); phppp thisypart of the null hypothesis was accepped at the .01 and .05 confidence levels. (4) There was no significant increase in predictive value when the Preferred Job Characteristics Scale (PJCS) was combinediwith the Linguistic score, a subscore of the American Council on Education Psycholggical Examination, (L—ACE) (rL-ACE = .3454, RL-ACE + PJCS = .3503; F = .6743, Table XV, Appendix, page 143, tabled values, 6.785 at .01 confidence level and 3.90 at .05 confidence levelBO); phppp this part of the null hypothesis was accepted at the .01 and .05 confidence levels. (5) There was no significant increase in predictive value when the General Situational Choice Inventory (GSCI) was combined with the Quantitative score, a subscore of the American Council on Education Psychological Examination, (Q-ACE) (r .4032, R .4201; F = 2.9370, Q-ACE = Q-ACE + GSCI = Table XVI, Appendix, page 144, tabled values, 6.785 at .01 confidence level and 3.90 at .05 confidence levelBl); thus, 29Ibid. 30Ibid. 311bid. lg" lh— 76 thisgpart of the null hypothesis was accepted at the .01 and .05 confidence levels. (6) There was no significant increase in predictive value when the Preferred Job Characteristics Scale (PJCS) was combined with the Quantitative score, a subscore of the American Council on Education Psychological Examination, (Q-ACE) (r .4032, .4037; F = .0831, Q-ACE = RQ-ACE + PJCS = Table XVI, Appendix, page 144, tabled values, 6.785 at .01 confidence level and 3.90 at .05 confidence leve132); thus, this part of the null hypothesis was accepted at the .01 and .05 confidence levels. Section II--Second Semester: The first null hypothesis tested was: Each predictor variable will not significantly predict achievement of eleventh grade girls in second semester of Gregg Shorthand (r = 0). The zero order correlations were used to test this hypothesis. The Table ”Value of the Correlation Coefficient for Different Levels of Significance"33 was used to determine significance of the zero order correlations. The correla— tions between the predictor variables and shorthand grades for the 171 subjects in the second semester analysis are summarized in Table XVII, Appendix, page 145. The tabled 32Ibid. 33Edwards, op. cit., p. 502. Fl-\ fl 77 value of r for 150 degrees of freedom at .05 confidence level is .159 and at .01 confidence level is .208. All except four of the zero order correlations between second semester shorthand grades and the predictor variables were found to be significant at the .01 confidence level since all were found to be greater than .208. The four zero order correlations found to be not significant were: (a) the correlation between General Situational Choice Inventory (GSCI) and shorthand grades; (b) the correlation between Preferred Job Characteristics Scale (PJCS) and shorthand grades; (c) the correlation between the Human Trait Inventory (HTI) and shorthand grades; and (d) the correlation between the total Michigan M-Scales and shorthand grades, Thus, the null hypothesis was rejected at the .01 level of confidence except for the parts which pertained to the total score of the Michigan M-Scales for which the null hypothesis was rejected at the .05 confidence level and for those parts which pertained to the General Situational Choice Inventory (GSCI), Preferred Job Characteristics Sca1e (PJCS), and Human Trait Inventory (HTI) for which the null hypothesis was accepted at .05 and .01 confidence levels. By comparing Table XII, Appendix, page 140, ”Zero Order Correlations Between First Semester Shorthand Achievement and Predictor Variables for the Population of the Present Study" with Table XVII, Appendix, page 145, "Zero Order A W?“ VJQI g)» in- 3'... I '4 Ya; n- v 9.- o 0 H 's b" E? '- 78 .Correlations Between Second Semester Shorthand Achievement and Predictor Variables for the Population of the Present Study" certain increases and decreases in correlations were noted. The noted increases were: (1) Tenth grade English increased from .5078 for first semester to .5253 for second semester. (2) American Council on Education Psychological Examination (ACE) increased from .4190 for first semester to .4704 for second semester. I3) American Council on Education Psychological Examination Linguistic subscore increased from .3454 for first semester to .4491 for second semester. The noted decreases were: (1) Grade Point Average (GPA) decreased from .6743 for first semester to .6011 for second semester. (2) Ninth grade English decreased from .5212 for first semester to .4684 for second semester. (3) M-Scales total score decreased from .2219 for first semester to .1593 for second semester. (4) General Situational Choice Inventory (GSCI) decreased from .1627 for first semester to .0905 for second semester. (5) Preferred Job Characteristics Scale (PJCS) decreased from .0592 for first semester to .0169 for second semester. n... - I.. ~ is F’v A ~. \ a 79 (6) Word Rating List (WRL) decreased from .2280 for first semester to .2216 for second semester. (7) Human Trait Inventory (HTI) decreased from .2235 for first semester to .1114 for second semester. (8) American Council on Education Psychological Examination Quantitative subscore decreased from .4032 for first semester to .2624 for second semester. The above changes in correlation from first semester shorthand grades to second semester shorthand grades indicate that a predictor variable may be of value in predicting achievement for one semester but of greater or lesser value in predicting shorthand achievement for another semester. The second null hypothesis tested was: The predictor variables as a group will not significantly predict achievement of eleventh grade girls in second semester Gregg Shorthand (R = O). The multiple correlation computed from grade point average, tenth grade English grades, ninth grade English grades, American Council on Education Psychological Exami- nation subscores, and M-Scales subscores was .6399. For the reason that the obtained value from the test (F = 12.7788; tabled value, 5.70 at .01 confidence leve134) was greater than the tabled value at .01 confidence level, the 34Edwards, op. cit., pp. 504-507. 80 assumption was made that the multiple correlation obtained in the present study was significantly greater than zero; thus, the null hypothesis was rejected at the .01 confidence 121E1- The third null hypothesis tested was: The Michigan M-Scales subscore or total score when combined with an estimate of mental ability will not significantly increase the precision of pre- diction of achievement for eleventh grade girls in second semester of Gregg Shorthand. Tables XVIII, XIX, and XX, Appendix, pages 146, 147, and 148, present a summary of the zero order correlations, multiple order correlations, and results of the F tests made to statistically test the third hypothesis for second semester. To test this hypothesis, a multiple correlation was computed by combining the total score or a subscore of the Michigan M-Scales with the total score or a subscore of the American Council on Education Psychological Examina- tion. The computed multiple correlation between the two independent variables and the dependent variable was then statistically compared, by means of an F test, with the zero order correlation between the total score or a subscore of the American Council on Education Psychological Examination and second semester shorthand grades to determine the significance of increase in predictive value. The F values obtained by the statistical tests were compared with the tabled values of F for l and 168 degrees 81 of freedom at the .01 and .05 confidence levels. Obtained F values greater than 6.78535, the tabled value of F for l and 168 degrees of freedom at the .01 confidence level, were Considered significant at the .01 confidence level. Obtained F values greater than 3.90, the tabled value of F for l and 168 degrees of freedom at the .05 confidence level, but less than 6.78536, the tabled value of F for l and 168 degrees of freedom at the .01 confidence level, were con- sidered significant at the .05 confidence level. Obtained F values less than 3.9037, the tabled value of F for l and 168 degrees of freedom at the .05 confidence level, were considered not significant. F Tests at the .01 Confidence Level: By using a series of F tests at the .01 confidence level, the following significant increases in predictive value were found when the Michigan M-Scales were used in combination with the Quantitative score, a subscore of the American Council on Education Psychological Examination, to predict achievement of eleventh grade girls in second semester Gregg Shorthand: 351bid. 36Ibid. 37Ibid. 82 (1) There was a significant increase inypredictive value when the Word Rating List (WRL) was combined with the Quantitative score, a subscore of the American Council on Educational Psychological Examination,(Q-ACE) (rQ-ACE = .2624, R .3281; F = 7.2860, Table XX, Appendix, Q-ACE + WRL = page 148, tabled value, 6.785 at .01 confidence leve138); thus, this part of the null hypothesis was rejected at the .01 confidence level. (2) There was a significant increase inypredictive value when the total score of the Michigan M-Scales was combined with the Quantitative score; a subscore of the American Council on Education Psychological Examination, (Q-ACE) (r .2624, R Q-ACE = Q-ACE + total M-Scales = '3347‘ F = 8.1541; Table XX, Appendix, page 148, tabled value, 6.785 at .01 confidence level39); thusL this part of the null hypothesis was rejected at the .01 confidence level. F Tests at the .01 and .05 Confidence Levels: By using a series of F tests at the .01 and .05 confidence levels, no significant increases in predictive value were found when the total score or subscores of the Michigan M-Scales were used in combination with the total score or subscores of the American Council on Education 38Ibid. 39Ibid. 83 Psychological Examination