MSU LIBRARIES “ RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES wiI] be charged if book is returned after the date stamped below. A STUDY OF THE RELATIONSHIPS BETWEEN READING AND PERSONALITY AND WAIS SCALES AT AN OPEN DOOR COMMUNITY COLLEGE By Ear] L. Menary, Jr. A DISSERTATION Submitted to Michigan State University in partiaI fquiIIment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Administration and Curricqum 1984 COpyright by EARL L. MENARY, JR. I98A To Walter E. Lockhart: distinguished teacher. scholar. and researcher. whose inquisitive mind explored many diverse subjects. the was a true philosopher to the end. I am fortunate indeed to have known Halter as an office partner. fellow instructor. research collaborator. and fellow Christian. Most of all. I am glad he was my friend. ACKNOWLEDGMENTS I first want to thank God. for without Him nothing would be possible. A special indebtedness is owed to my chairperson. Dr. Lois Bader. Without her support and guidance this paper would not have been written. I would also like to acknowledge the members of my committee. Drs. Ben Bohnhorst. Lawrence Lezotte. and Eugene Parnell. for their suggestions and guidance. I acknowledge the late Walter Lockhart for his skillful assist- ance in gathering the data for this study. Too often those who have inspired us and molded our character are overlooked when kudos are handed out. I would like to recognize two such people. Lawrence Ordowski and Edward McNally. Thank you for your knowledge. guidance. and friendship. I would also like to recognize Madalyn Kuhn. who typed the manuscript from handwritten copy. and Sue Cooley. who put it into final form. Finally. a most heartfelt thanks to my wife. Dawn. and my children. Tara. James. John. and Sarah. whose love and understanding were so necessary for the completion of such a project. ABSTRACT A STUDY OF THE RELATIONSHIPS BETWEEN READING AND PERSONALITY AND NAIS SCALES AT AN OPEN DOOR COMMUNITY COLLEGE By Earl L. Menary. Jr. The purpose of this study was to obtain and analyze data con- cerning the relationship between reading and personality and intelli- gence scores of a group of open door community college students. The writer also explored the development of a personality and achieve- ment profile that might help clinicians understand the strengths and weaknesses of the students with whom they work. Data were obtained on 127 open door community college students residing in the college dis- trict; they were Caucasian and predominantly middle class. 'The Nelson Denny Reading Test (Form C) was used to obtain students' total reading scores. The Clinical Analysis Questionnaire and the Wechsler Adult Intelligence Scale were used to measure personality and intelligence. respectively. The subjects were arranged into comparison groups based on their total reading achievement on the Nelson-Denny. Those obtaining scores one-quarter standard deviation above the mean (approximately the top 40%) were designated the better-reading students; those scoring Earl L. Menary. Jr. one-quarter standard deviation below the mean were designated the poorer-reading students. Correlations between total reading achieve- ment and performance on the CAD and between total reading achievement and WAIS subscale scores were obtained using a product-moment correla- tion procedure. Two-way analyses of variance were used in comparing better-reading and poorer-reading students. Results were as follows: Statistically significant relationships existed between reading and personality for the total group; the relationships were enhanced by intrasexual comparisons. Statistically significant relationships between male total reading achievement and personality characteristics indicated that better-reading males were more intelligent.lnore imagi- native. more liberal and experimental in their thinking. decisive. and resourceful. They had good vocabularies and reasoning abilities. Sta- tistically significant relationships between female total reading achievement and personality characteristics indicated that better- reading females were more intelligent. assertive and headstrong. quick and alert. secure and self-confident. liberal and experimental in their thinking. decisive and resourceful. restless and excitement seeking. had complacent attitudes toward anti-social behavior. were not hurt by criticism. and lacked self-insight. Comparison of better and poorer readers' scores on the NAIS scales indicated that a profile analysis did not yield useful information. The only information evident was a verbal deficiency. which is assessed more easily by means other than a NAIS test. TABLE OF CONTENTS LIST OF TMLES O I O I O O O O O I O O O O I O O C O O O 0 Chapter I 0 INTRODUCTION 0 O O O I O O O O O O I 0 O O I I O 0 Statement of the Problem and Purpose of the Study . . Hypotheses . . . . . . . Definitions of Terms . . Instrumentation . . . . Data Analysis . . . . . . . Organization of Subsequent Ch II. REVIEW OF RELATED LITERATURE . Studies Comparing Personality and Reading Need for the Study ACh1evment O O O O O I I O O O O O O O O O 0 Studies at the Elementary and Secondary Levels Studies at the College Level The Relationship Between Reading Achievement and Subscale Patterns . Wechsler Intelligence Test Summary III. METHODOLOGY OF THE STUDY . . . Introduction . . . Population . . . . Instrumentation . Data Collection . Comparison Groups Hypotheses . . . . Data Analysis . . Summary IV. PRESENTATION AND ANALYSIS OF DATA . . Results of Hypothesis Testing Hypothesis 1 . . . . . . . Hypothesis 2 . . . . . . . iv Page vi —0 \O ooqqosmxs... \JNO‘O 28 35 36 36 36 46 47 49 50 SO 52 52 52 53 Page Hypothesis 3 . . . . . . . . . . . . . . . . . . . . 54 Hypothesis 4 . . . . . . . . . . . . . . . . . . . 55 Hypothesis 5 . . . . . . . . . . . . . . . . . . . 55 Hypothesis 6 . . ..... . . . . . . . . . . . . . S7 Hypothesis 7 . . . . . . . . . . . . . . . . . . . . 58 Hypothesis 8 . . . . . . . . . . . . . . . . . . . . 63 Hypothesis 9 . . . . . . . . . . . . . . . . . . . . 64 Hypothesis l0 . . . . . . . . . . . . . . . . . . . 65 Hypothesis ll . . . . . . . . . . . . . . . . . . . 66 Hypothesis l2 . . . . . . . . . . . . . . . . . . . 67 Summary . . . . . . . . . . . . . . . . . . . . . . . 80 82 83 Summary . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . Comparison of the Present Study With Other Studies Discussion and Recommendations . . . . . . . . . . . . . . ‘O —J 96 APPENDIX 0 O O O O O O O O 0 O O O O O O O O O O O O O O O O O 0 l0] BIBLImRAH-‘Y O O O O O O O O O I O O O O O O O O O O O O O O O O '03 LIST OF TABLES Table Page 1. Patterns of WISC Subtest Scores Among Poor Readers . . . . 29 2. Means and Standard Deviations for Upper and Lower College Readers on the Nelson-Denny and the WAIS-FIG . . 48 3. Correlation of Total Reading Achievement With Personality: Combined Group . . . . . . . . . . . . . . S3 4. Correlation of Total Reading Achievement With Personality: Males . . . . . . . . . . . . . . . . . . . 54 5. Comparison of Total Reading Achievement With Personality: Females . . . . . . . . . . . . . . . . . . SS 6. Correlation of Group Total Reading Achievement With WAIS SUbSCII as I I I I I I I I I I I I I I I I I I I I I I I 56 7. Correlation of Males' Total Reading Achievement With WAIS SUbsca1 es I I I I I I I I I I I I I I I I I I I I I 57 8. Correlation of Females' Total Reading Achievement With WAIS SUDsca1 es I I I I I I I I I I I I I I I I I I I I I 58 9. Correlation of Group Total Reading Achievement and Personality. Before and After Controlling for ID . . . . 60 lo. Correlation of Male Total Reading Achievement and Personality. Before and After Controlling for ID . . . . 61 ll. Correlation of Female Total Reading Achievement and Personality. Before and After Controlling for ID . . . . 62 l2. Mean Sten Scores and Significance Levels for Better- Reading and Poorer-Reading Males on the CAD . . . . . . 64 I3. Sten Scores and Significance Levels for Better-Reading and Poorer-Reading Females on the CAD . . . . . . . . . 65 14. Significant Differences Between Better-Reading and Poorer-Reading Males on the WAIS . . . . . . . . . . . . 66 vi 15. 16. 17I l8. l9. 20. 21. Significant Differences Between Better-Reading and Poorer-Reading Females on the WAIS . . . . . . . . . . Significant Relationships Between CAD Scales and WAIS Subscales for the Total Group . . . . . . . . . . . . Significant Relationships Between CAD Scales and WAIS Subscales for Males . . . . . . . . . . . . . . . . . Significant Relationships Between CAO Scales and WAIS Subscales for Females . . . . . . . . . . . . . . . . Statistically Significant Relationships Between Males' Total Reading Achievement and Personality Char- acteristics . . . . . . . . . . . . . . . . . . . . . Statistically Significant Relationships Between Females' Total Reading Achievement and Personality Char- acteristics . . . . . . . . . . . . . . . . . . . . . WAIS Scale Scores for Combined Group. Males. and Females vii Page 67 68 73 76 86 87 89 CHAPTER I INTRODUCTION W The relationship that personality has to reading and to intelligence requires further exploration and extension. Previous work in this area needs to be replicated with similar instruments and design. In this way. theoretical insights can be gained. from which further studies can be generated. As a leading authority on reading wrote. “Almost everyone who has worked with or written about retarded readers has reacted to the personality problems that complicate treat- ment of the disability" (Spache. l976. p. 236). Many textbooks on reading diagnosis and remediation lend truth to this statement by including at least one chapter on the relationship between reading and personality. Although the relationship of reading to personality has been discussed at length. the topic entails a number of problems. One of the major problems is lack of replication of research studies investigating the relationship of reading to personality. Spache (l976) stated that the role of personality in the act of reading is difficult to assess "because of their bases in varying theories of personality and its measure" (p. 240). Also. each study seems to have unique characteristics. depending to some extent on the measuring devices used in the research (Ekwall. l966. p. 240). 'Throughout the literature. little commonality exists in either the measuring devices used. or in the age or grade level assessed. 0f 14 studies cited by Spache. only three used the same instrument--the Rorschach--but here the similarity ended. The research design of each study differed; thus the chance of similar findings was negated. Another problem. though of lesser degree. is that most studies investigating the relationship of reading to personality have dealt with the elementary-school population. Few such studies have been conducted at the middle school. secondary school. or college levels. Profiles of poor and good readers at all grade levels are needed if practical value is to result from the research. To study poor readers without comparing them with their good-reading counterparts will not help detect differences in the personality factors of readers. It will only measure a poor reader against the norms of the personality test used. Such information is useful in describing the poor reader. but it does not aid in identifying differences that may be involved as either incentives or deterrents in the act of reading. In addition to the foregoing problems. determining which research instrument is most appropriate for the age and grade level of the intended population is an important consideration. The ease of instrument use and interpretation of findings for other researchers should also be considered when selecting an instrument. In a study conducted at the community college level. Lockhart and Menary (1979) explored the use of the Cattell l6 Personality Factor Test (16 PF) in identifying differences between good and poor readers and its ease of interpretation. The researchers were satisfied that the instrument fulfilled these requirements. However. the inability to extrapolate from the 16 PF to the Minnesota Multi- phasic Personality Inventory UHMPI) led Menary and Lockhart to conduct a study with Cattell's Clinical Analysis Questionnaire. an extension of the 16 PF that measures abnormal traits. similar to the measurements obtained by the MMPT. The attempts to discover scales that can extend the knowledge of personality traits of good and poor readers at the college level is continued in the present study. in conjunction with an attempt to develop the IQ profiles of good and poor readers and to compare them. Like the personality studies. research on the 10 profiles of poor readers has mainly compared retarded readers as a group with the scales used. rather than with a sample of good readers. 'This method has led to controversy over what the researchers have labeled significant differences. Analyses of how poor readers perform on individual IO.tests such as the Wechsler Adult Intelligence Scale (WAIS) or the Wechsler Intelligence Scale Combined (WISC) do not indicate how the poor reader differs from the good reader on a given subtest. nor do they give a clear indication of whether the intragroup differences on the subtests are significant. Spache (1976) stated that The bases of these comparisons vary from one writer to another: some comparing mean scores of matched or unmatched groups: some allowing for the probable errors of the estimate. while others ignore these; some considering almost any variation among sub-test scores with his own average as a correction. In addition to the aforementioned difficulties in determining what scales seem to be associated with reading. or with retarded individuals' reading. to be more precise. environmental differences are also a legitimate concern. Are certain relationships attributed to reading when. in fact. they might be attributed to the persons' being from radically different environments? Thus a profile of differences between good and poor readers should be generated on homogeneous populations. Also. the population should be large enough to minimize the possibility of exaggerating differences that may be slight. at best. Despite the numerous problems in investigating the relationship between reading and personality and intelligence. an attempt should be made to understand this relationship among well-defined. homogeneous populations. .Adding to this body of knowledge will contribute to diagnostic advances in reading education. The present study is an attempt in this regard. W The open door community college operates under the assumption that all persons. given the opportunity. have a potential for academic learning. Reading is the foundation for academic growth and develop- ment. Therefore. information that will help practitioners understand the reading characteristics and processes of open door community col- lege students will aid in cultivating and developing their academic potential. To this end. the investigator's main purposes in conducting the study were to examine the relationships between reading and personality and intelligence scores of open door community college students and to explore the development of a personality and achievement profile that will help practitioners assist students to realize their potential. A third purpose in conducting this research was to study the relationship between reading and personality and intelligence within a homogeneous population. The subjects for the present study were selected from logic. psychology. and reading classes at a midwestern open door community college in a district that predominantly comprises middle—class managers and skilled workers. All subjects chosen for the study resided within the college district and were Caucasian. ‘To this extent. environmental differences should have been minimized and a reasonable homogeneity of the group achieved. flxmlbeses The following hypotheses. stated in the null form. were formu- lated to guide the collection of data in this investigation: : There will be no statistically significant relationship between group total reading achievement and group scores on the individual scales of the Clinical Analysis Questionnaire. : There will be no statistically significant relationship between males"total reading achievement and their scores on the individual scales of the Clinical Analysis Questionnaire. fiypgthesjspa: There will be no statistically significant relationship between femalesfi'total reading achievement and their scores on the individual scales of the Clinical Analysis Questionnaire. .flypgthesis_fi: There will be no statistically significant relationship between group total reading achievement and group performance on the individual subscales of the Wechsler Adult Intelligence Scale. Hyp91h351§_5: ‘There will be no statistically significant relationship between males' heading achievement and their performance on the individual subscales of the Wechsler Adult Intelligence Scale. flxnntbesis_§: There will be no statistically significant relationship between females' heading achievement and their performance on the individual subscales of the Wechsler Adult Intelligence Scale. W: There will be no statistically significant relationship between group reading achievement and group scores on the individual scales of the Clinical Analysis Question- naire when the effects of IQ have been controlled. ,Hypgthesjs_§: There will be no statistically significant difference between better-reading males and poorer-reading males on the individual scales of the Clinical Analysis Questionnaire. W: There will be no statistically significant difference between better-reading females and poorer-reading females on the individual scales of the Clinical Analysis Questionnaire. .Hypgtnesis_1fl: There will be no statistically significant difference between better-reading males and poorer-reading males on the subscales of the Wechsler Adult Intelligence Scale. HypothesstJJ: There will be no statistically significant difference between better-reading females and poorer—reading females on the subscales of the Wechsler Adult Intelligence Scale. ‘flypgthesis_12: There will be no statistically significant relationship between total group scores on the individual scales of the Clinical Analysis Questionnaire and total group scores on the subscales of the Wechsler Adult Intelligence Scale. W The following terms are defined in the context in which they are used in this dissertation. LnuuL;kxu;5:mmnun11y_college: A two-year college that requires no entrance examinations for admission to the school. Bettgn_neadens: 'The group of readers who scored in the upper 40% of the distribution of scores obtained on the Nelson-Denny Reading Test. ‘EQQL_L§fld§L§: The group of readers who scored in the lower 40% of the distribution of scores obtained on the Nelson-Denny Reading Test. It should be noted that use of psychological terminology in this study reflects the language and thought of the authors of the various studies cited and the terminology and explanations contained in the Clinical Analysis Questionnaire Handbook. Instnumentation Because this study is a replication and extension of the research conducted by Lockhart and Menary. the same instrumentation and methodology were used. The Nelson-Denny Reading Test (Form D). the Cattell Clinical Analysis Questionnaire (CAO). and the Wechsler Adult Intelligence Scale (WAIS) were used to obtain the scores for comparison purposes. We The subjects for this study were categorized into two groups according to the scores they obtained on the Nelson-Denny Reading Test. Students with scores in the upper 40% constituted the group designated as better readers or better-reading college students. Those students with scores in the lower 40% constituted the group designated as poorer readers or poorer-reading college students. The Pearson product-moment correlation technique was used to test Hypotheses 1 through 6 and 12. A partial correlation was employed to test Hypothesis 7. To test Hypotheses 8 through 11. a two-way analysis of variance was used. WW Chapter 11 contains a review of pertinent literature in two major areas: (1) studies comparing personality and reading achievement and (2) the relationship between individualized IQ test and reading achievement. .A description of the test materials and data-collection and analysis procedures employed in the study is found in Chapter III. The research hypotheses are also presented. Chapter IV reports the results of the data analysis and hypothesis testing. A summary of the investigation. appropriate conclusions. and recommendations for further study are contained in Chapter V. CHAPTER II REVIEW OF RELATED LITERATURE The review of related literature is organized under two major headings. These are: (1) studies comparing personality and reading achievement and (2) the relationship between reading achievement and Wechsler Intelligence Test subscale patterns. 'The review of research in these areas served as a basis for designing and conducting the present investigation. WWW StudieuLtDLElememzau W Although the current investigation is concerned with examining the relationship between reading and personality and intelligence in a community college population. studies conducted with elementary and secondary populations can lend insights into the relationship between reading and personality and achievement in younger students. In this section. studies done at the elementary level and the mixed elementary/secondary levels will be summarized before discussing the research conducted at the college level. In a 1954 study. Spache investigated the relationship between reading and personality in children attending a reading clinic. Acting 10 on a "clinical hunch" that many of the students in the reading clinic evidenced definite negativism about and resistance to learning. Spache administered the Rosenzweig Picture Frustration Test to 50 of the clinic's clients. Children ranged from 6 to 14 years of age. with an average age of 10.96. The subjects had a mean Wechsler VIQ of 93.2. When this group of subjects was compared with Rosenzweig's group of normal children. the clinic cases appeared significantly more aggressive and cocky. less insightful. and less apt to admit blame or fault than Rosenzweig's normative group. Also. the clinic cases showed strong tendencies toward intolerance in conflict situations. By modifying Rosenzweig's scoring method. Spache was able to separate situations of conflict into situations of child-child conflict and adult-child conflict. thereby enabling him to study how the poor readers reacted in conflict situations with their peers and with authority figures. Spache found that the clinic population reacted differently toward adults than they did toward their peers. With an adult. the poor readers were less aggressive and more inclined to accept blame. When dealing with their peers. they were aggressive. cocky. and less apt to try to solve a problem. They aggressively resisted their peers but passively resisted adults' suggestions. In conclusion. Spache confirmed that his clinic population manifested a resistance to learning. FHs.findings also confirmed that simple remedial work without attention to the learner's emotional attitudes often is unsuccessful. 11 Three years later. Spache (1957) administered the Rosenzweig Picture Frustration Test to a group of 125 children in grades 1 through 3 who were reading one year below grade level and children in grades 3 and above who were reading two years below grade level. He found five major personality patterns that he felt were important to understanding children's failure in reading. 1. an finanfififiixe or hostile group in conflict with authority figures. 2. an adjustive group that seeks to be inoffensive. 3. a defensive group that is sensitive and resentful. 4. a solution or peace-making type. 5. an autistic group characterized by blocking or withdrawal. Spache's findings strengthened his belief in the need for individual- ized study and treatment of children with reading problems. Chronister (1964) also conducted research on the relationship of personality to reading achievement. His study was an attempt to determine the relationship of certain measured personality variables to reading achievement; 'The personality variables he chose for study were self-reliance. personal worth. personal freedom. feeling of belonging. freedom from withdrawal. freedom from nervous symptoms. social standard. social skills. and freedom from anti-social tendencies. The subjects for Chronister's study were 167 fifth-grade pupils enrolled in a central Missouri school system. He administered the Iowa Every Pupil Test of Basic Skills--Silent Reading Comprehension Test (Form A) and the California Test of Personality and Behavior Preference 12 Record. Chronister concluded that personality factors have a slight but positive relationship to reading comprehension. He recommended that teachers who are evaluating the work of underachievers should give considerable attention to other factors than personality. In another study with students of similar ages and grade levels. Zimmerman and Allebrand (1965) compared poor readers with good readers in terms of their personality characteristics and attitudes toward achievement. They hypothesized that poor readers (subjects reading at least two years below grade level) would show less adequate personal adjustment toward achievement than good readers. The study population consisted of 71 poor readers. known as the remedial group. and 82 good readers. known as the contrast group. equated as nearly as possible for age. sex. ethnic composition. and intelligence. The children were predominantly of middle to lower socioeconomic status. and roughly half were of Mexican descent. Subjects were drawn equally from the fourth and fifth grades of an urban school district. ‘Twice as many boys as girls were represented in the remedial group. and the same proportion was maintained in the contrast grouD. Subjects in the remedial group were enrolled in a remedial reading program. Children in this program had been screened on the basis of having average or better-than-average intellectual ability and reading at least two years below grade level. ‘The contrast group comprised children who were reading at grade level or above. but otherwise matched to the remedial group. 13 The California Test of Personality was used to measure personal and social adjustment. 'To measure attitudes toward learning. each child was asked to tell a story about card I of the Thematic Appercep- tion Test. In the area of personality functioning. Zimmerman and Allebrand found significant differences between the good and poor readers. Spe- cifically. the poor readers characterized themselves as having nervous symptoms. limited personal freedom. and feelings of isolation. The good readers saw themselves as having personal worth. absence of with- drawal tendencies. and self—reliance. Major significant differences occurred in the personal rather than in the social adjustment areas. In regard to achievement attitudes. as revealed by the Thematic Apperception Test card 1. good readers embraced such middle-class goals as practice and study with a payoff of future success. Poor readers did not stress effort. and more than one-third of the poor readers' stories evidenced a negative tone. The authors concluded that the good readers appeared to reflect the concepts of adjustment and motivation prized by teachers and school psychologists. whereas the poor readers admitted feelings of inadequacy and nervousness and had short-lived goals related to immediate and social achievement. Hake (1969) conducted research to discover and isolate person- ality maladjustments among poor readers. He also sought to develop a projective test to evaluate covert motivation as contrasted with overt behavior related to reading situations. Hake's study group included 80 sixth graders selected from a population of more than 600 sixth graders 14 in a midwestern public schooL. They had average intelligence. based on standardized test scores. and represented a wide socioeconomic range. The 80 children were divided into below-average and above- average reading groups. based on their scores on the California Achievement Test. The below-average group comprised those who scored one or more years below their grade level. whereas the above-average group scored one or more years above their grade level. No pupil undergoing special psychological treatment was included in the study. Hake found significant differences (p < .01) between the above- and below-average readers on the following personality factors evalu- ated by the Reading Apperception Test. a test similar in format to the Thematic Apperception Test: 1. Poor readers more often saw their parents and homes as less warm and comforting than did the good readers. 2. Significantly more poor than good readers identified with story characters whose teachers punished them freely for their learning difficulties. A comparison of the overt classroom behavior of the two reading groups as rated by the Pupil Behavior by the Teacher Rating Scale. a measure of overt behavior. indicated that the poor readers 1. displayed more negative classroom behavior: 2. were characterized as more shy and withdrawn; 3. were more easily led. distracted. and selfish: 4. were more happy and carefree about their work; 15 5. were rated more emotionally unstable. nervous. and unreliable; and 6. were rated more depressed and more easily discouraged than their better-reading counterparts. Hake concluded: In general the findings of the study support those of earlier studies that poor readers exhibit significantly more negative desires and wishes along with more maladaptive classroom behavior than do good readers. Also. as in previous studies. this research points up the fact that classroom teachers and reading clinicians must.not only be concerned about the poor readers! word recognition problem. but must be equally sensitive to their emotional diffi- culties which are indeed considerable. (p. 738) Neville. Pfost. and Dobbs (1967) studied the relationship of several personality factors to silent-reading achievement gains. The subjects were 54 boys who were enrolled in a summer reading program offered by the Child Study Center of George Peabody College for Teachers. The subjects' ages ranged from 7 through 14. with a mean age of 10.5. Grade level ranged from 3 through 9. with a mean grade level of 5.5. Before instruction the students took the Peabody Picture Vocabulary Test (Form B). the Lorge-Thorndike Group Intelligence Test (Form A). the Test Anxiety Scale for Children (TASC) by Sarason. and either the Gates Basic or Advanced Primary Reading Test. Subjects were taught using the clinic's ordinary diagnostic teaching methods: teachers helped the children gain skills in areas of deficiency. The subjects were then tested with a parallel form of the Gates test and reading gain was calculated. Subjects were divided into three groups according to their scores on the TASC: High Anxiety (HA). Middle Anxiety (MA). and Low Anxiety (LA). The groups were not 16 different on several factors related to achievement. namely. IQ. age. and grade. 'Three simple analyses of variance comparing the three anxiety groups on each of these factors resulted in F-ratios well below that required for statistical significance. 