except as indicated above; thus, these parts of the null hypothesis were accepted at the .01 and .05 confidence levels. It should be pointed out that four combinations approached the tabled value at the .05 confidence level (tabled value for l and 168 degrees of freedom at .05 level of confidence is 3.9040). These com- binations were as follows: (1) Word Rating List (WRL) and total score of the American Council on Education Psychological Examination .4858; F = 3.2256; Table (ACE) ( .4704, R rACE = ACE + WRL = XVIII, Appendix, page 146, tabled value, 3.90 at .05 confi- dence levelul). (2) Total score of the Michigan M-Scales and total score of the American Council on Education Psychological Examination (ACE) ( .4704, R I'ACE 2 ACE + total M—Scales = .4868; F = 3.4560; Table XVIII, Appendix, page 146, tabled value, 3.90 at .05 confidence levelnz). (3) Word Rating List (WRL) and the Linguistic score, a subscore of the American Council on Education Psychological L-ACE = .4491, RL-ACE + WRL = '4671‘ F = 3.5457; Table XIX, Appendix, page 147, tabled value, Examination (L-ACE) (r 84 3.90 at .05 confidence 1eve143). (4) Total score of the Michigan M-Scales and Linguis- tic score, a subscore of the American Council on Education Psychological Examination,(L-ACE) (r .4491, R L-ACE z L-ACE + total M-Scales = .4686; F = 3.8534, Table XIX, Appendix, page 147, tabled value, 3.90 at .05 confidence leveluu). The above analysis indicates that the Word Rating List (WRL), Human Trait Inventory (HTI), and total score of the Michigan M-Scales significantly increased the precision of prediction of the American Council on Education Psychologi- cal Examination (ACE) in predicting achievement for eleventh grade girls in first semester Gregg Shorthand; however, in second semester, only the Word Rating List (WRL) and total score of the Michigan M-Scales significantly increased the precision of prediction of the Quantitative score, a sub— score of the American Council on Education Psychological Examination (Q-ACE). First Semester Uniform Shorthand Examination: Table XXI, Appendix, page 149, presents a summary of the zero order correlations between the predictor variables of the present study and Shorthand I uniform examination, Shorthand I teacher grades, Shorthand II uniform examination, ”31bid. qubid. 85 and Shorthand II teacher grades. The first semester uniform shorthand examination, which was administered to 68 subjects in the three schools using the test, was given to validate teacher grades. The total score of the test correlated with teacher grades at .7762 which indicates that teacher grades may be accepted as a valid measure of student achievement for first semester in the three participating schools. All but six of the correlations between the predictor variables and teacher grades for first semester were higher than the correlations between the predictor variables and the total score of the researcher's uniform shorthand exami- nation for first semester. The six correlations that were lower were: (1) The correlation between the total score of the uniform shorthand examination for first semester and the tenth grade English grades was .4779 (Table XXI, Appendix, page 149) while the correlation between the tenth grade English grades and the teacher grades for first semester shorthand was .4456 (Table XXI, Appendix, page 149). (2) The correlation between the total score of the American Council on Education Psychological Examination and the total score of the uniform shorthand examination for first semester was .5115 (Table XXI, Appendix, page 149) while the correlation between the total score of the American Council on Education Psychological Examination and teacher grades for first semester shorthand was .3880 (Table 86 XXI, Appendix, page 149). (3) The correlation between the Linguistic score, a subscore of the American Council on Education Psychological Examination,and the total score of the uniform shorthand examination for first semester was .4001 (Table XXI, Appendix, page 149) while the correlation between the Linguistic score and teacher grades for first semester shorthand was .2818 (Table XXI, Appendix, page 149). (4) The correlation between the Quantitative score, a subscore of the American Council on Education Psychological Examination, and the total score of the uniform shorthand examination for the first semester was .5313 (Table XXI, Appendix, page 149) while the correlation between the Quantitative score and teacher grades for first semester shorthand was .4658 (Table XXI, Appendix, page 149). (5) The correlation between the theory score, a sub- score of the uniform shorthand examination for first semester, and the total score of the uniform shorthand examination for first semester was .7994 (Table XXI, Appendix, page149) while the correlation between the theory score and teacher grades for first semester shorthand was .7017 (Table XXI, Appendix, page 149). (6) The correlation between the transcription score, a subscore of the uniform shorthand examination for first semesten and the total score of the uniform shorthand examination for first semester was .8635 (Table XXI, 87 Appendix, page 149) while the correlation between the transcription score, a subscore of the uniform shorthand examination for first semester, and Shorthand 1 teacher grades was .5773 (Table XXI, Appendix, page 149)- The correlation analysis for second semester for the 68 subjects in the three schools that used the uniform shorthand examination for first semester showed that the teacher grades correlated higher with the predictor variables than did the uniform shorthand examination for second semester in all but two cases. The two exceptions were: (1) The correlation between the brief form-theory score, a subscore of the uniform shorthand examination for second semester, and the total score of the second semester uniform shorthand examination was .5088 (Table XXI, Appendix, page 149) while the correlation between the brief form-theory score and teacher grades for second semester shorthand was .4378 (Table XXI, Appendix, page 149). (2) The correlation between the transcription score, a subscore of the uniform shorthand examination for second semester, and the total score of the uniform shorthand examination for second semester was .9632 (Table XXI, Appendix, page 149) while the correlation between the transcription score and teacher grades for second semester shorthand was .6551 (Table XXI, Appendix, page 149). A major portion of the uniform shorthand examination 88 for second semester consisted of transcription which accounts for the correlation of .9632 (Table XXI, Appendix, page 149) between the transcription subscore for second semester shorthand and the total score on the uniform shorthand examination for second semester for the 68 subjects who used the uniform shorthand examination for first semester. The correlation between the total score of the uniform shorthand examination for first semester and the teacher grades for first semester shorthand was .7762 (Table XXI, Appendix, page 149) which indicates that teacher grades were valid measures of Shorthand I student achievement in the three schools using the uniform shorthand examination for first semester. The correlation between the total score of the uniform shorthand examination for second semester and teacher grades for second semester shorthand was .7058 (Table XXI, Appendix, page 149) which indicates that teacher grades were valid measures of Shorthand II student achievement for the 68 subjects in the three schools that had used the uniform shorthand examination for first semes- ter. The correlation between teacher grades for first semester shorthand and teacher grades for second semester shorthand was .7528 (Table XXI, Appendix, page 149). This correlation of .7528 (Table XXI, Appendix, page 149) is: (1) higher than the correlations between the total score or subscores of the uniform shorthand examination for first semester and teacher grades for second semester 89 shorthand; (2) higher than the correlations between the total score or subscores of the uniform shorthand examination for first semester and thetotal score or subscores of the uniform shorthand examination for second semester; and (3) higher than all the zero order correlations between the predictor variables of the present study and Shorthand II teacher grades for the 68 subjects who took the uniform Shorthand I examination in the three participating schools. Analysis of Table XXII, "Intercorrelations Between Uniform Shorthand I Examination, Uniform Shorthand II Examination, Shorthand I Teacher Grades, and Shorthand 11 Teacher Grades,” Appendix, page 151. shows that the corre- lation between the first semester transcription score and the second semester transcription score was .2125 (Table XXII, Appendix, page 151) which indicates that the relation- ship was not significantly different from zero between the two variables for the 68 subjects in the present study who took the uniform Shorthand I examination (tabled value for 60 degrees of freedom at the .05 confidence level is 025045). The correlation between the theory score,a sub— score of the uniform shorthand examination for first semes- ter, and the transcription section of the uniform shorthand examination for second semester was .6315 (Table XXII, 45Ibid., p. 502. 