'Thus it was assumed that any effect these factors might have had on reading gains could be attributed to chance. When comparisons were completed. it was found that high test anxiety did have an inverse relationship to comprehension gain but not to vocabulary gain. However. both very high and very low test anxiety were associated with little or no gain in silent-reading comprehension. whereas a medium level of anxiety seemed to be associated with greater gain. In attempting to account for this finding. the authors stated: It seems reasonable to assert that comprehension is a more complex task than vocabulary. and that the complexity of the task and the high anxiety resulted in inferior performances among the HA group. It is difficult to explain the LA group in this light. but it appears possible that the successful performance of a complex task requires more personal involvement than does the successful performance of a more simple task. ‘Thus. the LA group had enough involvement to perform as well as the MA group on the Voc. task. but not enough to perform as well on the Comp. task. (p. 49) Neville et ale results indicated a curvilinear relationship between anxiety and achievement gains. Moderate amounts of anxiety were associated with greater gains in reading achievement than were either high or low amounts of anxiety. As can be noted from the studies conducted at the elementary- school level. some personality characteristics appear to be more commom than others in poorer readers. Some of these characteristics are aggressiveness among boys. negative attitudes. impulsiveness rather 17 than goal-directed behavior. and anxiety factors. Other identified personality factors seemed to be restricted to a particular study and thus may have reflected the measuring instrument used in the investi- gation. WW Joseph and McDonald (1964) conducted an exploratory study at Marquette University to determine if a discernible relationship existed between personality needs. as measured by a forced-choice instrument. and reading performance. as measured by a standardized group test. The study group comprised 1.475 Marquette University freshmen. 'The SAT scores for the group were above the national average. and about two- thirds of the group had been in the upper one-third of their high school graduating classes. Subjects were given the Diagnostic Reading Test and the Edwards Personal Preference Schedule. 0n the basis of the reading test scores. Joseph and McDonald divided the population into the following sub- groups: Group l--High Comprehension (N = 200) Group 2--Low Comprehension (N = 200) Group 3--Top Comprehension (N = 55) Group 4--Bottom Comprehension (N = 53) Group 1 comprised the top 15% of the population in reading comprehen- sion. and group 2 constituted the lowest 15% of the population. Groups 3 and 4 were subsamples of Groups 1 and 2 and reflected the top and bottom decile. respectively. based on local norms. 18 The investigators then created three additional groups based on reading rate. Membership in these groups was determined by students' scores on the Reading Versatility Test. The group designations were as follows: Group 5--Efficient Readers (N = 6) Group 6--Effective Readers (N = 19) Group 7--Ineffective Readers (N = 38) No effort was made to equalize the number of males and females in each group. but the two sexes were about equally represented. Tests of significant differences in the mean were used to compare the data. Many of the comparisons in the Joseph and McDonald study were not significant beyond the .20 level. This might have been a result of the good overall ability of the Marquette University freshman class. Because most of the students in the population had been in the upper one-third of their graduating classes. they were a fairly select group of individuals. This. in effect. would minimize the qualitative dif- ferences that may have existed within a more heterogeneous population. The lack of correlations might also have been a result of the small sample size in the final groupings. Among the significant findings of this study. Group 1 compared with Group 2 showed a tendency toward autonomy. whereas Group 2 had a greater tendency toward deference and order. Readers in the top 15% of the group showed greater independence and flexibility than the lowest 15% of the readers. whereas the poorer readers tended to be dependent and lacked self-confidence. 19 Brunkan and Shen (1966) were stimulated by the results of the Joseph and McDonald study and sought to replicate it. They used the Reading Versatility Test (RVT) and substituted the Adjective Check List (AOL) for the Edwards Personal Preference Schedule (EPPS) because the ACL would provide information on a greater number of scales than the EPPS. The subjects were selected from Marquette University's 1964 incoming freshman class on the basis of their RVT scores. SAT scores were already available on these students. The selected students fit into three categories of reading performance: 1. Efficient readers. 60% comprehension and a rate/ratio of more than 1 to 1.5 between fiction and nonfiction (N = 38). 2. Effective readers. 60% comprehension and a rate/ratio of less than 1 to 1.5 between fiction and nonfiction (N = 226). 3. Ineffective readers. less than 60% comprehension and rate/ ratio of less than 1 to 1.5 between fiction and nonfiction (N = 57). Each of these qualitative categories was then subdivided into three reading-rate levels on the basis of the ratio between fiction and nonfiction. These levels represented the upper (N = 146). middle (N = 84). and lower (N = 91) one-third of the freshman class. With a high. middle. and low reading-rate category within each of the qualitative groups. a total of nine comparison groups resulted. Brunkan and Shen found that a relationship existed between quality and rate of reading and one's sel f-description. Significant interactions on five of the variables indicated that both rate and 20 quality must be considered. Variables without interaction effects indicated twice as many significant differences due to rate than due to quality. When considering the results according to the following subgroups. certain characteristics emerged: WW: High on: self-confidence. dominance. exhibition. autonomy. and SAT Verbal. Low on: succorance. abasement. deference. and counseling readiness. WW: High on: self-confidence. autonomy. dominance. exhibition. and SAT Verbal. Low on: heterosexuality. abasement. succorance. deference. and counseling readiness. Wallace: High on: heterosexuality. deference. abasement. and succor- ance. Low on: self-confidence. dominance. autonomy. and SAT Verbal. The ineffective low-rate readers portrayed the poorest personality patterns for success in school. Their scores indicated they were passive. dependent. seeking reassurance. lacking leadership. and lower in general ability as measured by the SAT. In conclusion. the authors felt their research generally supported the Joseph and McDonald study because a relationship was found between personality characteristics and reading characteristics. In addition. their study identified desirable traits in good readers as well as indicating less- desirable characteristics of poor readers. "Thus not only were the 21 results of previous studies reaffirmed but also a broader range of personality and reader characteristics were investigated and described" (Brunkan & Shen. 1966. p. 842). Raygor and Wark (1964) examined poor readers' personalities and the difference between typical poor readers and typical normal students in a college sample. Their personality measure was the Minnesota Multiphasic Personality Inventory (MMPI). a test developed to aid in diagnosing psychiatric patients. The comparison groups comprised 151 male and 67 female students who voluntarily sought aid and remained in the program at the Reading and Study Skills Center at the University of Minnesota and for whom MMPI scores were available. and a random sample of 608 males and 508 females drawn from all entering freshmen in the College of Science. Literature and Arts (SLA). to whom the MMPI was administered as a routine part of registration. The Center sample represented the poor readers. and the SLA sample represented the normal readers. An average score on each of the scales was computed for each of the groups. and a t-test of the differences was conducted. The data were analyzed by looking at the specific scales that were significantly different across sex. and also by using the high-point code. which groups profiles according to the three highest scale scores above 55. Students having similar profiles have common adjustment problems. The results of the Raygor and Wark study indicated that the Center-sample males had a high-point code of 7. 4. and 8. A code of this kind indicates a person lacks skill with the opposite sex. has 22 conflicts at home. is nonverbal. complains of insomnia. and has poor rapport with counselors. The SLA-sample males had a high-point code of 5. 8. and 9. which indicates CODfIlCtS‘Wlth mother. complaints of insomnia. and poor rapport with counselors. The only apparent differ- ence on high-point comparison was that the poor-reading males in the Center sample had fewer social skills with girls and were less verbal. 0n individual scales on which the difference between the two groups of males was significant. the poor readers complained more about health and general physical condition. tended to be more immature. and had a greater need for social acceptance. 'They also seemed somewhat more depressed. more irresponsible. and more shy and withdrawn than the SLA sample. Among females. there were no scores over 55 for the SLA sample and only one scale score over 55 for the Center sample: hence there was no high-point coding. When individual scales were compared. Center females and SLA females were significantly different from each other on three of them. Surprisingly. the poor-reading females tended to be emotionally healthier than the females in the normal sample. Poor- reading females were less depressed. less withdrawn and introverted. and slightly less anxious. tense. and uncomfortable. Raygor and Wark concluded that there were small but statis- tically reliable differences between the students in the two samples. They pointed out. though. that the differences might be interpreted as distinctions between the personalities of volunteers and nonvolunteers rather than between good and poor readers. as the Center students were 23 volunteers. The investigators anticipated that the results were gen- eralizable because other reading groups could be expected to contain students who are shy and withdrawn. Anderson (1961) investigated the possibility of a relationship between the subscale measures of the Cooperative Reading Test (Higher Level Form Y) and the personality factors of the Cattell l6 Personality Factor Test (FormTAL His sample comprised 290 males with a mean age of 18.0 years and 125 females with a mean age of 17.8 years. All subjects were freshmen admitted to the University of Western Australia. The mean intelligence of the group was 126. standard deviation 7. as measured by the ACER B-40 Intelligence Test. In correlations obtained for the males. females. and combined group. certain factors showed a consistent relationship with reading subscales and with Total Reading Score. These factors were intelli- gence. conscientiousness. sensitivity. introversion. radicalism. and self-sufficiency. Anderson described those who scored high on the reading test as more intelligent. less conscientious and persevering. more sensitive. more introverted. more radical. and more self- sufficient than those who scored low on the test. Lockhart and Menary (1978) used Cattell's l6 Personality Factor Test to study the relationship between personality traits and reading scores of a community college population. They administered the Nelson-Denny Reading Test (Form D) to 150 students in logic and psy- chology cl asses in the community college. Thirteen top readers were selected for comparison with 13 students selected from a reading and 24 study-skills class. Nine readers from the logic and psychology classes roughly approximated the scores of the study-skills group and were included in the low-scoring reading group. and nine more top readers were selected for the high-scoring reading group. The total size of the group was 44; coincidentally. the sample contained 22 males and 22 females. Rank-order correlations were computed. and comparisons were carried out between sexually homogeneous groups. Good-reading males tended to be higher on B. intelligence; lower on G. less conscientious; lower on 01. less insecure: higher on 02. self-sufficiency; and higher on 04. more tense and withdrawn. than the poorer-reading males. Sig- nificant correlations with Total Reading Score on the Nelson-Denny for the male group were with A- (.05). 8+ (.01). and 02+ (.05). Good-reading females tended to be lower on A. more reserved: higher on C. more stable; lower on E. more dependent: lower on F. more serious: lower on G. less conscientious: higher on L. more suspicious; lower on 0. less insecure; higher on 01, more experimental: higher on 02. more self-sufficient: higher on 03. more lax; and lower on 04: more relaxed and composed. than their poorer-reading counterparts. Signifi- cant correlations with Total Reading Score on the Nelson-Denny for the female group were with M (.01). 0- (.05). and Q4— (.05). An interesting finding of the Lockhart and Menary study was that differences in personality scores existed between poorer readers who sought help at a reading center and poorer readers of similar abilities who did not seek help at such a center. .Among males. poorer 25 readers in the general population were more happy-go—lucky. more group dependent. less exacting. and more lax than those in the study-skills class. The poorer-reading females from the general population seemed to be more intelligent. more cheerful. less conscientious. and more socially naive than females in the study-skills class. In this pilot study. Lockhart and Menary sought to determine whether Cattell's 16 PF could differentiate profiles of community college students. Because the number of subjects was small. direction of difference rather than magnitude or significance of difference between groups was reported. In a second study. Menary and Lockhart:(l979) increased the size of their sample. hoping to strengthen their original findings. They also included a measure of pathology by using Cattellls Clinical Analysis Questionnaire (CAQ). which retains the 16 personality factors and adds 12 pathology scales. The sample was again obtained by administering the Nelson-Denny Reading Test (Form D) to students enrolled in logic. psychology. and reading cl asses at a midwestern community college. The students were grouped according to the scores they obtained on the Nelson-Denny. The personality scores for the respective groups were tabulated and aver— aged: no attempt was made to correlate the results. A trait was deemed significant if the between-group score was a sten or better. Comparison of the results of this study with those of the first study indicated that the good reader had been profiled remarkably well; ten scales for the males remained the same and eight scales for the 26 females remained the same. In addition. scales that changed. altered in the direction of expectancy stated in the first study; only the B scale for the males shifted in the opposite direction. On the other hand. the profile of poorer readers did not fare as well. The profile of the poorer male reader exhibited shifts oppo- site to expectations on Scales H. 0. and 04; eight scales remained the same and five changed in the direction of expectancy. Among the female poorer readers. five scales shifted in the opposite direction (A. C. F. 03, and 04). whereas seven remained and same and three changed in direction of expectancy. The investigators found the better male readers were lower on A. reserved versus warm: higher on B. intelligence; higher on E. assertive versus dependent; lower on F. introspective versus happy-go- 1ucky; lower on G. self-indulgent versus conscientious; higher on 01' experimental versus conservative; higher on 02, selfesufficient versus group dependent; higher on 03. excitement seeking; higher on 02. avoids involvement with people; and higher on Pd. had complacent attitude toward self or others' anti-soci a1 behavior. than were the poorer male readers. Significant differences (by the authors' definition of a sten score or better) between the better and poorer male readers were as follows: Better readers scored significantly higher on 8. Q1, 02, 03, and Pd. Poorer male readers scored significantly higher on F. happy- go-lucky VGPSUS serious; 04. tense and easily upset; 06. feelings of guilt; and As. repetitive ideas and impulse to do certain acts. 27 Good female readers had a personality profile that was higher on 8. intelligence; E. assertive versus dependent; F. happy-go—lucky versus introspective; H. adventurous versus shy; I. imaginative versus practical: L. suspicious versus trusting; M. unconventional versus conventional; 0}. experimental versus conservative; and 82. self- sufficient versus group dependent. than that of the poor-reading female. The good-reading females were profiled lower on N. socially climbing versus socially aware: 0. secure versus insecure; and Pd. has complacent attitudes toward one another‘ianti-social behavior. than were the poor-reading females in the study. Significant differences between the good-reading female and her poorer-reading counterpart were as follows: Better readers scored significantly higher on B. F. I. M. 01, 02, and Pd, whereas poorer female readers scored significantly higher on 0. insecure and worrying: D4. tense and easily upset; 05. feelings of guilt: Pa. persecution: and As. repetitive ideas and impulses. The authors concluded that the CAQ did distinguish between the better and poorer readers of the study and that extending the scales of a previous study resulted in a strong set of discriminating scales. The studies conducted at the college level reflected some of the effects of replication of design or device. Brunkan and Shen. for example. extended Joseph and McDonald's findings. The results of the studies using Cattell's 16 PF were almost identical. even though one of the studies was conducted in Australia at an earlier time. According to these studies. the good reader is more intelligent. autonomous. 