90 Appendix, page 151) which indicates that there is a consider- able amount of relationship between the two variables. The correlation between the brief form score, a subscore of the uniform shorthand examination for first semester, and the transcription score, a subscore of the uniform shorthand examination for second semester, was .0064 (Table XXII, Appendix, page 151) which indicates there is no relation- ship between the variables. The above analysis of the uniform Shorthand I exami- nation indicates that a predictor variable for Shorthand I may be of greater or lesser value in predicting Shorthand II achievement. §econd Semester Uniform Shorthand Examination: Table XXIII, Appendix, page 152. presents a summary of the zero order correlations between the predictor variables of the present study and the Shorthand II uniform examination, Shorthand I teacher grades, and Shorthand II teacher grades. The second semester uniform shorthand examination was administered to 153 subjects in the eight participating schools. The correlation between the total score of the uni- form examination for second semester shorthand and Shorthand II teacher grades was .6198 (Table XXIII, Appendix, page 152) which indicates that teacher grades may be accepted as valid measures of student achievement for second semester. All but two of the correlations between the predictor 91 variables and teacher grades for second semester shorthand were higher than the correlations between the predictor variables and the uniform shorthand examination for second semester. The two correlations that were lower were: (1) The correlation between the Preferred Job Charac— teristics Scale (PJCS) and teacher grades for second semes- ter shorthand was -.0215 (Table XXIII, Appendix, page 152) compared to a correlation of .0434 (Table XXIII, Appendix, page 152) between the Preferred Job Characteristics Scale (PJCS) and total score of the uniform shorthand examination for second semester. (2) The correlation between the transcription score, a subscore of the uniform shorthand examination for the second semester, and teacher grades for second semester shorthand W88‘.5614 (Table XXIII, Appendix, page 152) compared to a correlation of .9853 (Table XXIII, Appendix, page 152) between the transcription score, a subscore of the uniform Shorthand examination for second semester, and the total score of the uniform shorthand examination for second semester. This extremely high correlation of .9853 (Table XXIII, Appendix, page 152) is due to the composition of the uniform Shorthand II examination. The examination consisted of 100 points for the brief form-theory section and a maximum of 360 points for the transcription section of the test (Appendix, pages 167-177). These two scores were added to make the total score; thus, the high 92 correlation of .9853 (Table XXIII, Appendix, page 152) between the transcription score and the total score of the uniform shorthand examination for second semester. Table XXIV, Appendix, page 153, presents a summary of the intercorrelations between the first semester shorthand grades, second semester shorthand grades, and the subscores and the total score of the uniform shorthand examination for the second semester for the 153 subjects who took the exami- nation. The correlations presented in Table XXIV, Appendix, page 153, indicate that there is a high degree of commonalty among the predictor variables. The correlation between the Shorthand II teacher grades and the brief form-theory sub- score of the uniform shorthand examination for second semester shorthand was .5986 (Table XXIV, Appendix, page 153) which is higher than the correlation between the teacher grades for second semester shorthand and the tran- scription subscore which was .5614 (Table XXIV, Appendix, page 153). The correlation between the brief form-theory subscore of the uniform shorthand examination for second semester and the transcription score of the uniform short— hand examination for second semester was .4373 (Table XXIV, Appendix, page 153) which indicates that there is a relation- ship between the brief form-theory section of the uniform shorthand examination and the transcription section of the uniform shorthand examination. This correlation of .4373 (Table XXIV, Appendix, page 153) is lower than the (A In 93 correlation found between the theory score of the first semester uniform shorthand examination and the transcription section of the second semester uniform shorthand examination which was .6315 (Table XXIV, Appendix, page 153) for the 68 subjects who took the first semester uniform shorthand examination; however, this correlation of .6315 (Table XXIV, Appendix, page 153) is much higher than the correlation of .0064 (Table XXIV, Appendix, page 153) between the brief form subscore of the first semester uniform shorthand examination and the transcription subscore of the uniform shorthand examination for first semester. The difference between .4373 and .6315 (Table XXIV, Appendix, page 153) might be due to differences in number of peOple (68 for first semes- ter and 153 for second semester) or to a suppressing effect of the brief form score when used in combination with the theory score. Summary of Findings: The summary of findings is divided into two parts-- Part I is a summary of the general findings; Part II is a summary of the findings for first and second semester. Part I--General Findings: The general findings are summarized as follows: (1) The subjects' responses to the Michigan M—Scales were found to be reliable. The estimates of reliability in the present study were higher than the estimates of 94 reliability for the female population of the Farquhar Study for the Preferred Job Characteristics Scale (PJCS) and for the Word Rating List (WRL); however, the estimates of reli- ability were lower than the estimates of reliability for the female population of the Farquhar Study for the General Situational Choice Inventory (GSCI) and Human Trait Inven— pgry (HTI). (2) No significant differences were found between variances (distribution of scores) of grade point average and responses to the Michigan M-Scales between the female population of the Farquhar Study and the population of the present study. (3) The mean score of the total score of the Michigan M-Scales and the mean score of each of the subscores of the Michigan M-Scales was significantly less for the population in the present study than the mean score of the female population of the Farquhar Study; however, the mean score of the grade point average for the pOpulation in the present study was significantlyygreater than the mean score of the grade point average of the female population of the Farquhar Study. Part II--Summary of Findings for First and Second Semester: Findings for first and second semester are summarized as follows: (1) Grade point average, tenth grade English grades, ninth grade English grades, total score and subscores of the 95 American Council on Education Psychological Examination, and the Word Rating List,which is a subscore of the Michigan M-Scales, individually correlated with shorthand teacher grades significantly greater than zero for first and second semester of shorthand. The total score of the Michigan M- Scales, General Situational Choice Inventory, and Human Trait Inventory,which are subscores of the Michigan M-Scales, individually correlated with shorthand teacher grades sig- nificantly greater than zero for first semester shorthand. The Preferred Job Characteristics Scale, which is a subscore of the Michigan M—Scales, did not correlate with shorthand teacher grades significantly different from zero for first and second semester shorthand. (2) The predictor variables used in the present study when used as a group did predict achievement significantly different from zero of eleventh grade girls in first semester of Gregg Shorthand. (3) The combination of the total score of the American Council on Education Psychological Examination with either (a) the Word Rating List, a subscore of the Michigan M-Scales, or (b) the Human Trait Inventory, a subscore of the Michigan M-Scales, or (c) the total score of the Michigan M-Scales, produced a significant increase in precision ofgprediction over the use of the American Council on Education Psycho- logical Examination as a single predictor in predicting achievement of eleventh grade girls in first semester Gregg 96 Shorthand; however, these combinationsdid notyproduce a significant increase for second semester. (4) The combination of the Linguistic score, a sub- score of the American Council on Education Psychological Examination,with either (a) the Word Rating List, a subscore of the Michigan M-Scales, or (b) the Human Trait Inventory, a subscore of the Michigan M-Scales, or (c) the total score of the Michigan M-Scales produced a significant increase in precision ofyprediction over the use of the Linguistic score, a subscore of the American Council on Education Psychological Examination, as a single predictor in predict- ing achievement of eleventh grade girls in first semester Gregg Shorthand; however, these combinations did not produce significant increases for second semester. (5) The combination of the Quantitative score, a sub- score of the American Council on Education Psychological Examination, with either (a) the Word Rating List, a sub- score of the Michigan M-Scales, or (b) the Human Trait Inventory, a subscore of the Michigan M-Scales, or (c) the total score of the Michigan M-Scales produced a significant increase ingprecision of prediction over the use of the Quantitative score as a single predictor in predicting achievement of eleventh grade girls in first semester Gregg Shorthand; and the combination of the Quantitative score with the Word Rating List or total score of the Michigan M-Scales produced a significant increase in precision ofyprediction 97 for second semester also. (6) The correlation between the uniform examination for first semester Gregg Shorthand and teacher grades was significant. The correlations between the uniform examina- tion for first semester Gregg Shorthand and the predictor variables of the present study were lower than the corre- lations between teacher grades and the predictor variables except for the correlation between the total score of the uniform examination with (a) the American Council on Edu- cation Psychological Examination total score and subscores; with (b) tenth grade English grades; with (c) theory score, a subscore of the uniform examination for first semester; and with (d) the transcription score, a subscore of the uniform examination for second semester. (7) The correlation between the uniform examination for second semester Gregg Shorthand and teacher grades was significant. The correlations between the uniform examina— tion for second semester Gregg Shorthand and the predictor variables of the present study were lower than the correla— tions between teacher grades for second semester Gregg Shorthand and the predictor variables, except for (a) the correlation between the Preferred Job Characteristics Scale and the total score of the uniform examination for second semester Gregg Shorthand, and (b) the correlation between the transcription score, a subscore of the uniform examina— tion for second semester Gregg Shorthand. 98 (8) The present study found that the best single pre- dictors of Shorthand I achievement were: (a) grade point average (.6743); (b) ninth grade English grades (.5212); (c) tenth grade English grades (.5708); and (d) mental ability as measured by the American Council on Education Psychological Examination (.4190) (Table XII, Appendix, page 139). The best single predictors of Shorthand II achievement (in order) were: (a) Shorthand I teacher grades (.7459); (b) grade point average (.6011); (c) tenth grade English grades (.5253); and (d) mental ability as measured by the American Council on Education Psychological Examination (.4704) (Table XVII, Appendix, page 144). Summary of Findings of the Hypotheses: A summary of the findings of the hypotheses is as follows: (1) Each predictor variable did significantly predict achievement of eleventh grade girls in first semester Gregg Shorthand except for the Preferred Job Characteristics Scale (PJCS), a subscore of the Michigan M-Scales. Each predictor variable did significantly predict achievement of eleventh grade girls in second semester Gregg Shorthand except for the Preferred Job Characteristics Scale (PJCS), the General Situational Choice Inventory (GSCI), and the Human Trait Inventory (HTI), subscores of the Michigan 99 (2) The predictor variables as a group did signifi- cantly predict achievement of eleventh grade girls in first and second semesters of Gregg Shorthand. (3) The Michigan M-Scales when combined with an estimate of mental ability did significantly increase theyprecision of prediction of achievement for eleventh grade girls in first semester of Gregg Shorthand. However, the Michigan M-Scales when combined with an estimate of mental ability did not significantly increase the precision of prediction of achievement for eleventh grade girls in second semester Gregg Shorthand. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Chapter V is divided into three parts. Part I presents a summary of the study; Part II presents the conclusions of the study; and Part III presents the recommendations of the study. Part I Summary Statement of the Problem: The problem of the present study was to find a criterion for predicting achievement of students in first and second semesters of Gregg Shorthand that would be useful to high school teachers. The present study was specifically concerned with the relationship of academic motivation to achievement in first and second semester shorthand and the extent to which motivation might add to the precision of predicting achievement. Purpose of the Study: The purpose of this specific study was to determine the predictive value, if any, of the Michigan M-Scales total score or subscores for predicting achievement of eleventh grade girls in first and second semesters of Gregg Shorthand when used individually or in combination with the total score or a subscore of an esti- mate of mental ability. lOO 101 Need for the Study: A review of the studies in short- hand prognosis showed that there were no known reliable predictors of shorthand achievement; however, many researchers concluded that a measurement of motivation, interest, desire, and need might aid in predicting success. Delimitations: The study was delimited to eleventh grade girls in eight participating high schools in central and southern Michigan who enrolled in and completed the first and/or the second semester of Gregg Shorthand during the 1964-1965 school year. Review of Literature: The review of literature showed that considerable effort has been spent on analyzing short- hand achievement through tests of mental ability, English tests and grades, grade point average, and a host of other criteria. The review showed that the best known predictors of shorthand achievement are grade point average, English grades, and mental ability. There is disagreement among researchers as to the value of these three predictors; however, the criteria used for each analysis appears to cause the discrepancies. Several researchers mentioned the need for an instrument to measure interest, motivation, or desire of a student. The researcher utilized a test designed to measure academic motivation which had been created by Dr. W. W. Farquhar and his associates of the Michigan State University faculty in 1963. The Farquhar Study identified 102 overachievers and underachievers by means of mental ability and grade point average based on a regression analysis. From the discrepant achievers, certain differences were noted. Questions were then formulated to measure the noted differences. The questions were grouped into four objective tests which were labeled the Michigan M-Scales. Dr. Farquhar and his associates tested the Michigan M-Scales and found them to produce a significant increase in predictive efficiency of a test of mental ability in predicting grade point average. Procedures: In order to achieve the purpose, the Michigan M—Scales, a test to measure academic motivation, were administered to the beginning shorthand students in the. eight participating schools in the fall of 1964. The American Council on Education Psychological Examination for high school students, a test to measure mental ability, was also given. Academic grades were obtained from the students'permanent records in the school offices. Teacher grades in shorthand were collected at the end of the first semester and at the end of the second semester. A uniform shorthand examination for first semester was administered to the students in three schools at the end of the first semester and a uniform exami- nation for second semester was administered in all eight participating schools. The collected data were statistically analyzed by (a) an F test of homogeneity of variances between the pOpulation of 103 the present study and the female population of the Farquhar Study; (b) a "t” test to compare mean scores of the popula— tion of the present study with the mean score of the female population of the Farquhar Study; and (c) a correlation analysis to determine the linear relationship between the predictor variables of the present study and teacher grades for first and second semesters. Findings: (1) Zero-Order Correlations: Grade point average, ninth grade English grades, tenth grade English grades, total score and subscores of the American Council on Education Psychological Examination, total score of the Michigan M-Scales, and the Word Rating List, which is a subscore of the Michigan M-Scales, each correlated with achievement, as measured by teacher grades, of eleventh grade girls in first and second semester of Gregg Shorthand significantly greater than zero. (2) Multiple Order Correlation: Grade point average, English grades, mental ability, and motivation score produced a multiple correlation significantly greater than zero. (3) Increase of Precision of Prediction by Michigan M-Scales: The total score of the Michigan M-Scales and the subscores, Word Rating List (WRL), and Human Trait Inventory (HTI), increased the precision of prediction 104 produced by an estimate of mental ability in predicting achievement of eleventh grade girls in first semester Gregg Shorthand. The Word Rating List (WRL), a subscore of the Michigan M-Scales, correlated with achievement of eleventh grade girls in second semester Gregg Shorthand significantly greater than zero. The Human Trait Inventory (HTI) signifi— cantly increased the precision of prediction produced by the quantitative score, a subscore of the American Council on Education Psychological Examination, in predicting achievement of eleventh grade girls in second semester Gregg Shorthand. (4) Predictors for Shorthand I and II: Grade point average and English gradespredicted achievement of eleventh grade girls in first or second semester Gregg Shorthand better than did the combination of academic motivation test scores and an estimate of mental ability. The present study found that the best predictors for first semester shorthand grades, in order, were (a) grade point averages, (b) ninth grade English grades, (c) tenth grade English grades, and (d) mental ability; however, the best predictors for second semester shorthand grades, in order, were (a) Shorthand I teacher grades, (b) grade point average, (c) tenth grade English grades, and (d) mental ability. .1. .-. 1. '1 #4 a. .l A.