28 self-confident. and free thinking than his poorer-reading counterpart. The poor reader is more passive. abased and deferential. has fewer verbal skills. and is depressed and withdrawn. The negativism and impulsiveness that were evident among the poorer readers at the elementary-school level seem also to be present among the older poorer readers. The aggressiveness may simply be of a different nature. disguised in the measured passivity of the poorer reader who disregards the advice of others and has poor rapport with counselors. The subtest patterns of the Wechsler Intelligence Test have been used for diagnostic purposes almost as long as the scales them- selves have been in existence. Zimmerman and Woo-Sam (1973) stated. Wechsler himself (1958) while warning of the unreliability of the individual subtests (a point basic to profile analysis). notes that the qualitative data to be obtained. even though largely inferen- tial. is actually the essence of a good test. (p. 16) In the area of reading. many individuals have attempted to use the "essence of a good test" to develop profiles for predicting reading ability or for identifying retarded readers based on pattern analysis. Ekwall (1966) reported that in the 20 years preceding his study more than 19 statistical studies "reported on the relation of distinctive subtest patterns to reading retardation" (p. 2). Spache (1976) provided a table summarizing the patterns of WISC subtest scores of poor readers. as found in 26 research studies. (See Table lJ He pointed out that although the methods various researchers 29? .Lcmu >4 vou_u moz so uu._u mad .mcovmoc teem coca Lo;m_; >_4cmu_u_cm_m mcoumuc cocoa .mcuvmoL teem coca L036. >_ucmu_u_cu_m mcoumoL Looam .Awmm_ .couom uco c>__< “acumOmv mo_u___amm_o m:_umox mo moamm. ozu m:_umm44mo>c_ .onumom omcoou "oucaom 4 cmEIumcum 4 c_zcoU I soon 4 4 m_zo4 cmEm.oaz 4 Laym04 a sauna Lu_:cm co 4 ncov_u:m 4 ccOmuLm:u_¢ Lo050m a o_o¢ z 4 UL_:: 4 v004ut 4 tcmo4uz IOmmm w cmEo_oU 4 xuon0x uu_mc0ouz umccoumm 4 __ucxuoo 4 mo__mx 4 u___>oz umc_: : ozumam ucOmuLmzu_¢ u.=s_< ounce a mxcsm Emzmco 4444 44444 I I 4 44 I I 4 4 4 I 4444444444444 4444 4 4 4 I I I IIII I I I II IIII IIIII 4 4 4 444444444444 IIII 4 4444 > I I .neomm< LL< cm_moo _a504 .4muo> cmam mc_aou ouuqno «Lauu_a xuo_m oLsuu_a u_m_o .__e_m .;._L< .aeoo .oec_ >essm .mcovmoL Loon chEm moLoOm um0443m um.) Io meccauma--._ o_om» 30 used to identify patterns of poor readers differed greatly. there was considerable agreement in the patterns themselves. 0f the 26 studies summarjzed in the table. 19 found poor read- ers scored low on the Information subtests of the W180. Nine studies reported poor readers scored high on the Comprehension subtest. whereas one reported they scored low. Twenty-five studies reported poor read- ers scored low on the Arithmetic subtest. and only one found no differ- ence. Few and mixed results were reported for the Similarities sub- test: only seven studies reported significant differences. with five showing the good reader doing well and two reporting the poor readers scoring high on the subtest. Fourteen studies reported low scores on Digit Span. Only eight studies reported differences on the Vocabulary subtest; surprisingly. two of those reported the poor readers did better than good readers on this subtest. Fifteen studies reported the poor reader scored high on the Picture Completion subtest. Nine studies indicated higher scores on Block Design. whereas one study reported low performance on this subtest. Poor readers did well on Picture Arrangement in nine studies. 0n Object Assembly. seven studies reported that poor readers scored high. Finally. 20 studies reported that poor readers did poorly on Coding. Converting the foregoing information into simple percentages and arranging the results according to the percentage of studies indicating scale differences. it was found that poor readers scored higher than good readers on Picture Completion in 57% of the studies. 31 on Comprehension in 34%. on Block Design in 34%. on Picture Arrangement in 34%. on Object Assembly in 26%. on Similarities in 19%. and on Vocabulary in 7% of the studies. The poor readers scored lower than good readers on Arithmetic in 96% of the studies. on Coding in 76%. on Information in 73%. on Digit Span in 53%. on Vocabulary in 23%. on Similarities in 7%. and on Comprehension and Block Design in 3% of the studies. Viewed in this light. 50% or more agreement among investigators occurred on only five scales. These scales. in order of magnitude of agreement. were Arithmetic (96%). Coding (76%). Information (73%). Picture Completion (57%). and Digit Span (53%). The low scores came on the verbal portion of the test: Arithmetic. Information. and Digit Span. ‘The high scores on the performance portion of the test were in Picture Completion. the low in Coding. Seemingly better scores by poorer readers on the performance section of the test have led some researchers to believe that higher performance scores are indicative of the poorer reader. According to Newland and Smith (1967). for a dif- ference to be significant at the .05 level between Verbal and Perform- ance 105. a 15-point difference must occur. and to reach the .01 level of confidence there must be a 20-point gap in obtained scores. 0n the whole. the methodologies that have been used to compare WISC IQs and subtest patterns to reading level have varied considerably from one study to another. Spache (1976) pointed out that bases of these comparisons vary from one writer to another: some comparing mean scores of matched or unmatched groups; some allowing for the probable error of the estimate. while others ignore these; some considering almost any variation among subtest scores as 32 significant. while others compared each pupil's subtest score with his own average as a correction. (p. 140) Differences occur regardless of methodology. But what is a significant difference. and if a significant difference occurs what does it mean? Newland and Smith (1967) attempted to answer the question concerning significant differences between subtest scores on the Wechsler tests. Using the properties of a normal distribution curve and the basic reliability data from the Wechsler manuals. the authors computed the magnitudes of the differences between subtest scores needed for the .05 and .01 levels of significance. They presented tables for test-score differences necessary for significance at the.05 and .01 levels for the WISC and the WAIS. Newland and Smith also presented in tabular form median differences within six age groups. and within and between the verbal and performance subdivisions of the WISC and WAIS. ‘The authors indicated that larger differences on the WISC were necessary at younger age levels because of the problem of relia- bility in testing younger children. One can only wonder what results would have been reported among the WISC studies had similar methodolo- gies and statistical-significance tables have been used. Now that the magnitude needed for significance is known. the question of the meaning of the difference must be addressed. Cohen (1957) conducted a factor-analytic study of the WAIS on a normal population over a wide age range. 'The groups studied were ages 18-19 (N = 200). 25-34 (N = 300). 45-54 (N = 300). and 60-75+ (N = 352). Cohen found that the same factors operated over the entire age range 33 and were the ones identified for clinical populations on the Wechsler- Bellavue. The major factors were A. Verbal Comprehension. B. Perceptual Organization. and C. Memory. Two minor factors were also found: Factor 0. a Picture Completion specific found in all four age groups. and Factor E. a Digit Symbol specific absent only in the oldest group. The subtests that measure Factor A. Verbal Comprehension are Information. Comprehension. Similarities. and Vocabulary. ‘The subtests that constitute Factor B. Perceptual Organization are Block Design. Object Assembly. and Picture Arrangement. Factor C. Memory includes Arithmetic and Digit Span. Factor 0 is unequally Picture Completion. and Factor E. is unequally Digit Symbol. A result of Cohen's study is that the meaning of what the same T-scales measure is imparted to the user. This gives some rationale upon which clinical decisions and interpretation can be based. McDonald (1964) incorporated both the magnitude for statistical significance and a meaning of the results based on factorial studies in his investigation of intellectual characteristics of older disabled readers. He studied disabled readers at the high school and college levels because virtually all previous studies had involved the WISC and younger readers. He randomly selected a group of 60 disabled readers from files on adolescents referred to the Marquette University Reading Clinic. The age range was 16.1 to 19.4 years. All subjects were male; therefore. sex was eliminated as a variable. Reading was assessed by various instruments; most commonly used were the Durell Analysis of 34 Reading Difficulties. the Oral Word Attack section of the DRT. and the Gates Analysis of Reading Difficulty. McDonald wrote. To minimize flaws arising from varying intercorrelations and reliabilities among the subtests of the WAIS. each subtest score was compared only with the mean subtest score within the same scale and by employing a value to be considered indicative of a signifi- cant deviation which was greater than the standard deviations of the subtests. (p. 98) The findings indicated that the study group had verbal 105 of 95.8. Performance 105 of 105.3. and Full Scale 105 of 99.8. The 9.5- point difference between Verbal and Performance IQ was statistically significant. according to McDonald. The significant differences in subtest scores showed plus deviations for disabled readers on the Comprehension. Picture Completion. Picture Assembly. Block Design. and Object Assembly subtests. Minus deviations occurred on the Informa- tion. Arithmetic. Digit Span. and Digit Symbol subtests. McDonald compared these findings to similar findings from other WISC studies and analyzed the results in light of Cohen's factor analysis. He stated. "The disabled readers in this group performed most poorly on subtests reflecting concentration and attention" (p. 100). Cohen had suggested that these subtests are a measure of the person's "freedom from distractibility." By relating his significant factors to suggestions of what the scale measures. McDonald indicated that results of WAIS patterns could be useful to the reading instructor if he uses patterning as "suggestive" only and weighs it against all the available evidence. Spache (1975) reported that Belmont and Birch criticized pattern analysis because it is 35 based on clinical populations rather than on a true sample of the entire population. for use of groups not homogeneous in age or sex; and in the comparison of good and poor readers. for selection of cases from different schools. different social classes and different ages.... . [Bijou] considers pattern analysis useful only in the comparison of equated groups. (p. 143) Nevertheless. pattern analysis can aid the clinician or reading instructor in weighing the strengths and weaknesses of a student and in deciding upon a course of remediation. But as the preceding discussion has shown. many factors need to be explored before the WISC or WAIS can or should be used independently in determining types of reading dis- abilities or remedial treatments. Sumanv This chapter contained a review of related literature on the relationship between reading and personality and between reading and subscale patterns on the WISC and WAIS scales. The methodology of the study. including sample selection. instrumentation. the design. and hypotheses tested. is discussed in Chapter III. CHAPTER III METHODOLOGY OF THE STUDY .Intcoduction This chapter describes the methodology employed in conducting the study. which replicates and extends the procedures used in the Lockhart and Menary (1979. 1980) studies on personality and reading achievement at the community college level. The current study extends their work by including individualized intelligence measures and a more sophisticated statistical design. Emulation The community college district from which the sample was drawn consists mainly of white middle— and working-class families. Many heads of households are managers or skilled workers whose income places them economically in the middle class. The sample reflects the nature of the college district. ‘The sample was totally Caucasian; 44% came from homes of professionals. iue. engineers. pharmacists. teach- ers. bank managers. and funeral directors: 10% from homes of profes- sional sales people; and 34% from homes of highly skilled workers. i.e.. aircraft mechanics. foremen. and office managers. Only 8% came from homes of unskilled workers. whereas 3% did not indicate any paren- tal occupation. 36 37 The subjects for the study were drawn from introductory psy- chology. human relations. introductory and advanced logic. and reading and study skills classes. Although a cross-section of the student population was desired. it might not have been obtained voluntarily. Sample selection relied on both the willingness of the instructor to allow his classes to participate in the study and the issuance of extra credit to students who would take part in the study. An effort was made to include night-class students in the study so that the large night-school enrollment of the community college would be represented. That this effort was successful is reflected in the average age of the sample: 24.56 years. Males in the sample averaged 23.67 years. whereas females averaged 25.12 years of age. One way the investigator had hoped to control the distribution of the population was to select classes that had high enrollments year after year. and were applicable to the requirements for all degrees offered by the college. Psychology and logic classes met both of these criteria. 0n the other hand. the reading classes were selected so as to tap those students who felt they needed help with reading and study habits. Their representation contributed to the inclusion of a cross-section of reading levels appropriate to an open door community college. Thus. although the writer believes the sample reflects the population from which it was drawn. there were definite deficiencies in the manner of sample selection. on which the study can be criticized. The final configuration of the sample was 127 subjects. 49 males and 78 females. 