‘ J. A: I: 105 Part II Conclusions The conclusions of the present study are as follows: (1) Analysis of the responses of the subjects of the present study to the Michigan M—Scales produced reliability estimates that compared favorably with reliability estimates of the Farquhar Study on the Preferred Job Characteristics Scale (PJCS); however, lower reliability estimates were found for the General Situational Choice Inventory (GSCI), for the Human Trait Inventory (HTI), and for the Word Rating List (WRL). Therefore, the conclusion was made that the short- hand students respond, in certain cases, in a manner more reliabile than the female pOpulation of the Farquhar Study, and in certain cases, in a manner less reliable than the female pOpulation of the Farquhar Study. (2) By comparing the distributions of the total score and subscores of the Michigan M-Scales of the population of the present study to the distributions of the total score and subscores of the Michigan M-Scales of the female pOpula— tion of the Farquhar Study, the conclusion was made that the distributions of scores were similar which indicates that the shorthand students' responses to the Michigan M-Scales were distributed in a manner similar to the female pOpulation of the Farquhar Study. (3) By comparing the mean score of the grade point 106 average of the pOpulation of the present study with the mean score of the grade point average of the female pOpulation of the Farquhar Study, the conclusion was made that shorthand students have a higher grade point average than the female pOpulation of the Farquhar Study which might be due to the courses taken in the ninth and tenth grades. This signifi- cant difference indicates that shorthand students in this study tend to have better academic records than did the female population of the Farquhar Study. (4) By comparing the mean scores of the total score and subscores of the Michigan M-Scales of the population of the present study with the mean scores of the total score and subscores of the Michigan M-Scales of the female population of the Farquhar Study, the conclusion was made that the shorthand students were not as academically motivated as the female population of the Farquhar Study. This significant difference in mean scores to the Michigan M-Scales might be due to the composition of the Michigan M-Scales. For example, a question in the Human Trait Inventory (HTI) is "I think I would like the work of a teacher."1 The shorthand student with educational goals leading to employment as a secretary would tend to answer this question in a manner that would indicate low academic 1William W. Farquhar, The_Michigan M-Scales (Form C - Female) "Human Trait Inventory" Item #128, (East Lansing: Michigan State University College of Education), 1961, p. 12. 107 motivation, which would not necessarily be true. (5) From the analysis of the combination of the Michigan M-Scales with the estimate of mental ability as measured by the American Council on Education Psychological Examination, the conclusion was made that academic motivation, as measured by the Michigan M-Scales, is a factor in learn- ingpin first semester Gregg Shorthand but is not a factor in second semester Gregg Shorthand. The Michigan M-Scales total score did increase thegprecision of prediction by the American Council on Education Psychological Examination, an estimate of mental ability, in predicting first semester shorthand achievement; however, the Michigan M-Scales total score did not produce a significant increase for second semester. The increase for first semester was not suffi- cient to warrant usage of the Michigan M—Scales in the present form by high school teachers. The lack of increase in predictive efficiency for second semester might be due to the manner of grading by the teachers of second semester Gregg Shorthand or to the homogeneity of the pOpulation of the second semester shorthand classes. (6) The Human Trait Inventory, a subscore of the Michigan M—Scales, was found to be the most valuable sub- score of the Michigan M—Scales inypredicting achievement in first semester shorthand; however, it did not correlate significantly different from zero for second semester. The researcher concluded that the personal characteristics of 1 -H v Pan 108 the student, as measured by the Human Trait Inventory, are factors of learning for the first semester of Gregg Shorthand but are not factors of predicting learning for second semester except when used in combination with a measure of quantitative ability. This difference in correlation between the Human Trait Inventory and Shorthand I Teacher Grades and Shorthand II Teacher Grades might be due to difference in learning load between Shorthand I and Shorthand II. Shorthand I is pre— dominantly memory work--a large amount of new material must be memorized and automatized--whereas in Shorthand II, memorization is reduced and is replaced by learning of rapid responses and by fusion of other skills such as typing, spelling, punctuation, and reading imperfect notes. As there is a larger amount of teacher-pupil interaction in Shorthand 1 than in Shorthand II, the Shorthand I teacher might become more aware of the personal characteristics as measured by the Human Trait Inventory than the teacher of Shorthand II where a majority of the activity is drill work in the fusion of skills to gain speed in taking dictation and to produce a usable copy. It was further concluded that the Human Trait Inventory and the Quantitative score, a subscore of the American Council on Education Psychological Examination, are measuring different abilities and can advantageously be combined in predicting shorthand achievement. (7) The subtest, Word Rating List (WRL), was found to 0 II.» aka _ 6..- vi». a: - Cu - 109 be the second most valuable subtest of the Michigan M-Scales in predicting achievement in first semester Gregg Shorthand and was found to be the most valuable subtest of the Michigan M-Scales in predicting achievement in second semester Gregg Shorthand. This subtest of the Michigan M-Scales was found to be a signifiCant predictor and a consistently significant predictor of shorthand achievement for both first and second semesters of Gregg Shorthand. It was concluded that the academic self—concept of the student is a factor in learning in the first and second semesters of Gregg Shorthand. (8) The Preferred Job Characteristics Scale and the General Situational Choice Inventory were found to be of little or no value in predicting shorthand achievement. The researcher concluded that these two subtests in their present form should not be used to predict shorthand achievement. (9) The aptitude measure, American Council on Education Psychological Examination for high school students, in combination with the motivation score as measured by the Michigan M-Scales did not predict shorthand achievement for the subjects in this study as well as the other predictor variables--grade point average and English grades. It was concluded that grade point average is the most accurate of the variables used in the present study followed by English grades; however, the correlation did not warrant the usage of grade point average as a reliable predictor of shorthand 110 achievement. (10) The correlation between theory subscore of the uniform Shorthand I examination and achievement of students in first and second semesters was suppressed by the presence of brief forms. The researcher concluded that knowledge of brief forms is not an aid in predicting achievement of second semester and is not an aid in predicting transcrip- tion ability as measured by the uniform examinations in the present study. The score of the brief form section of the test had a very small variance which might account for this suppressing effect. The brief form score when used with the theory score tended to suppress the relationship with Shorthand II teacher grades. The reason for this suppressing effect of the brief form score might be due to the short range of scores. The test had a high mean score and a low standard deviation which indicates that this particular subtest was not giving a high degree of discrimination. The relationship between the brief form subscore of the uniform Shorthand I examination and the transcription sub- score of the uniform Shorthand I examination was .3496, which indicates that there is a tendency for those who did well on the brief form test in Shorthand I to do well on the tran- scription subtest of the uniform Shorthand I examination; however, the relationship between the brief form subscore of the uniform Shorthand I examination and the transcription subscore of the uniform Shorthand II examination was .0064 TV :43 J. ll) ‘“ A . V . .J (I) ‘11 II) 111 which indicates that there is no relationship. The conclusion, based on the analysis of the 68 subjects who took the uniform Shorthand I examination, was made that, if second semester shorthand achievement is to be predicted, knowledge of brief forms should not be included as a predictor variable for second semester and that knowledge of brief forms is of little or no value in predicting second semester tran- scription skills. This difference in relationship between the brief form score and transcription score for first semester and the transcription score for second semester might be due to the difference in the vocabulary of the dictation. The Shorthand II dictation may contain more non- brief—form vocabulary; therefore, the automatization of outlines has not taken place and the student can not create these outlines as rapidly in the second semester of short- hand where the dictation reaches speeds of 80 to 100 words per minute or higher. (11) The predictor variables used in the present study correlated in some instances higher and in some instances lower with first semester teacher grades than with second semester teacher grades. Thus, the researcher concluded that different predictors should be utilized for different semesters of achievement; that is, if first semester short- hand achievement is to be predicted, a set of predictor variables could be utilized; whereas, if second semester shorthand achievement is to be predicted, another set of 112 predictor variables could be utilized. The objectives of the prediction should determine the criteria to be employed in making the prediction. For the reason that Shorthand I is of little value without Shorthand II, the conclusion was made that a criterion should be sought to predict achievement in Shorthand II (second Semester) instead of a criterion to predict achievement in Shorthand I (first semester). (12) The total score or