38 Instcumeniation The tests used in this study were selected because of their applicability to the research conducted by Lockhart and Menary using the Nelson-Denny Reading Test. Cattellhs 16 Personality Factor Test. and Cattefll's Clinical Analysis Questionnaire (CAQ). The Wechsler Adult Intelligence Scale (WAIS) was included to control for intelli- gence and to examine the WAIS subscales among and between groups of college students with differing reading achievement. The reliability coefficients of the instruments as reported are more than adequate for an exploratory study. The Examiner's Manual for the Nelson-Denny Reading Test contains tables of reliabilities for individual grade levels 9 through 16. (See Appendix for a copy of the table.) The reliabilities were derived by the split-halves technique. and the authors of the manual related that the technique is more suitable for power tests than for the Nelson-Denny. on which speed plays an important role. They went on to state that the approach tends to err on the high side. resulting in a substantial overestimate of reliability. But on the Comprehension Tests. because of the smaller number of items. some believe that the split-halves approach tends to underestimate reliability. In any event. the reliabilities seem more than adequate for the purpose as stated (p. 29). The Nelson-Denny provides a measure of reading ability on three individual scales and on one combined scale; a separate score is avail- able for vocabulary. comprehension. and rate. The total reading score 39 is a derived score of vocabulary plus a double comprehension score. The total score is the best indicator of reading ability when using the Nelson-Denny and was the principal measure used in the present study. The vocabulary section contains 100 multiple-choice items. .An example of one of the items contained in the practice exercises is ”A chef makes (a) bricks. (b) dishes. (c) clothes. (d) food. (e) statues." Students have ten minutes to work on the vocabulary section of the test. The comprehension portion of the test contains 36 questions. based on eight reading selections. The first selection is a long passage and is the one on which the reading rate is measured. ‘The reading rate makes up the first minute of the second part of the test: then the student has 19 minutes to complete the comprehension section. TOtal administration time of the Nelson-Denny Test is 30 min- utes. A group can easily be tested in one 50-minute class period. The test is widely used for screening purposes. ‘The Clinical Analysis Questionnaire (CAQ) was used to obtain the personality profiles of the readers in the study. According to the CAQ manual. the reliabilities were calculated as test-retest coeffi- cients. The median test-retest coefficient on all 28 scales was reported as .73. the lowest coefficient being the N scale at .51 and the highest the Sc scale at .90. The clinical scales are somewhat more reliable (test-retest .80) than the normal personality scales (p. 27). 40 The CAQ measures 28 primary traits on two scales. the normal personality scale and the clinical scale. The normal personality scale consists of 16 factors with high score meaning as follows: Factor A. Warmth L. Suspiciousness B. Intelligence M. Imagination C. Emotional Stability N. Shrewdness E. Dominance 0. Insecurity F. Impulsivity 01. Radicalism G. Conformity 02. Self-sufficiency H. .Boldness 03. Self-discipline I. Sensitivity Q4. Tension The clinical scale consists of 12 factors with high score meanings as follows: Factor 01. Hypochondriasis D7. Boredom and withdrawal 02° Suicidal Depression Pa. Paranoia D3. Agitation Pp. Psychopathic Deviation D4. Anxious Depression Sc. Schizophrenia 05. Low Energy Depression As. Psychosthenia D5. Guilt and Resentment Ps. Psychological Inadequacy Second-order factors can also be calculated by combining certain pri- mary scales if it is desirable to do so. The CAQ contains 272 items; 128 items measure the normal per- sonality traits. and 144 items measure the clinical traits. It is easily administered in a group setting and requires only about two hours to administer. Parts I and II can be administered in different sittings. The manual reports that the test requires only a reading level of grade 6-7. which means remedial secondary and college popula- tions can take the test without special accommodation. The test is not 41 a forced-choice instrument; rather. each of the CA0 items has three choices with an uncertain or in-between category that the examinee can select. The Wechsler Adult Intelligence Scale (WAIS) was employed to measure and control for I0. It is widely used in clinical settings and in psychological research. The test was designed specifically for administration to individuals 16 years of age and above. It must be administered by an examiner trained in the administration and interpre- tation of the test. The reliability as reported in the WAIS Manual (1957) was based on the spl it-half technique for three age groups: 18-19. 25-34. and 45-54 years of age. Reliability coefficients were computed for Full Scale IQ (FIQ). Verbal IQ (VIQ). and Performance IQ (PIQ) for each age group. The reliability coefficient for the F10 was .97 for each age group; for the V10 a reliability of .96 for each age group was obtained; and for the PI0 a reliability of .93 was obtained for the 18-19 and 25-34 age groups. and a reliability of .94 was computed for the 45-54 age category. Individual subtest reliabilities ranged from .65 on Object Assembly to .94 on Vocabulary. Thus caution must be employed when making judgments concerning scores on a subscale or differences between subscale scores. Newland and Smith's (1967) tables of significant differences among subscale scores should be considered when interpreting differences. Administration of the WAIS generally takes between one and one and a half hours. depending on the skill of the examiner and the examinee. Testing with the WAIS requires a well-lighted room free from 42 intrusions and noise. with enough space for the tester to be comfort- able when manipulating the materials. The WAIS contains a Verbal scale with six subtests and a Per- formance scale with five subtests. Separate IQs can be computed for each scale. The Full Scale IQ is the sum of the Verbal and Performance scores. The Verbal scale consists of the following subtests: ‘lnignmatlon: This subtest contains 29 items and is designed to measure general information the subject has accumulated. Possibly long-term memory. reading habits. awareness about or interest in events. and interest areas are being measured. Performance on the scale is greatly affected by the subject's age. High scores are char- acteristic of college-level or gifted high school students. Low scores can result from cultural bias and anxiety. as well as a nonachievement orientation (Zimmerman 8. Woo-Sam. 1973). Comprehension: This subtest contains 14 items and is designed to measure how well the subject can grasp past experiences and apply them to social situations. Many of the questions deal with moral or ethical judgments the subject is asked to make. Formal learning is not as important in this subtest as in the Information subtest. High scores suggest a socially aware individual who can apply reasoning ability and common sense to social situations. Low scores can reflect a lack of verbal ability by those subjects who are concrete in their judgments or have anti-social thought patterns. mm: The Arithmetic subtest contains 14 timed items. The subject is not permitted to use pencil and paper but must solve the 43 problems in his head. Computational level does not exceed the seventh grade. The test measures the cognitive skills involved in solving word problems. as well as memory and concentration. Success indicates a task orientation as well as school learning. Low scores could indicate situational anxiety. carelessness in computation. or simple inability to concentrate on word problems. ,Similanities: The Similarities subtest contains 13 items designed to tap a person's ability to classify and abstract. An example of one of the easier questions is "In what way are an orange and a banana alike?" ‘The relationships grow more difficult from the concrete to the more abstract. .A high score on the Similarities subtest indicates good abstracting and conceptualizing ability. A low score indicates poor skills in these areas. Digit_§pan: This subtest contains 14 items; seven are Digits Forward and seven are Digits Backward. The test measures immediate recall. attention to detail. and freedom from distractibility. The subtest requires that the subject listen to a series of numbers and then repeat them. The items increase from a series of three digits to a series of nine digits in Digits Forward. In Digits Backward the items increase from two to eight digits. High scores indicate good immediate recall of a nonassociative task. Low scores can indicate boredom. distractibility. and poor concentration. 0n the whole. Digit Span is the poorest measure of intelligence in the WAIS. .Vggabulany: The Vocabulary subtest contains 40 items and is generally accepted as the best single indicator of intelligence. The 44 test measures a person's ability to use words; therefore. school learning and life experience are both tapped by this subtest. The test is oral. so reading ability may be tapped only by the more difficult items that come later in the test. High scores reflect good verbal comprehension. sophistication. and general intelligence. Low scores indicate low intelligence. cultural deprivation. and/or psychological problems. The Performance section of the WAIS comprises the following subtests: W: The Digit Symbol subtest contains 90 items. Subjects must substitute a symbol for a number in the 90 spaces pro- vided. They have 90 seconds to work on this task. Digit Symbol is a measure of hand-eye coordination; another aspect of intelligence that could be measured is memorization under pressure. High scores indicate speed of operation. accuracy. and freedom from distractibility. Low scores could result from poor hand-eye coordination. physical problems. compulsiveness. and left-handedness. .E1ctune_§gmmletion: The Picture Completion subtest contains 21 items designed to test the subject's awareness of common things in the environment. The subject is shown pictures with some significant element missing and is asked to identify that element. Picture Comple- tion is the best meassure of intelligence in the Performance section of the WAIS. High scores on Picture Completion indicate an awareness of environmental surroundings. good perception. and a broad base of gen- eral information. which knowledge of the surroundings would indicate. 45 Low scores could indicate poor attention to details. lack of interest in surroundings. and a suspicious nature that insists nothing is missing. Wu: The Block Design subtest contains ten items that are designed to measure reasoning ability and hand-eye coordination under time constraints. The subjects are required to look at a design pattern presented in a booklet and to reproduce the design with red and white blocks. Time bonuses are awarded for designs 7 through 10. High scores indicate good analytical reasoning ability of a nonverbal nature. Low scores may reflect a speedy. careless approach. anxiety. and compulsiveness. W: The Picture Arrangement subtest contains eight items that are designed to measure a subject's ability to compre- hend a story or situation from a group of scrambled clues and to arrange the clues into a whole. Sequential thinking. visual organiza- tion. and social knowledge are involved in this subtest. Subjects are presented with cartoon cards in a prescribed scrambled arrangement. The subjects are then asked to put the pictures into a sequence that tells a story. The test is timed. and bonus points are awarded. The subject who is successful on Picture Arrangement pays attention to small details and is logical with good sequencing abilities. Low scores may indicate a lack of attention to details. impulsiveness. and poor social awareness. .iject_A§§embly: The Object Assembly subtest contains four items and appears to measure visual organization and synthesis of parts 46 into familiar wholes. Object Asssembly is a timed test with bonuses for speedy work; partial points are awarded if time runs out before the subject completes an item. The subject must put together jigsaw puz- zles of four cut-up figures presented in a prescribed placement. This subtest can be compared to Block Design as both include perceptual organization and speed as components of the measurement. but the Object Assembly tasks may measure»a sociahunemory component or mental alert- ness. Object Assembly is the least reliable and poorest measure of intelligence in the WAIS. Success on this subtest calls for perceptual organization and integration of parts into wholes. Low scores may reflect anxiety or poor perceptual skills. IBUELLEfljssfldsul The subjects were tested both in groups and individually. The reading and personality tests were administered in group settings. ‘The WAIS was administered to students individually by one of the two test- ers trained in administering the WAIS. To facilitate data collection. cooperating professors assigned the testing sessions as a project for which the students earned a grade. Students received a general explanation of the nature of the study before the testing project. In addition. students received a copy of their scores on each test. and a class session was set aside for an explanation of what the test scores measured. If students had further questions. special office hours were set aside for private consultation. 47 Each student had to take all three tests to receive credit for the class project. If students objected to taking part in the class endeavor. a related project was provided. The tests were administered to all students in the cooperating classes. but only Caucasians born in the United States and residing in the district were included in the study sample. Students who did not complete all of the tests or whose WAIS protocols were invalid were not included in the study. Administration of the WAIS proved the most difficult aspect of the study. Only two examiners were used. to insure tester consistency. Appropriate testing rooms had to be secured for the duration of the data gathering and a flexible schedule worked out to permit adequate blocks of time for testing. Appointments for testing were made at the students' convenience. At the end of a testing day. the examiners consulted one another on scoring-protocol decisions on the Comprehen- sion. Similarities. and Vocabulary subtests to insure consistency in scoring. Complete test files were obtained on 140 students. One hundred twenty-seven students met the criteria for inclusion in the sample. 49 males and 78 females. W Based on the total reading score obtained on the Nelson-Denny Reading Test (Form D). the subjects were divided into a better-reading group (upper 40%) and a poorer-reading group (lower 40%) for comparison of scores obtained on the CA0 and on the WAIS. Two additional 48 subdivisions were made for intrasexual comparisons. The females were divided into better-reading females and poorer-reading females. Males were divided into better-reading males and poorer-reading males. Total reading score obtained on the Nelson-Denny was compared with scores obtained on the CA0 and the WAIS according to the following groupings: Jgtgl study population (N = 127) total reading x CAQ total reading x Female study population (N = 78) total reading x CAQ total reading x WAIS Male study population (N = 49) total reading x CAQ total reading x WAIS Better-reading females (N = 32) versus poorer-reading females (N = 32): difference in scores obtained on CAQ and WAIS Better-reading males (N = 19) versus poorer-reading males (N = 22): difference in scores obtained on CAQ and WAIS The means and standard deviations for the upper and lower groups of college readers on the Nelson-Denny and the WAIS-FIQ.are presented in Table 2 for the combined group. for the males. and for the females. Table 2.--Means and standard deviations for upper and lower college readers on the Nelson-Denny and the WAIS-FIQ. Total Reading Score WAIS-FIQ Upper Lower Upper Lower Mean 5.0. Mean 5.0. Mean 5.0. Mean S.D. Combined group 106.60 12.75 48.00 15.47 119.60 7.49 105.26 7.33 Males 109.52 13.82 45.47 16.56 122.42 6.64 106.85 6.53 Females 104.21 11.85 50.87 14.56 117.93 7.55 104.19 7.73 49 Hypotheses The following research hypotheses. stated in the null form. were formulated to guide the collection of data in this research: ‘flypgthesis_1: There will be no statistically significant relationship between group total reading achievement and group achievement on the individual scales of the Clinical Analysis Questionnaire. .flypgthe§1§_2: There will be no statistically significant relationship between males' total reading achievement and their achievement on the individual scales of the Clinical Analysis Questionnaire. ‘ .flypgthe51§_3: There will be no statistically significant relationship between females' total reading achievement and their achievement on the individual scales of the Clinical Analysis Questionnaire. Hypothe51544: 'There will be no statistically significant relationship between group total reading achievement and group achievement on the individual subscales of the Wechsler Adult Intelligence Scale. ‘Hypgthesis_5: There will be no statistically significant relationship between males' reading achievement and their achievement on the individual subscales of the Wechsler Adult Intelligence Scale. hypothesis_6: There will be no statistically significant relationship between females' reading achievement and their achievement on the individual subscales of the Wechsler Adult Intelligence Scale. Hypothe§1§_1: There will be no statistically significant relationship between group reading achievement and group achieve- ment on the individual scales of the Clinical Analysis Question- naire when the effects of IQ have been controlled. .flypgthesis_8: There will be no statistically significant difference between better-reading males and poorer-reading males on the individual scales of the Clinical Analysis Questionnaire. W: There will be no statistically significant difference between better-reading females and poorer-reading females on the individual scales of the Clinical.Analysis Questionnaire. 50 : There will be no statistically significant difference between better-reading males and poorer—reading males on the subscales of the Wechsler Adult Intelligence Scale. .flypgthesls_11: There will be no statistically significant difference between better-reading females and poorer-reading females on the subscales of the Wechsler Adult Intelligence Scale. : There will be no statistically significant relationship between the individual scales of the Clinical Analysis Questionnaire and the subscales of the Wechsler Adult Intelligence Scale. W The Pearson product-moment correlation technique was used to compare total reading scores with measures of personality and intelli- gence (Hypotheses 1-6 and 120. To compare reading with personality. controlling the effects of intelligence (Hypothesis 7). a partial correlation was employed. Hypotheses 8-11. the intrasexual compari- sons. were tested with a two-way analysis of variance. Sum The methodology used in the study was described in this chapter. The population comprised students attending a midwestern open door community college. They were predominantly from middle-class homes. were Caucasian. and were Americans by birth. The instruments used for measuring reading ability. personality traits. and intelligence were described individually. The reliability coefficients of the instruments and their derivation. as well as a description of the individual subscales and what they purport to measure. were included. A description of how the study population was 51 grouped for comparison and the statistical procedures used to test the hypotheses was also included. Chapter IV contains a presentation and (analysis of the data gathered in this study. CHAPTER IV PRESENTATION AND ANALYSIS OF DATA The purpose of this study was to analyze the relationships between reading and personality and intelligence scores of selected open door community college students and to explore the development of a personality and achievement profile that will help practitioners assist students in realizing their potential. The methodology used in collecting and analyzing the data was described in Chapter III. This chapter contains the findings of the statistical analyses of the data as they relate to the hypotheses formulated for the study. WW The data concerning the relationship between total reading achievement and measures of personality and achievement were statis- tically compared using two-tailed Pearson correlation coefficients. Hypotheses 1 through 6 were tested with this procedures. WEI-21.5.1 There will be no statistically significant relationship between group total reading achievement and group scores on the individual scales of the Clinical Analysis Questionnaire. 52 53 Slight but statistically significant relationships (p < .05) were noted between the total grouphs reading achievement and the total group's scores on 9 of the 28 personality scales of the CA0: Domi- nance. Impulsivity. Imagination. Insecurity. Radicalism. Self- Sufficiency. Intelligence. Agitation. and Psychopathic Deviation. Therefore. Hypothesis 1 was rejected. Table 3 contains the correlation coefficients and the levels of significance for the nine CAQ scales with group total reading. Table 3.--Correlation of total reading achievement with personality: combined group. Degree of CAQ Scale Correlation p = (E) Dominance .269 .002 (F) Impulsivity .264 .003 (M) Imagination .274 .002 (0) Insecurity -.284 .001 (Q1) Radicalism .375 .001 ( ) Self-Sufficiency .317 .001 (B Intelligence .447 .001 (D3) Agitation .180 .043 (Pp) Psychopathic Deviation .250 .005 W There will be no statistically significant relationship between males' total reading achievement and their scores on the individual scales of the Clinical Analysis Questionnaire. Statistically significant relationships (p < .05) were found between mal es' total reading achievement and their scores on 4 of the 54 28 CAQ personality scales: Imagination. Radicalism. Self-Sufficiency. and Intelligence. Therefore. Hypothesis 2 was rejected. Table 4 contains the correlation coefficients and levels of significance for these four CAQ scales with males' total reading achievement. Table 4.--Correlation of total reading achievement with personality: males. Degree of CAQ Scale Correlation p = (M) Imagination .409 .003 (01) Radicalism .414 .003 (Q ) Self-Sufficiency .409 .003 (B1 Intelligence .428 .002 W There will be no statistically significant relationship between females"tota1 reading achievement and their scores on the individual scales of the Clinical Analysis Questionnaire. Statistically significant relationships (p < .05) were found between females"total reading achievement and their scores on 9 of the 28 CAQ personality scales: Dominance. Impulsivity. Shrewdness. Insecurity. Radicalism. Self-Sufficiency. Intelligence. Agitation. and Psychopathic Deviation. Therefore. Hypothesis 3 was rejected. Table 5 contains the correlation coefficients and the level of significance for these nine CAQ scales with females' total reading achievement. 55 Table 5.--Correlation of total reading achievement with personality: females. Degree of CA0 Scale Correlation p = (E) Dominance .343 .002 (F) Impulsivity .339 .002 (N) Shrewdness -.366 .001 (O) Insecurity -.420 .001 (O1) Radicalism .355 .001 (0.) Self-Sufficiency .279 .013 (8 Intelligence .475 .001 (D3) Agitation .277 .014 (Pp) Psychopathic Deviation .338 .002 11192111351541 There will be no statistically significant relationship between group total reading achievement and group performance on the individual subscales of the Wechsler Adult Intelligence Scale. Total group performance on all scales of the WAIS except Digit Symbol correlated significantly (p <:.05) with group total reading achievement. Therefore. Hypothesis 4 was rejected. Table 6 contains the correlation coefficients and level of significance for group performance on each of the WAIS subscales with group total reading achievement. W There will be no statistically significant relationship between males' reading achievement and their performance on the individual subscales of the Wechsler Adult Intelligence Scale. Statistically significant relationships (p < .05) were found between males"total reading achievement and their performance on 12 of 56 the 14 WAIS subscales. The only subscales that were not found to be significantly correlated with males' total reading achievement were Digit Symbol and Object Assemblyu Therefore. Hypothesis 5 was rejected. Table 7 contains the correlation coefficients and the levels of significance for males' performance on these 12 WAIS subscales with males' total reading achievement. Table 6.--Corre1ation of group total reading achievement with WAIS subscales (N = 127). Degree of WAIS Subscale Correlation p = Full Scale 10 .724 .001 Verbal IQ .753 .001 Performance 10 .466 .001 Information .653 .001 Comprehension .498 .001 Arithmetic .491 .001 Similarities .481 .001 Digits Forward .270 .002 Digits Backward .266 .002 Combined Digits .338 .001 Vocabulary .763 .001 Digit Symbol Not significant Picture Completion .423 .001 Block Design .350 .001 Picture Arrangement .396 .001 Object Assembly .247 .001 57 Table 7.--Correlation of males' total reading achievement with WAIS subscales (N = 49). Degree of WAIS Subscale Correlation p = Full Scale IQ .765 .001 Verbal 10 .823 .001 Performance IQ .443 .001 Information .737 .001 Comprehension .582 .001 Arithmetic .564 .001 Similarities .564 .001 Combined Digits .288 .045 Vocabulary .783 .001 Picture Completion .369 .009 Block Design .316 .027 Picture Arrangement .455 .001 amnesia There will be no statistically significant relationship between females' reading achievement and their performance on the individual subscales of the Wechsler Adult Intelligence Scale. Females' performance on all WAIS subscales except Digit Symbol correlated significantly (p < .05) with females"tota1 reading achieve- ment. Therefore. Hypothesis 6 was rejected. Table 8 contains the correlation coefficients and the levels of significance for females' performance on each of the WAIS subscales with females' total reading achievement. 58 Table 8.--Correlation of females' total reading achievement with WAIS subscales (N = 78). Degree of WAIS Subscale Correlation p = Full Scale IQ .718 .001 Verbal 10 .735 .001 Performance IQ .488 .001 Information .626 .001 Comprehension .471 .001 Arithmetic .495 .001 Similarities .411 .001 Digits Forward .286 .011 Digits Backward .335 .003 Combined Digits .380 .001 Vocabulary .762 .001 Digit Symbol not significant Picture Completion .485 .001 Block Design .421 .001 Picture Arrangement .356 .001 W There will be no statistically significant relationship between group reading achievement and group scores on the individual scales of the Clinical Analysis Questionnaire when the effects of IQ have been controlled. Hypothesis 7 was tested using zero-order partial correlations. Slight but statistically significant relationships between group total reading achievement and group scores on the CA0 were noted on 10 of the 28 personality scales before FIQ was controlled and on 8 of the 28 scales after FIQ was controlled. When the group was divided according to sex. statistically significant relationships were found between males' reading achieve- ment and their scores on 4 of the 28 personality scales before 59 controlling for FIQ. Slight but significant relationships existed between males' reading achievement and their scores on 5 of the 28 personality scales after the effect of IQ was controlled. In addition. certain CAQ scales that were significantly related to reading achieve- ment for the males before IQ was controlled were not statistically significant after FIQ was controlled. Sinfllar before-and-after scale shifts were noted for the female group. Before IQ was controlled. the females' reading achievement was significantly related to their scores on 12 of the 28 CAQ scales. After controlling for 10. statistically significant relationships were noted between females reading achieve- ment and their scores on 8 of the 28 CAD scales. Tables 9 through 11 contain comparisons of the correlation coefficients and the levels of significance obtained for the total group. for males. and for females between total reading achievement and scores on the CAQ personality scales. before and after IQ was controlled. As shown in Table 9. with FIQ controlled. the factors Emotional Stability and Conformity were slightly but significantly related to group total reading achievement. whereas Warmth. Dominance. Imagina- tion. Self-Sufficiency. and Intelligence were no longer significantly related to reading achievement. For the male group. control of IQ resulted in the factors Conformity. Shrewdness. Boredom and Withdrawal. and Psychosthenia being significantly related to total reading achievement. Imagination. Self- Sufficiency. and Intelligence were no longer significantly related to 60 a a 8mm. .>mo u_cumao;u>ma A .4 Noo. omN. .>oo u_gomao;u>ma 4 a4 Noo. mNO. me_. =o_omu_m< Amov ONO. cm_. co_umo_m< Amov .oo. Ass. Q8:03:85 Amv .00. “.m. >ucu_u_..=m-._om Awav mmo. _e_. Em__mu_am¢ 4.04 .00. mxm. sm__mu_emx 4_u4 mmo. wN~.- >D_Lsuamc_ 404 .oo. smN. >D_.=um.c_ 404 _oo. mum. co_umc_mme_ 4:4 mNo. mn_. mmace_om 4:4 _oo. $4.. .macu_om 4:4 owe. mw_. >._eLo.coo Auv .oo. mmN. mocmc_eoo 444 omo. 54.. .stm .mco_oo2m 484 ~mo. os_. EosLmz Aqv . . 1%.. 2... s. T . Emma 23. s. eo__oL.cou a. ea__oL.=ou .oz 0. .Am~_ u 24 o. Lem mc___oLucou Leumm ncm OLOWOB .>u__mc0mcoa ncm acesm>o_£um mc_cmoL .maOD osocm mo co_4m_oLLOuun.m o_nmh 61 omo. NNN. m_co;.mo;u>ma 4 <4 . . .mzmce;._3 mso msN . eoeocom ANGV _oo. mNs. 08.323.: Amy MNo. omN. s.__mu_emm A_a4 Noo. mos. suco_u_..=m-c_mm ANov mNo. smN. .mmcezoLcm 424 N00. s_s. sm__mu_em¢ 4.04 s_o. m_m. >._eLoLEou Auv Noo. mo:. co_Dm:_mme_ 4:4 I .LLoU I .LLOQ I a mo ooLmoo o_mum du__mcomcoa vcm acmeo>o_;um mc_umoL .muOH o_mE mo co_um_oLL0uII.o_ o_nmh 62 NOO. .Nm. .>OO u.;.mao;u>m. .O.. .OO. Omm. .>OO u.:.mao;u>m. .O.. O.O. NsN. m.Ocm.m. .m.. OMO. NON. .m3m.nc..3 O soOoLoO .NO. ONO. MON. co..mO.O< .MO. NOO. .NN. OO.OmO.O< .mO. .OO. ONO. OOOOO...OOO_ .O. .OO. ONN. >uco.u...sm-..mm .NO. ONO. NON.- co..co. .OO. .OO. mmm. s...mu.emx ..O. .OO. mmm.- >.7582. .O. .OO. ON..- >u..suo.c_ .O. .mO. .O..- mmacezm.;m .2. .OO. OOM.- .moceza.;m .2. OOO. Om.. OO.OOO.OOE_ .z. mmO. m... mmmce.oO .1. N.O. OmN. >£3.39... ... .OO. mmm. >D.>...OOE_ ... NOO. .ON. OOOOO.EOO .O. .OO. mam. v2.2.8.. .4. u a wo.mmmmoo o_mom du__mc0m.oa ccm ucoso>o_£um mc_umo. .mHOH o_mEom we co_um_oL.oUII.._ v.4mh 63 reading achievement when IQ was controlled. (kfly'Radicalism remained as statistically significant when 10 was controlled. (See Table 10.) For the female group. when 10 was controlled. Tension and Paranoia were significantly related to total reading achievement. whereas Boldness. Imagination. Radicalism. Self-Sufficiency. Intelli- gence. and Boredom and Withdrawal no longer were significantly related to reading achievement. (See Table 11.) Whether the study population was considered as a combined group or grouped by sex. statistically significant relationships were observed between their reading achievement and their scores on the CAQ personality scales when IQ was controlled or uncontrolled. When IQ was controlled. a slightly different relationship existed between person- ality and reading achievement: nevertheless. the relationship was sta- tistically significant. Therefore. Hypothesis 7 was rejected. Statistical comparisons for Hypotheses 8 through 11 were con- ducted using the analysis of variance technique. Results are reported in the following paragraphs. Hypothesjfl There will be no statistically significant difference between better-reading males and poorer—reading males on the individual scales of the Clinical Analysis Questionnaire. Statistically significant differences (p < .05) were found between better-reading and poorer-reading males on four CAQ scales: Imagination. Radicalism. Self-Sufficiency. and Intelligence. There- fore. Hypothesis 8 was rejected. Table 12 shows the mean sten scores for the better-reading and poorer-reading males on the CAQ scales on 64 which there were significant differences and the levels of significance for each factor. Table 12.