subscores of the Michigan M—Scales did not correlate sufficiently high to warrant the use of the M-Scales in their present form by shorthand teachers as accurate predictors of achievement in either Shorthand I or Shorthand II. It should be noted, however, that the correlations were higher in Shorthand I than in Shorthand II. This difference in correlations between the Michigan M-Scales with Shorthand I and with Shorthand II indicates the need for different predictors of Shorthand II success than those used for predicting success in Shorthand I. M Recommendations The recommendations of the present study are as follows: (1) A subsequent study should be made to determine the value of the Michigan M-Scales in combination with more A recent measures of mental ability in predicting shorthand achievement in first semester and second semester Gregg 113 Shorthand. (2) A study should be made to determine whether the personality factors as measured by the Human Trait Inventory (HTI) and Word Rating List (WRL) are factors which successful secretaries possess. (3) Further studies should be made of the academic self-concept to refine the Word Rating List and to deter— mine the effect self—concept has on learning in various types of programs; such as, academic and vocational. (4) Further study of the General Situational Choice Inventory (GSCI) should be made to determine potential revisions to make it more useful in predicting achievement of a particular subject area. (5) Further study of the Preferred Job Characteristics Scale (PJCS) should be made to determine potential revisions to make it more useful in predicting achievement of a particu- lar subject area. (6) An instrument similar to the Michigan M-Scales should be created for use with adults and business college students to measure motivation in a vocationally oriented group. (7) Studies should be made to discover a device to more accurately predict achievement in transcription in shorthand. (8) Studies should be made of the effect knowledge of brief forms and theory has on achievement in transcription K\') (I) In (h n": 114 and speed in taking dictation. (9) A series of studies should be made to analyze the learning activities of theory, speed building, and tran- scription in shorthand. As the findings of the present study indicated, a predictor variable, in some instances, had a low correlation with shorthand achievement in one semester and had a high correlation in another semester; thus, it is recommended that the learning activities of each semester be thoroughly analyzed to determine what activity or activities are unique to that particular semester and what activity or activities are similar to the subsequent or preceding semester or semesters of shorthand. A means of measuring these activities could then be selected or created to measure these particular factors in order to predict transcription ability which is the ultimate goal in shorthand. BIBLI OGRAPHY 115 BIBLIOGRAPHY ON SHORTHAND Anderson, Ruth I. An Analysis and Classification of Research in Shorthand and Transcription. Unpublished doctoral dissertation, Indiana University, 1946. Bollinger, Edgar. "The Establishment and Implementation of Guiding Principles for Motivation in Teaching Shorthand,” National Business Education Quarterly, XXX (October, 1961), 11. . "The Establishment and Implementation of Guiding Principles for Motivation in Teaching Shorthand.” Unpublished Master's thesis, University of North Dakota, 1960. Breuch, E. Margaret. "An Analysis of the Dropouts in First Year Shorthand Classes.” Unpublished Master's thesis, Colorado State College of Education, 1948. Byers, Edward E. "Construction of Tests Predictive of Success in First-Year Shorthand.” Unpublished doctoral dissertation, Boston University, 1958. Cheney, Truman M. and Goodish, Naomi. "An Analysis Between Certain Variables and Achievement in Beginning Shorthand," Journal of Business Education, XXXVIII (May, 1963). 317-319- DiBona, Lucille J. ”Predicting Success in Shorthand," Journal of Business Education, XXXV (February, 1960), 213-214 0 Dodson, Mary H. "A Study in Typewriting and Shorthand Prognosis." Unpublished Master's thesis, University of Kentucky, 1943. Frink, Inez. "A Comprehensive Analysis and Synthesis of Research Findings and Thought Pertaining to Shorthand and Transcription, 1946—1957." Unpublished doctoral dissertation, Indiana University, 1961. Goodenow, Jean E. “A Review of Professional Literature Relating to Stenographic Prognosis.” Unpublished Master's thesis, State University of Iowa, 1948. Hutson, Billy T. ”Prognosis of Achievement in First-Year Gregg Shorthand Simplified." Unpublished Master's thesis, University of Tennessee, 1951. 116 117 , and Vincent, Nicholas M. "Motivation and Prognosis in Shorthand," Journal of Business Education, XXXIII (October, 1957), 29-31. Lee, Carole Dawn. ”A Study of the Effect of Reward and Punishment on Learning in Shorthand.” Unpublished Master's thesis, Brigham Young University, 1962. . "A Study of the Effect of Reward and Punishment on Learning in Shorthand," National Business Education Quarterly, XXXII (October, 1963), 28. Leslie, Louis A. Methods of Teaching Gregg Shorthand. New York: Gregg Division, McGraw-Hill Book Company, 1953- Morrison, Robert L. ”Prognosis of Shorthand Achievement of Watertown High School." Unpublished Master's thesis, Boston University, 1949. Osborne, Agnes E. ”Guidance Through Prognosis," National Business Education Quarterly, XIII (May, 1945), 19-26. . The Relationship Between Certain Psychological Tests and Shorthand Achievement. New York: Bureau of Publications, Teachers College, Columbia University, 1943. Pauk, Walter. "What's the Best Way To Predict Success in Shorthand?" Business Education World, XLIII (April. 1963). 7—8,34. Powell, Georgia Faye. ”An Analysis of Shorthand Dropouts at Ottawa Township High School," National Business Education Quarterly, XXXI (October, 1962), 52-53. . "An Analysis of Shorthand Dropouts at Ottawa Township High School." Unpublished Master's thesis, Illinois State Normal University, 1961. Rogers, H. W. ”Psychological Tests for Stenographers and Typists,” Journal of Applied Psychology, I (January, 1917), 268-274. Sanders, Celene Honeycutt. "A Study of the Relationship Between Certain Radford College Students' ACE Scores, Years of Shorthand in High School, and Achievement in Shorthand!‘ National Business Education Quarterly, XXXI (October, 1962), 58. Q} 118 Sherman, Marsdon A. "An Exploratory Study of Certain Factors in the Prognosis of Achievement in Gregg Shorthand." Unpublished Master's thesis, Leland Stanford Junior University, 1942. Rowe, Clyde E. "Significant Research in Shorthand," National Business EducationyQuarterly, VIII (Summer, 1940), 48. Tschider, Irene R. "A History of Selected Studies in Shorthand Prognosis from 1914 to 1960,” National Business Education Quarterly, XXX (October, 1961), 57. . "A History of Selected Studies in Shorthand Prognosis from 1914 to 1960.” Unpublished Master's thesis, University of North Dakota, 1960. Veon, Dorothy. The Relationship of Learning Factors Found in Certain Modern Foreign-Language Aptitude Tests to phe Predigtion of Shorthand Achievement in College. Delta Pi Epsilon Research Award Series. Stillwater, Oklahoma: Oklahoma Agricultural and Mechanical College, 1950. Whittle, Marie. "The Relationship Between Certain Variables and Achievement in Beginning Shorthand at the University of Texas,” National Business Education Quarterly, XXIX (October, 1960), 70—71. . "The Relationship Between Certain Variables and Achievement in Beginning Shorthand at the University of Texas." Unpublished Master's thesis, University of Texas, 1959. BIBLIOGRAPHY ON MOTIVATION Anderson, Jack O. "A Study of the Relationships Among Academic Motivation Level of Aspiration, Level of Expectation, and Gain or Loss in Achievement in a College Reading Improvement Situation." Unpublished doctoral dissertation, Michigan State University, 1962. Bindra, Dalbir. Motivation: A Systematic Reinterpretation. New York: The Ronald Press Company, 1959. Bonnell, Jane A. "The Use of Depth Interviewing to EXplore the Motivational Factors Underlying Achievement of Eleventh Grade High School Girls." Unpublished doctoral dissertation, Michigan State University, 1962. 119 Burros 4th Yearbook of Mental Measurements (Highland Park, New Jersey: The Gyphon Press, 1949). Burros 5th Yearbook of Mental Measurements (Highland Park, New Jersey: The Gyphon Press, 1959). Edwards, Allen L. Statistical Methods for the Behavior Sciences. New York: Holt, Rinehart and Winston, 1961. Guilford, James P. Fundamental Statistics in Psychology and Education. New York: McGraw-Hill Book Company, 1956. Farquhar, William W. Motivation Factors Related to Academic Achievement. Final Report of COOperative Research Project No. 846 in cooperation with U. S. Office of Education and Michigan State University, East Lansing, Michigan, 1963. Green, Robert L. ”The Predictive Efficiency and Factored Dimensions of the Michigan M-Scales for Eleventh Grade Negro Students - An Exploratory Study.” Unpublished doctoral dissertation, Michigan State University, 1962. Guthrie, John A. ”A Study of Male Academic Achievement Motivation in a Self-Economy Context." Unpublished doctoral dissertation, Michigan State University, 1961. Johnson, Van. C. "An Assessment of the Motivation Factor in the Estimation of Academic Achievement of Eleventh Grade Indian Students and the Factored Dimensions of the M—Scales, An Exploratory Study." Unpublished doctoral dissertation, Michigan State University, 1963. Kagan, Jerome, and Lesser, Gerald S. Contemporary Issues in Thematic Apperceptive Methods. Springfield, Illinois: Charles C. Thomas, Publishers, 1961. Kipfmueller, Mark K. "The Predictibility and Factored Dimensions of the M-Scales for Eleventh Grade Parochial School Students.” Unpublished doctoral dissertation, Michigan State University, 1963. Madsen, K. B. Theories of Motivation. Copenhagen, Denmark: Munksgaard, 1959. McClelland, David C., et al. The Achievement Motive. New York: Appleton-Century-Crofts, Inc., 1953. 