--Mean sten scores and significance levels for better-reading and poorer-reading males on the CAQ. CAQ Scale Upper (N=19) Lower (N=22) X Sten Score X Sten Score Signif. (M) Imagination 6.263 5.142 .039 (0]) Radicalism 6.789 5.285 .018 ( ) Self-Sufficiency 7.105 5.476 .004 (8 Intelligence 6.578 5.285 .014 .HxnfliheitiJi There will be no statistically significant difference between better-reading females and poorer-reading females on the individual scales of the Clinical Analysis Questionnaire. Statistically significant differences (p < .05) were found between better-reading and poorer-reading females on nine CAQ scales: Dominance. Impulsivity. Shrewdness. Insecurity. Radicalism. Self- Sufficiency. Intelligence. Agitation. and Psychopathic Deviation. Therefore. Hypothesis 9 was rejected. Table 13 shows the mean sten scores for better-reading and poorer-reading females on these nine CAQ scales and the levels of significance for each factor. 65 Table 13.--Sten scores and significance levels for better-reading and poorer-reading females on the CA0. CAQ Scale Upper (N=32) Lower (N=32) X Sten Score X Sten Score Signif. (E) Dominance 6.281 5.250 .051 (F) Impulsivity 6.343 5.218 .012 (N) Shrewdness 4.375 5.593 .002 (O) Insecurity 4.375 5.593 .011 (01) Radicalism 6.250 4.687 .003 (Q 1 Self-Sufficiency 6.281 4.937 .012 (8 Intelligence 6.312 4.812 .0001 (03) Agitation 6.156 4.906 .043 (Pp) Psychopathic Deviation 6.687 5.125 .032 .H¥291h§§1&_lfl There will be no statistically significant difference between better-reading males and poorer-reading males on the subscales of the Wechsler Adult Intelligence Scale. Statistically significant differences (p < .05) existed between better-reading and poorer-reading males on all of the WAIS subscales except Digit Span and Object Assembly. Therefore. Hypothesis 10 was rejected. Table 14 shows a comparison of the mean scale scores for each group of male readers. the numerical differences between the mean scale scores. and the levels of significance for those differences. 66 Table l4.--Significant differences between better-reading and poorer- reading males on the WAIS. Upper (N=19) Lower (N=22) X Scale Score X Scale Score Diff. Signif. WAIS Subscale FIQ 122.42 106.86 15.56 .0001 V10 122.74 104.48 18.26 .0001 PIQ 118.74 108.67 10.07 .0014 Information 13.58 9.57 4.01 .0001 Comprehension 15.43 12.43 3.00 .0002 Arithmetic 13.53 10.19 3.33 .0001 Similarities 14.21 11.67 2.54 .0002 Digit Span 12.47 11.42 1.05 n.s. Vocabulary 14.26 9.71 4.55 .0001 Digit Symbol 12.21 11.57 0.64 n.s. Picture Completion 13.37 11.86 1.51 .0170 Block Design 13.90 12.29 1.61 .0521 Picture Arrangement 12.21 9.71 2.50 .0005 Object Assembly 12.47 11.71 0.76 n.s. flxnothes1s_11 There will be no statistically significant difference between better-reading females and poorer-reading females on the subscales of the Wechsler Adult Intelligence Scale. Statistically significant differences (p < .05) were found between better-reading and poorer-reading females on all of the WAIS subscales except Digit Symbol. Therefore. Hypothesis 11 was rejected. Table 15 shows a comparison of the mean scale scores for each group of female readers. the numerical differences between the mean scale scores. and the level of significance for those differences. 67 Table 15.--Significant differences between better—reading and poorer- reading females on the WAIS. Upper (N=32) Lower (N=32) X Scale Score X Scale Score Diff. Signif. WAIS Subscale FIQ 117.94 104.47 13.47 .0001 V10 116.34 101.81 14.83 .0001 PIQ 117.31 107.41 9.90 .0001 Information 11.97 8.97 3.00 .0001 Comprehension 14.56 11.72 2.84 .0001 Arithmetic 11.72 8.97 2.75 .0001 Similarities 13.19 11.81 1.38 .0027 Digit Span 12.13 10.65 1.48 .0217 Vocabulary 13.13 9.41 3.72 .0001 Digit Symbol 14.00 13.59 0.41 n.s. Picture Completion 12.75 10.66 2.09 .0001 Block Design 12.47 10.25 2.22 .0002 Picture Arrangement 10.88 9.72 1.16 .0235 Object Assembly 12.19 10.10 2.09 .0010 thauflflnflihi_LZ There will be no statistically significant relationship between total group scores on the individual scales of the Clinical Analysis Questionnaire and total group scores on the subscales of the Wechsler Adult Intelligence Scale. Statistically significant relationships (p < .05) were found between the total group's scores on the individual scales of the CA0 and the group's scores on the subscales of the WAIS. Tables 16. 17. and 18 contain the correlation coefficients and levels of significance between scores on the CAQ personality scales and scores on the WAIS subscales for the combined group. males. and females. respectively. Only those scales are included on which a statistically significant relationship was found. 68 Table 16cm-Significant relationships between CAQ scales and WAIS subscales for the total group. CAQ Scale WAIS Subscale (A) Warmth FIQ: -.191; p = .031 VIQ: -.171; p = -.171 (E) Dominance (F) Impulsivity (G) Conformity (H) Boldness (I) Sensitivity (M) Imagination (N) Shrewdness Arithmetic: -.244; p = .006 FIO: .271; p = .002 VIQ: .242; p = .006 FIG: .220; p = .013 Information: .218; p = .014 Comprehension: .262; p = .003 Vocabulary: .261; p = .003 Digit Symbol: .183; p = .038 Picture Completion: .251; p = .004 FIQ: .226; p .010 VIQ: .235; p .008 Information: .194; p = .029 Comprehension: .329; p = .001 Vocabulary: .211; p .017 Digit Span: .173; p .038 Digit Symbol: -.209; p = .018 Picture Completion: -.246; p = .005 Picture Arrangement: -.205; p = .02 Comprehension: .232; p = .008 Digit Symbol: -.172; p = .053 FIQ: .253; p = .004 VIO: .308; p = .001 Information: .270; p = .002 Comprehension: .300; p = .001 Arithmetic: .175; p = .049 Similarities: .206; p = .020 Vocabulary: .336; p = .001 FIQ: -.218; p .014 P10: -.244; p .006 Vocabulary: -.l74; p = .050 Picture Completion: -.201; p = .023 Block Design: -.180; p = .040 Picture Arrangement: -.252; p = .004 Table 16 # CAQ # (O) Ins ((11) Rac (Q4) Ter (B) In: Table 16.-~Continued. 69 CAQ.Scale WAIS Subscale (O) Insecurity (Q1) Radicalism (02) Self-Sufficiency (Q4) Tension (B) Intelligence Digit Symbol: -.236; p = .008 FIG: .375; p = .001 VIO: .381; p = .001 P10: .278; p = .002 Information: .364; p = .001 Comprehension: .254; p = .004 Arithmetic: .312; p = .001 Similarities: .229; p = .009 Vocabulary: .410; p = .001 Block Design: .257; p = .003 Picture Arrangement: .174; p = .050 Object Assembly: .287; p = .001 FIQ: .369; p = .001 VIO: .325; p = .001 P10: .301; p = .001 Information: .288; p = .001 Arithmetic: .346; p = .001 Digits Forward: .214; p = .015 Vocabulary: .383; p = .001 Picture Completion: .232; p = .009 Block Design: .280; p = .001 Picture Arrangement: .225; p = .011 Object Assembly: .265; p = .003 FIQ: .216; p .015 VIQ: .217; p .014 Information: .200; p = .024 Comprehension: .201; p = .230 Arithmetic: .201; p = .024 Similarities: .200; p = .024 Vocabulary: .181; p = .041 Block Design: .236; p = .008 FIQ: .570; p = .001 VIQ: .563; p = .001 P10: .410; p = .001 Information: .549; p = .001 Comprehension: .348; p = .001 Arithmetic: .448; p = .001 Similarities: .313; p = .001 70 Table 16.--Continued. CAQ Scale WAIS Subscale (8) Intelligence (cont'd) Digits Forward: .218; p = .014 Digits Backward: .193; p = .029 Vocabulary: .516; p = .001 Picture Completion: .387; p = .001 Block Design: .337; p = .008 Picture Arrangement: .264; p = .008 Block Assembly: .358; p = .001 (Dz) Suicidal Depression Picture Completion: .175; p = .049 (D7) Boredom and Information: .177; p = .046 Withdrawal Vocabulary: .177; p = .046 Block Design: .168; p = .059 Object Assembly: .175; p = .049 Table 16 shows that primarily slight correlations existed between the total group's scores on the CAQ scales and their scores on the WAIS subscales. Some CAQ scales correlated with more than half of the WAIS subscales. Cattell's Intelligence scale correlated moderately with all WAIS scales except Digit Symbol. The Intelligence scale is a power measure and requires logical-reasoning capabilities and some reading ability. Correlations between the Intelligence scale and cer- tain WAIS subscales such as Similarities. Block Design. and Object Assembly could indicate a common factor of logical-reasoning ability. This might help explain why poor readers have mixed results on the Similarities subtest. Those poor readers who have logical-reasoning abilities and who order their thinking in this manner should do well on the subscales demanding logical relationships. It is interesting that 71 the Intelligence scale correlated with the Arithmetic subscale more highly than it did with PIQ. This lends some support to the suggestion that the Intelligence scale measures a logical-reasoning component of intelligence. The CAQ scales Radicalism and Self-Sufficiency correlated with 11 WAIS measures. Radicalism correlated most highly with Vocabulary. VIQ. Information. and Arithmetic. ‘This finding fits the description of the Radicalism scale in the Handbook for the 16 PF (1970)--that "[Radicalism +1 persons are more well informed. more inclined to experiment with problem solution. less inclined to moralize. less unquestioning about views generally. eth'(p. 104). Self-Sufficiency correlated most strongly with Vocabulary. FIQ. VIQ. Arithmetic. and P10. Again. these findings support the statement in the Handbook for the 16 PF--that "[Self-Sufficiency] is a constant and very significant contributor to scholastic success" (p. 105). Dominance and Imagination correlated with eight and seven WAIS subscales. respectively. Imagination correlated more strongly with the Vocabulary. VIQ. and Comprehension subscales than did Dominance. Negative correlations were found between five CAQ scales-- Warmth. Sensitivity. Conformity. Shrewdness. and Insecurity-~and subscales of the WAIS. Shrewdness was slightly negatively related to seven WAIS subscales; the strongest inverse relationship was with Picture Arrangement. Conformity was negatively related to three performance subscales. The strongest relationship was with Picture Completion. 72 Tension attained slight but significant relationships with eight WAIS subscales. the strongest of which was with Block Design. a speeded test. Cattell called the Tension scale the "ergic tension factorJ' High-scoring individuals are tense. frustrated. driven. and overwrought. The CAQ Manual (1980) suggests that higher scores on Tension can be "associated with frustrated motivation" q» 170. In any event. the relationships were slight and could suggest a curvilinear relationship. such as was suggested for Anxiety. Of the clinical traits. Suicidal Depression and Boredom and Withdrawal had slightly significant relationships with subscales of the WAIS. The relation of Suicidal Depression with Picture Completion and of Boredom and Withdrawal with Information.‘Vocabu1ary. Object Assembly. and Block Design can also be interpreted on an introversion- extraversion continuunL High-scoring individuals would be less apt to interact with other people (CAQ Manual. 1980). As shown in Table 17. males' scores on the CA0 were correlated with their scores on the WAIS. Scores on 7 of the 16 normal person- ality scales had statistically significant relationships with scores on the various subscales of the WAIS. Intelligence correlated with 11 subscales. followed by Self- Sufficiency with eight correlations and Imagination with six correlations. Radicalism and Dominance followed with four and three correlations. respectively. whereas Impulsivity and Boldness correlated with one subscale each. in! 73 Table 17.--Significant relationships between CAQ scales and WAIS subscales for males. CAQ Scale WAIS Subscale (E) Dominance (F) Impulsivity (H) Boldness (M) Imagination (Q1) Radicalism (02) Self-Sufficiency (B) Intelligence (01) Hypochondriasis FIQ: .313; p .028 FIG: .316; p .027 Digit Symbol: .287; p = .045 Digit Symbol: .445; p = .001 Comprehension: .293; p = .041 FIQ: .363; p .010 VIO: .423; p .002 Information: .431; p = .002 Comprehension: .303; p = .034 Arithmetic: .279; p = .052 Vocabulary: .435; p = .002 FIQ: .308; p = .031 VIQ: .315; p = .027 Information: .337; p = .018 Vocabulary: .320; p = .025 FIQ: .446; p = .001 VIQ: .346; p .015 P10: .418; p .003 Information: .320; p = .025 Arithmetic: .331; p .020 Vocabulary: .508; p .001 Picture Completion: .408; p = .004 Picture Arrangement: .424; p = .002 FIQ: .627; p = .001 VIO: .588; p = .001 PIQ: .489; p = .001 Information: .576; p = .001 Arithmetic: .569; p = .001 Similarities: .426; p = .002 Vocabulary: .536; p = .001 Picture Completion: .368; p = .009 Block Design: .334; p = .009 Picture Arrangement: .484; p = .002 Object Assembly: .333; p = .019 VIQ: -.270; p .035 .052 74 CAQ Scale WAIS Subscale (D3) Agitation Digits Forward: -.281; p = .050 (D4) Anxious Depression FIQ: -.332; p = .020 VIQ: -.308; p .031 (05) Low Energy Depression Digits Backward: -.303; p = .034 (D7) Boredom and Withdrawal Digits Backward: -.380; p = .007 (Sc) Schizophrenia Digits Backward: -.283; p = .048 (As) Psychosthenia Digits Backward: -.306; p = .032 On the clinical scale. males' scores on six scales had statistically significant correlations with their scores on WAIS subscales; all of the correlations were negative. Anxious Depression was negatively correlated with three subscales. whereas Hypochondriasis was correlated negatively with two subscales. The rest of the clinical scales correlated negatively with digit subtests; Low Energy Depres- sion. Boredom and Withdrawal. Schizophrenia. and Psychosthenia corre- lated with Digits Backward; Agitation correlated negatively with Digits Forward. It is interesting that Self-Sufficiency correlated more highly with PIQ than it did with VIQ even though Self-Sufficiency was more highly correlated with Vocabulary than with any other subscale. Also. Picture Arrangement correlated more highly with Self-Sufficiency than 75 with PIQ. When the correlation between Intelligence and Picture Arrangement was considered. the strength of the correlation was greater than that of any of the other performance subscales except P10. The strength of the correlation was slightly greater than for Similarities and again suggests that a common element may be present among Similari- ties and Picture Arrangement; it could be a verbal-reasoning or logical-reasoning element. but it is only suggestive. Another interesting relationship was the moderate correlation between Impulsivity and the Digit Symbol subscale. The speeded nature of the Digit Symbol test requires both speed and accuracy of response. Impulsivity may include quickness but not necessarily accuracy. A seemingly better description of this scale was contained in the old 16 PF Handbook (1970). in which the general scale high score was entitled Surgency and included as a description "quick and alertP (p. 87). This makes the relationship more understandable. Considering the CAQ scales that correlated positively or negatively with FIQ. a picture emerges of the male in the study who did well on the WAIS. He was intelligent. self-sufficient. imaginative. and radical or free thinking; exhibited dominance; viewed himself as healthy; and was relatively free of anxious depression. As shown in Table 18. females' scores on the CAQ were corre- lated with their scores on the WAIS. Scores on 12 of the»l6 normal personality scales had statistically significant relationships with scores on the various subscales of the WAIS. 76 Table 18.--Significant relationships between CAQ scales and WAIS subscales for females. CAQ Scale WAIS Subscale (A) Warmth Arithmetic: -.273; p = .015 (E) Dominance Comprehension: .235; p = .038 Vocabulary: .269; p = .017 Digit Symbol: .263; p = .020 Picture Completion: .248; p = .028 (F) Impulsivity FIQ: .230; p .042 VIQ: .274; p .015 Comprehension: .381; p = .001 Vocabulary: .292; p = .009 Picture Completion: .263; p = .020 (H) Boldness Vocabulary: .231; p = .041 (I) Sensitivity Arithmetic: -.221; p = .051 Digit Symbol: .215; p = .058 (L) Suspiciousness FIQ: .242; p = .033 VIQ: .217; p = .055 Information: .314; p = .005 Comprehension: .250; p = .027 p (M) Imagination Comprehension: .262; = .020 Vocabulary: .251; p = .026 (N) Shrewdness FIQ: -.345; p = .002 VIQ: -.310; p = .006 Information: -.277; p = .014 Comprehension: —.221; p = .052 Similarities: -.266; p = .018 Vocabulary: -.279; p = .013 Picture Completion: -.295; p = .009 Block Design: -.244; p = .031 Picture Arrangement: -.281; p = .012 (0) Insecurity FIQ: -.248; p = .028 VIQ: -.247; p = .029 Information: -.232; p = .041 Similarities: -.228; p = .045 Vocabulary: -.301; p = .007 Digit Symbol: -.248; p s .028 Picture Completion: -.253; p = .025 Table 18.