120 McDonald, Keith H. ”An Investigation into the Relationship of Socio-Economic Status to an Objective Measure of Motivation-—The Michigan M-Scales.” Unpublished doctoral dissertation, Michigan State University, 1962. Payne, David A. "A Dimension Analysis of the Academic Self- Concepts of Eleventh Grade Under- and Overachieving Students.” Unpublished doctoral dissertation, Michigan State University, 1961. Pierce, James V. "The Educational Motivation Patterns of Superior Students Who Do and Do Not Achieve in High School." Research report performed pursuant to a contract with U. S. Office of Education. University of Chicago, 1959. Taylor, Ronald G. ”Personality Factors Associated with Eleventh Grade Male and Female Discrepant Achievement." Unpublished doctoral dissertation, Michigan State University, 1962. Thorpe, Marion D. ”The Factored Dimension of an Objective Inventory of Academic Motivation Based on Eleventh Grade Male Over- and Underachievers.” Unpublished doc- toral dissertation, Michigan State University, 1961. Thweatt, Roger Clay. "The Development and Validation of an F Scale for an Objective Test Battery on Motivation." Unpublished doctoral dissertation, Michigan State University, 1961. APPENDIX 121 N I O u, y: I: \ AA “2" " 1‘ 3‘ 'v I 'LLINKAX“;‘III ‘x'i"~'"."""' Iu‘.\.‘.I\I.I ‘ 3 I; £3 I-L Eu is; salys Have: 1 113V! Ecor higa and}. Tue ; he 122 TABLE I FACSIMILE OF LETTER TO SCHOOL ADMINISTRATORS Mr. XXXXX XXXXXXXXXX 1317 Loraine XXXXXXXXXX Public Schools Lansing, Michigan XXXXXXXXXX, Michigan July 13, 1964 Dear Mr. XXXXXXXXXX: For my doctoral research project at Michigan State University, I am analysing the value of a motivation test in predicting achievement of eleventh grade girls in first and second semester Gregg Shorthand. May I have your permission to collect the necessary data from the student records and administer two class periods of tests in the XXXXXXXXXX High School of XXXXXXXXXX, Michigan, to be used as primary data for the study? The items needed from the students' permanent records are as follows: 1) ninth and tenth grade scholastic averages 2) tenth grade English grades 3) ninth grade English grades. The tests I would like to administer are as follows: 1) American Council on Education Psychological Examination for high school students-~time 50 minutes. 2) Motivation Battery created by Dr. Farquhar of the Michigan State University faculty-—time 45 minutes. I will need from the teacher the grades at the end of the first semester of Gregg Shorthand and the grades at the end of the second semester of Gregg Shorthand as well as general information on standards. All information will be held in strict confidence and no school, students, or teachers will be identified in the study. This study is being directed by Dr. Helen H. Green, Professor of Business Education, Michigan State University. Your permitting me to obtain the above information from the XXXXXXXXXX High School will make my study possible. I will look forward to receiving your reply regarding my request. Sincerely yours, /s/ Leonard J. Varah Leonard J. Varah : a Ti 123 TABLE II FREQUENT REASONS FOR DROPPING SHORTHAND AS CHECKED BY FORTY-FIVE STUDENTSa Rank Reason Frequency 1 I could not keep up in taking dictation 20 2 I did not work hard enough 18 3 I did not have enough time to study shorthand because I had to con- centrate on other subjects 17 4 I changed my plans about what I want to do when I leave school 16 5 There wasn't much likelihood of using a shorthand on the job 15 6 The class went too fast; I could not keep up with the rest of the class 14 6 I didn't get started right lb 6 My teacher recommended that I drOp shorthand 14 7 The course was too difficult; I could not understand it 13 8 I found it difficult to read shorthand 12 8 I found it difficult to write shorthand 12 9 My homeroom teacher advised me to drop shorthand 11 10 I could not find room for shorthand in my college preparatory schedule 10 aGeorgia Faye Powell, "An Analysis of Shorthand DrOpouts at Ottawa Township High School" (unpublished Master's thesis, Illinois State Normal University, 1961), p. 71. 124 .mxpma momsomme 0p pmme downphosm HmmSoHEHmo Hoonom Swan mama mmum. connobmo .mxsma mHoSommB 0b pmoe mcaompmanaaom hHmm mdbo Hoosom swag :dma mmm. w©w¢mH>mQ .pmma wcacammq dcmspnozm amazoaahmo on wcadmma womanhonm owmaaoo mama mmau. ocom> .H UQmSpHonm 0p nomamz smacmm Hoo£om Swfin omma mm. mmHmCmmHU .UsmspHOSm Ca mmomsw Honomme Hoosom smfin Hmma mmd. boompsm .mmomhw osm£ppo£m nonsmoo define mmeHoo Hmma wmmm. wmammmnoh .mmomnw nonommp . nmpmmsmm Hanna on mfipo Hooaom awn: Hmma mm. *emmnmaaomam .mpwmp pcma um>oa£om com mmowpw Honomoe Hoonom swan mmma on. manomam mxnmemm Hm>mq open Coapmamhhou pogohmmmmm Asmhm> .h camcomq mp omaaoeoo mmv mZOHB4UHBmm>ZH mDOH>mmm 2H Dmsmommm Bzmzm>mHmo¢ Qz¢m9momm 92¢ mEQDmmm EmMB MozmoHQQMBZH zmm3Emm ZOHEanb .mammSp m.nmpmmz dogmaaboazsv :.omamaamaam comnpnosm mwmnw Hmmwupmnam CH pnmambmH£o¢ mo wamoswonm: .nompsm .9 maafimb .eamumam .aa .Aomma .anmsnpmmv >xxx soapmosdm mmmnamsm do Hmsnsoh :.©zm£pno£m EH mmmoosm wQHpOHuon: .mCOmam .h mHHfiosqm :mpmhooom HOOSom me m.m .mocmam nonommp comb HmoaQSOQb Hmma ©mpoa©mpd= Hmmfih mo mwmnm>¢ Hoonom swan mmma mm. mmmmonmm mxnmamm Hoboq open Coaumamnnoo noncommmmm Admssapfloov HHH mqm¢B 127 .mo5pm SH momma camHoQO opmoaona mambaoz .moopm monopmonuo “adopm Compomum whoopm Moanmnm “hoopm comnmos¢n¢ .soabmahomnfi mo condom mpmodonfi mam: wsasoaaom newness one whopme* .Amemfl .apamnm>acp manasaoo .mwmaaoo mongoose .noabmoaabom mo ommnom "390% zozv unmambmH£o¢ ocmspnosm one momma Hmoawoaosohmm Campmmo comzpmm QHSmQOHpmHom one .mcnopmo .m mosw< U .Aomma .mmanmm oamz< nonmmmom noaamom Hm spawm‘wmwmaaoo Hmoasmsomz one HmHSpHooahw¢ maonmaxo «maosmaxo .HmpmsaaflpMP owmaaoo EH psoao>mano< downpnosm mo noapoaoonm one on momma moopapm< ammowmmq Inwampom :nmdoz Samphmo Ca osoom mhoocmm wmasnmmq no Qasmzoapmfimm one .Qoo> hnpOHoDo Admzflapfloov HHH mqm¢H 128 .mmompw downpnonm nonwoaow oncomm on mmomam Smaawsm Hoo£om Swan mama mm. .mmdmaw downpaogm nopmmamm panda on mmemnm smaamcm Hoosom smug name on. osomsm m.mxnsom.o Hoonom swan :sma 5mm. ommmQ .mmomaw enmSpMOSm ampnmso snags on soapfimoasoo nmfiamsm mwmaaoo Hmma memm. osammmsoe .mmompw one: IpHOSm 0p mmwm9m>m nmaawsm Hoo£om swan Nmmfi Hume. mmamxoflozmppom .mmmao mo mxomz NH Hosam wnfinoo pmmp psmam>mazom UQmQPMOSm cu mmthw Swaamcm Hoosom swag Hmma was. Dmompom .Hopwmaom Spooog ego mo one mnb pm umop noabmahom .sme on mmemnw amfiawsm Hooaom smug smmfl Hm. manomfin mxnmamm Hm>mq mama Soapmaohhoo Honohmmmom Asmnm> .h chanced an coaamaoo may mZOHB4UHBmm>ZH mDOH>mmm 2H QMBmommm Bzmzm>mHmU< QZH mqm< Hoosom swan mama mam. mau¢omaho£m %£anmosmpm no mason ozu ca mxsma on pmme nmaawsm .Qooo mwoaaoo mama mud. mwm¢nmnsno mthamm Hobmq comm soapmamnhoo Honoammmmm AdeQHonov >H mqm¢8 O 3 1 .eamumam .ao .Aomma .anmsspmmv >xxx sodpmosem wmmsflmom mo Hmnnoow :.osm£pnosm SH mmmooom mnapoaomhm: .mSOmaQ .h mHHHooqm .wmomnw ocmnbnonm nmpmmamm cocoon 0p oncom pmmp nmaawsm modem canon .mmemnw smfiawsm wemnm spams Hooaom ems: emma **mmm. .wmdmsw UQmeHOSm HmpmmBmm osoomm on smfiamam memnw epsm>mfim Hoogom awn: emma mamm. .mmcmhw csmzpnosm Hopmmamm uncomm 0p whoom pmmp macaosgom Hoo£om smHS :mma mow. .mmdmnw osmSpHOSw Monmoamm esoomm on emdamsm memes spams Hoogom swan smaa Nme. sweeppmawmm .wmcmnw nmfiamnm on mmomnw downpnonm nmpmmamm oncomm Hoosom swan mmma mm. .mmomaw Smaawom on mmdmnw downpnosm Hmpmmamm pmnam Hoosom swan mmma mm. .mwmmo Smaamsm on mmomnw downpnosm snowmamm oncomm H00£om swan mmma mm. .mwmwo Smfiawsm 0p mmomnw USmSphonm Hmpmmamw pmnam Hoosom swan mmma on. oammmwwam mxnmamm Hm>mq open Scapmamhnoo nmsonmmmmm Afimdflapcoov >H mqm¢a 131 .noapmamhnoo oHQHpHsz** .aaSpm seanmum “meson nompmdsasb .mawosp m.nmpmmz dogmaansmsov :.omamaaoafim. downpnozm wwmho nmmwnpmham SH psmaoboH£o< mo mamoswoame .nompom .e kHHHmD Admosapsoov >H mqm¢8 132 .mmdmhw UQmSphosm osmsoofiQSm Hoosom swam Hoosom awn: mead ms. .mmomnm downpnosm on mpomfibsm Hoosom swam Hooaom awn: memfi ea. Hsmam5aoxo «mu aoosom swag emma as. ommmmnaammfia wxnmamm Hm>mq open moapmamhpoo nononmmmmm Asmnm> .h osmsomq an dmaagaoo may mZOHBduHBmm>ZH mDOH>mmm ZH DMBmommm Bzmzm>mHmo< QZ¢mBmomm Qz< Bzmzm>mHmo¢ DHSMQ mqm<9 133 .sespm xefinmum “sespm somnmosanb .mdmmnp m.hmpmm2 dogmdansonov ..eoacfifiaaam semaphonm hams-pmnnm ca pzoso>mdsoe no mfimosmonm: .compsm .a sHHHmm AdeCHonov > mam HH exogenoem HH oooeonosm H ecospnogm H oomnonoem nooofioonm Aee u ze mmoemo mmm6 moaoHommm zomzamm mZoHa HH ososonogm HH ooosononm H ozononoem H osoeononm nopoHooHH All» Al Admssfipfloov Hxx mqm\o oa~o O\\n C'ko N>F4(O 23. 24. 25. harm appears themselves perplexing exportation self-made (53. becoming cold furniture circulation function surest quiet postponed $500,000 justification qualify consumed ounces terminations conveniently entertain misinform amounts 12 18 24 3O 36 42 48 54 l min. 12 18 24 30 36 42 48 54 2 min. 12 18 24 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. kingdom afternoon patiently depending investment exceedingly encourages skates ultimate family skillful township creative whenever 3 o'clock checks subdivide seriously earth musical supervisor let us if you want at a loss to know 5 min. 30 36 42 48 54 3 min. 6 12 18 24 30 36 42 48 54 4 min. 6 12 18 24 30 36 42 48 54 stOp! 166 FIRST SEMESTER SHORTHAND EXAMINATION Brief form Section Words to be dictated every 6 seconds 1.4 O\OCD\'JO\U\-{—‘\JON 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. after send bill never advertise newspaper consider time glad enclose to but purpose am for purchase correspond have govern street correct important part could Object 12 18 24 30 36 42 48 54 l min. 12 18 24 30 36 42 48 54 2 min. 12 18 24 26. 27. 28. 29. 30. 31. 32. 33- 34. 35- 36. 37- 38. 39- 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. success are world conclude refer every one right and organize advantage soon company satisfy character put under what was yet long go please they ever 5 min. 30 36 42 48 54 3 min. 6 12 18 24 30 36 42 48 54 4 min. 6 12 18 24 30 36 42 48 54 stOp! 167 TABLE XXVIII UNIFORM EXAMINATION FOR SECOND SEMESTER SHORTHAND DIRECTIONS FOR THE TRANSCRIPTION SECTION: Attached are speed takes at 60, 80, 100, and 120 WPM. These are three minute takes. After the take is dictated, the students have 15 minutes for transcribing notes for gggh speed. At the end of the 15 minute period say: STOP. (GO ON TO THE NEXT SPEED TAKE. YOU HAVE 15 MINUTES TO TRANSCRIBE YOUR NOTES FOR THIS SPEED. BEGIN. If you wish to give the students more time for transcribing, please have the students prepare a carbon copy of their transcript. At the end of the prescribed 15-minute period, collect the original, leaving the carbon copy with the student so that she may continue her transcription on the carbon COpy. Note: The four speeds may be given at any time with any number dictated on a particular day. However, the transcript for each speed should be made during the same class period that the dictation is given. 168 60 WPM DICTATION Dear Mr. Kelly: If you are tempted to buy a car with little or nothing down / and take three or four years to pay, be careful. These are not easy terms at all. They are, / in fact, very harsh terms. The cheapest way to buy a car is to pay cash for it. In / that way you have no interest charges to pay. The next Cheapest way to buy a car (1) is to pay as much down as you can afford and the rest as soon as possible. It / is true, Of course, that you can arrange payments that are too high for comfort; and, as a / result, your car may be a burden instead of a pleasure. On the other hand, if / you pay less than you can comfortably afford, you are paying for financing service (2) that you do not need. Consequently, it is wise to arrange your payments to suit / your income. Pay as much down as you can, and pay the rest as soon as you can. If you / are planning to purchase a car on time, why not ask one of our finance experts to / advise you on the terms of payment that will be best for your budget. Sincerely yours, (3) 169 80 WPM DICTATION Dear Mr. Jones: DO you waste time every month standing in line to pay taxes and other bills? Do you sometimes / neglect to get a receipt for bills you have paid? DO you risk loss or theft of money by carrying large sums of / cash with you? You can avoid these unnecessary risks and nuisances by Opening a special checking account. / With one Of our special checking accounts, you do not have to waste time standing in line to pay bills. You just write (1) out your check in the comfort of your home and send it through the mails. It is as simple as that. Your canceled checks / represent the best kind of proof that you have paid your bills. By using our special checks, you will not have to carry / a large sum of money with you and risk losing it or having it stolen from you. You may find it easier / to live within your budget, too, because your checkbook shows where the money goes. It helps you to find out whether you (2) are spending too much money on certain items. It makes good sense to Open a special checking account, / particularly in view Of the fact that it costs so little. The cost of each check is 10 cents, plus a fee of 25 / cents a month for maintenance. You do not have to keep a minimum balance at any time. Why not open / a special checking account today and start profiting from its many advantages. Very cordially yours, (3) 170 100 WPM DICTATION Dear Mr. Green: Thousands of people think that they will never have to go to court. But it can happen to you. For example, the other fellow / drives his car like a maniac, while you drive carefully. There is a crash, and you have a lawsuit on your hands. You suddenly inherit / some money, and unknown relatives appear from all over the place. The next thing you know, you are consulting lawyers and you are in court. You / write a book, and then you find yourself in court because you are accused of stealing someone else's material. You invent something, and you (1) are in trouble because someone is taking you to court for using his ideas. You leave something out of a contract or make some /_ perfectly innocent mistake, and you are in a court action. "A Day in Court,” by Harry Shaw, is the layman's guide to the law. Even if you / never tangle with the law, you will have a fine time reading this witty book, which tells clearly and cheerfully how to stay out of court and what / to do if you happen to get there. Whether you are a plaintiff or a defendant, this book will show you how to find a lawyer, how to act (2) in the courtroom, how to act on the witness stand, and how to help your lawyer win your case. This book will Show you exactly what to do, from the / time your trial begins to the moment you are awaiting the decision of the jury. 171 A copy of this book will be sent to you for seven / days' free reading. If you are not completely satisfied, send it back and you will owe us nothing. Otherwise, remit $3.50 / plus a few cents for postage. A stamped and self- addressed enveIOpe is enclosed for your convenience in sending us your order. Cordially yours, (3) 172 120 WPM DICTATION Dear Mr. Gates: Suppose you were in a mood to look for a new car. It is the middle of the afternoon and the temperature is about 95 degrees. You drive / past a new car display that catches your eye, but there is no Shade. Would the idea of getting out in the hot sun appeal to you? Farther down the street, however, there is / another display of new cars; but there is a difference there. This outdoor display is completely under cover. The cars can be looked at and bought in cool, comfortable / shade. Is there any question where you would stOp to look and trade? Wouldn't you stop where you could be cool and comfortable? That is the reason why more and more progressive (1) car dealers all over the country are covering their outdoor display area. They do not want to lose good prospects to their competitor down the street. If you want / to be sure that you do not lose car prospects to your competitors who already have covered outdoor displays, you will be interested in learning about Smith Carports. / For the first time in America a modern carport is available at an amazingly low cost. The reason that our carport can be Offered at such a / low price is that it is made on a production line. Consequently, we can sell it for almost 35 per cent less than our competitors can. The enclosed photos give you (2) an idea of the kind of 173 display you can have with our carports. Show these photos to your salesmen. See if they do not all agree that they can make more and better trades / with prospects who keep cool and comfortable. You can easily find out, without any Obligation on your part, how little it will cost you to have the finest car / display in town. Indicate on the enclosed card the number Of cars you want to cover, and we will give you the cost by return mail. Just fill out the card and drop it in / the mail today. You can pay for your covered display as it earns extra profits for you. If you prefer to do this, please indicate that fact on the reply card. Sincerely yours, (3) 174 SHORTHAND II — BRIEF FORM TEST NAME Part I: Write the longhand word for each of the following shorthand outlines in the space provided at the right of the outline. H C“ ‘31 RRfiMYYI) C” new 3133 O\OCD\IO\U\-PKA)I\) H H H H N H 4:- H U1 H O\ I..I fl H (I) H \O 20. 21. 22. 23. 24. 25. TURN PAGE AND CONTINUE TEST. 175 SHORTHAND II - BRIEF FORM TEST NAME Part II: Write the shorthand outline for each of the following words in the space provided at the right of the longhand word. H gentlemen 0V8 I' government thank which advantage progress question suggest O\OCD\10\U‘l-{-‘-‘\A)N responsible ll. thing 12. Mr. 13. Opportunity 14. subject 15. advantage 16. importance 1?. quantity 18. particular l9. recognize 20. idea 21. satisfactory 22. organize 23. publish 24. time 25. year STOP! DO NOT TURN THIS PAGE UNTIL YOU ARE TOLD TO DO SO. 176 SHORTHAND II — THEORY TEST NAME Part I: Write the longhand word for each of the following shorthand outlines in the space provided at the right of the outline. 1. 2, Cu. / 3. U 4 . e 5. ,2( 6. /7 7. 8- 3:: 9. 11. 12. 13. 14. ""f’ 15. “1(7 16. 17. B 18. 19. (1%:EEHJKr 2O. " 21. M/ 22. W 23. W 24. g\ 25' CT‘\__,/ TURN PAGE AND CONTINUE TEST. 177 SHORTHAND II - THEORY TEST NAME Part II: Write the shorthand outline for each of the following words in the space provided at the right of the longhand word. H inaccurate surest daily deficient diagram assist helpful forwarded duplicated township ll. join 12. managed O\OCD\10\U\-PKAJN l3. prOpose 14. refuse 15. threats 16. eXpire 17. $500 18. August 19. article 20. dependable 21. science 22. appreciated 23. underpaid 24. Washington 25. addition STOP! DO NOT TURN THIS PAGE UNTIL YOU ARE TOLD TO DO SO. 178 TABLE XXIX EXPLANATION FOR GENERALIZABILITY In order to Obtain generalizability for the results Of the present study, no analysis was made of the teaching methods, motivation déyices used in the classroom, equip- ment for teaching, objectivesof the courses for each teacher, standards of the course, standards for grades by the teachers, professional preparation of the teachers, or the teacher's personality. Such analysis would have created a series of subgroups--one subgroup for each teacher—- which may have created homogeneous groups in certain schools, if not in all schools. Homogeneity would have rendered the results of the Michigan M-Scales, the test that was analyzed in the present study, useless for the reason that an important prerequisite of the Michigan M-Scales is a heterogeneous pOpulation.