--Continued. 77 CAQ Scale WAIS Subscale (Q1) Radicalism (02) Self-Sufficiency (8) Intelligence (02) Suicidal Depression (03) Agitation (D4) Anxious Depression (05) Low Energy Depression FIQ: .406; p = .001 vm: .414; p = .001 P10: .288; p = .010 Information: .370; p = .001 Comprehension: .281; p = .013 Arithmetic: .378; p = .001 Similarities: .245; p = .030 Vocabulary: .301; p = .007 Block Design: .367; p = .001 Object Assembly: .345; p = .002 FIQ: .306; p = .006 VIQ: .282; p = .012 P10: .237; p = .036 Information: .233; p = .040 Arithmetic: .317; p = .005 Digits Forward: .252; p = .026 Vocabulary: .297; p = .008 Block Design: .303; p = .007 Object Assembly: .268; p = .017 FIQ: .519; p = .001 VIQ: .530; p = .001 P10: .357; p = .001 Information: .513; p = .001 Comprehension: .372; p = .001 Arithmetic: .344; p = .002 Similarities: .219; p = .053 Digits Forward: .246; p = .029 Digits Backward: .216; p = .057 Vocabulary: .489; p = .001 Picture Completion: .377; p = .001 Block Design: .304; p = .007 Object Assembly: .357; p = .001 Block Design: .303; p = .007 Similarities: .222; p = .050 Picture Arrangement: .214; p = .059 Information: .248; p = .028 Object Assembly: .226; p = .046 Block Design: .251; p = .02 Object Assembly: .276; p = .014 78 Table 18.--Continued. CAQ Scale WAIS Subscale (D7) Boredom and FIQ: .339; p = .002 Withdrawal VIQ: .311; p = .006 PIQ: .276; p .014 Information: .321; p = .004 Arithmetic: .321; p = .004 Similarities: .215; p = .058 Digits Forward: .269; p = .017 Vocabulary: .246; p = .030 Block Design: .424; p = .001 Object Assembly: .396; p = .001 (Pa) Paranoia Block Design: .299; p = .008 (Pd) Psychopathic Deviation Vocabulary: .274; p = .015 (Sc) Schizophrenia Block Design: .355; p = .001 (AS) Psychosthenia Block Design: .276; p = .014 (P5) Psychological Inadequacy Block Design: .220; p = .052 Object Assembly: .249; p = .028 As with the males. Intelligence correlated with the most subscales of the WAIS among the females-~13. counting the Digit Span sections separately. Here the comparison of females with the males ends. The females' next four normal CAQ scales that correlated significantly with WAIS subscales were Radicalism (10 subscale correlations). Shrewdness (10 subscale correlations). Self-Sufficiency (9 subscale correlations). and Insecurity (7 subscale correlations). Most of the correlations were slight; a few were moderate. Apart from Intelligence. the strongest correlations for the normal CAQ scales were between Radicalism and the WAIS subscales. 79 Among the clinical CAQ scales. the femal es' scores on the CAQ scales were significantly correlated with their scores on 10 of the 12 WAIS subscales. Boredom and Withdrawal was correlated with 10 WAIS subscales; some of the correlations approached moderate strength. Agitation. Anxious Depression. Low Energy Depression. and Psychological Inadequacy correlated with two WAIS subscales each. Apart from Boredom and Withdrawal. most of the correlations between the clinical scales of the CA0 and the WAIS subscales were with performance subscales of the WAIS. Most of these correlations were with Block Design and Object Assembly. It should be noted that. to be clinically significant. the scales of the CAQ require scores in the l to 3 range at the low end and in the 8 to 10 range at the high end (CAQ Manual. 1970. p. 11). Of the females in the study group. only a few had clinically significant scores at either end of the spectrum. In this case the correlations with Boredom and Withdrawal could more accurately mean the sample females tended to be introverted (CAQ Manual. 1970. p. 19). The interesting aspect of the females' CPO scores correlating with the WAIS subscales is the number of scales between which correla- tions existed. The two CAQ scales that were negatively related to a number of WAIS subscales indicated that insecure. socially astute women seemed to do poorly on the WAIS. whereas secure. socially unpretentious women seemed to do better. Taking this analysis a step further. by looking at the CAQ scales that correlated positively or negatively with FIQ. it might be stated that the female in this sample who did well on 80 the WAIS was intelligent. radical or free thinking. self-sufficient. introverted. socially unpretentious. and self-assured. Statistically significant relationships existed between sub- jects' scores on the CAQ and their scores on the WAIS subscales. whether the sample was examined as a combined group or by sex. Therefore. Hypothesis 12 was rejected. Summanx Hypotheses 1 through 6 were tested using a two-tailed Pearson correlation coefficient procedure. All six null hypotheses were rejected. {Statistically significant relationships were found between subjects' total reading achievement and their performance on the CAQ personality scales. and between subjects' total reading achievement and their scores on WAIS subscales. The statistically significant rela- tionships existed when the study population was examined as a total group and when grouped by sex. When subjects were grouped by sex. the relationships between total reading achievement and personality factors were intensified and clarified. Hypothesis 7 was tested with a zero-order partial correlation technique. The hypothesis was rejected. Statistically significant relationships were found between subjects' total reading achievement and their performance on individual scales of the CAQ when the effects of IQ were controlled. This relationship existed when the study population was combined into a single group and when grouped by sex. 81 Certain scale shifts were noted in the IQrcontrolled condition as compared to the IQ-uncontrolled condition. CAQ scales that were significant in the uncontrolled condition were no longer significant in the controlled situation. but other. less-significant. relationships were found. This shift occurred for the combined group and for both the male and female groups. Hypotheses 8 through 11 were tested using an analysis of variance technique. Each of these hypotheses was rejected. Statis— tically significant differences existed between the better—reading and poorer-reading students in terms of their scores on the CAQ scales and their performance on the WAIS subscales. 'These differences were found for the combined group of better-reading versus poorer—reading sub- jects. as well as for intrasexual comparisons of better and poorer readers. Hypothesis 12 was tested using a two-tailed Pearson correlation coefficient. The hypothesis was rejected. Statistically significant relationships existed between subjectsfl scores on the CA0 personality scales and their performance on the WAIS subtests. These relationships existed when the study population was combined into a single group and when grouped by sex. These relationships were intensified and clari- fied in the intrasexual comparisons. Chapter V contains a summary of the findings and recommenda- tions for further research. In addition. the findings of this study are compared with those of previous research. CHAPTER V SUMMARY. CONCLUSIONS. AND RECOMMENDATIONS Summary The purpose of this study was to analyze the relationships between reading and personality and intelligence scores of selected open door community college students. The writer also explored the development of a personality and achievement profile for open door community college students that could help clinicians understand the strengths and weaknesses of the students with whom they work. Such a profile would help the practitioner assist students in realizing their potential. The review of the literature contained findings of research on the relationship between reading achievement and personality factors at the elementary-school. secondary-school. and college levels. Also included was an overview of reading achievement and individualized intelligence measures. especially focused on studies involving the Wechsler scales. In this study. data were obtained on 127 open door community college students. 'These students were Caucasian. predominantly middle class. and resided in the college district. The Nelson-Denny Reading Test (Form C) was used to obtain the subjects' total reading scores; the Clinical Analysis Questionnaire»(CAQ) and the Wechsler Adult 82 83 Intelligence Scale (WAIS) were used to measure personality and intelligence. respectively. The subjects for the study were categorized into comparison groups according to their total reading achievement on the Nelson- Denny Reading Test. Those obtaining scores one-quarter standard devia- tion above the mean (approximately the top 40% of the sample) were designated the better-reading students. whereas those scoring one- quarter standard deviation below the mean were designated the poorer- a reading students. Intrasex comparisons were made using the same cri- ‘3 terion. Correlations between total reading and the CA0. and between total reading and WAIS subscale scores. were obtained using a Pearson product-moment correlation. Intragroup comparisons between the better- reading and the poorer-reading students were made using the two-way analysis of variance technique. Other statistical procedures and groupings were used for comparisons with other research. conclusions The following 12 hypotheses were formulated to examine the relationships between reading and personality and reading and WAIS subscale scores: Hypothesis_1: There will be no statistically significant relationship between group total reading achievement and group scores on the individual scales of the Clinical Analysis Questionnaire. ‘Hypotnesjs_2: There will be no statistically significant relationship between males' total reading achievement and their scores on the individual scales of the Clinical Analysis Question- naire. 84 W: There will be no statistically significant relationship between femalesfi'total reading achievement and their scores on the individual scales of the Clinical Analysis Question- naire. .flypotnesis_4: There will be no statistically significant relationship between group total reading achievement and group performance on the individual subscales of the Wechsler Adult Intelligence Scale. Hypothesis_5: There will be no statistically significant relationship between males' reading achievement and their performance on the individual subscales of the Wechsler Adult Intelligence Scale. :-- . .flypothesis_§: There will be no statistically significant relationship between females' neading achievement and their performance on the individual subscales of the Wechsler Adult Intelligence Scale. .flypothe§1§_1: There will be no statistically significant relationship between group reading achievement and group scores on the individual scales of the Clinical Analysis Questionnaire when the effects of IQ have been controlled. hypothesis_8: “There will be no statistically significant difference between better-reading males and poorer—reading males on the individual scales of the Clinical Analysis Questionnaire. : There will be no statistically significant difference between better-reading females and poorer-reading females on the individual scales of the Clinical Analysis Questionnaire. .flypothesis_lfl: There will be no statistically significant difference between better-reading males and poorer-reading males on the subscales of the Wechsler Adult Intelligence Scale. tbuuuiuufljLJJ5 There will be no statistically significant difference between better-reading females and poorer-reading females on the subscales of the Wechsler Adult Intelligence Scale. ‘flypothesis_12: There will be no statistically significant relationship between total group scores on the individual scales of the Clinical Analysis Questionnaire and total group scores on the subscales of the Wechsler Adult Intelligence Scale. 85 All of the hypotheses were rejected. The correlational data indicated that statistically significant relationships between students' reading achievement and their scores on both the CAQ and the WAIS subscales did exist for this group of open door community college students. These relationships were clarified when the combined group was resolved into female and male groups. The relationships that were found between total reading achievement and the personality scales of the CA0 for this group of open door community college students are presented below. The trait descriptions are a condensed version of those contained in the Handbook for the 16 PF (1970. pp. 80-109). In the following discussion. the statistically significant personality traits are listed under the better-reader and poorer-reader categories. To possess all the charac- teristics of a trait. students would have to have a scale score of 10; for the total group. no score of 10 was recorded. Therefore. a scale characteristic of the better or poorer reader was the tendency toward the behavior of a particular personality scale. Because the relation- ships of the combined group were somewhat weaker than those reported by intrasexual comparison. the characteristics are reported only for males and females. The statistically significant relationships that existed between males' total reading achievement and personality characteris- tics are shown in Table 19. The personality variables shown in the table maintained a consistent relationship with total reading achieve- ment of male college readers except when IQ was controlled. With IQ 86 controlled. only Radicalism remained significant. whereas a set of weaker relationships became evident. It appears the personality vari- ables associated with the better—reading male students had a shared component with Intelligence. Table 17 indicated that the personality variables Imagination. Self-Sufficiency. and Intelligence correlated with the most WAIS subscales. Looking beyond the data. one might speculate that the common factor related to personality and IQ. as measured by these instruments. may be an organizing principle that orders perceptions and catalogues memory. In any event. at the college level. intelligence and personality variables need to be considered in tandem. Table 19.--Statistically significant relationships between males' total reading achievement and personality characteristics. Factor Upper-Reading Male Lower-Reading Male M Imaginative. unconventional. Practical. conventional. absent-minded and concerned with the immediate Q1 Experimental. free thinking. Conservative. respecting analytical established ideas 02 Self-sufficient. resource- Group dependent. follower full. prefers own decisions and joiner 8 Higher general mental Lower general mental capacity. fast learning and adaptable capacity. less able to handle abstract problems 87 The statistically significant relationships between females' total reading and personality characteristics are shown in Table 20. Table 20.--Statistically significant relationships between females' total reading achievement and personality characteristics. Factor Upper-Reading Female Lower-Reading Female E Assertive. aggressive. com- Submissive. dependent. mild. petitive. stubborn accommodating. easily led F Enthusiastic. talkative. Serious. full of cares. quick and alert slow. and cautious -N Socially clumsy. spontane- Polished. socially aware. ous. simple tastes. lacks artful. insightful. cuts self-insight corners -O Placid. secure. vigorous. Lonely. brooding. self- self-confident reproaching. insecure and worrying Q1 Experimental. analytical. Conservative. respecting and free-thinking established ideas 02 Self-sufficient. resource- Group dependent. a joiner ful. prefers own decisions and follower B High general mental Lower general mental capacity. fast learning capacity. less able to and adaptable handle abstract problems 03 Seeks excitement. is rest- less. takes risks. tries new things Pp Complacency from one's or others' anti-social behav- ior. is not hurt by criti- cism. enjoys emergencies and quarrels. 88 The personality variables shown in Table 20 maintained a consistent relationship with reading achievement of the females in the sample. When IQ was controlled using a zero-order partial correlation. only Radicalism. Self-Sufficiency. and Intelligence ceased being significant. as Self-Sufficiency. Imagination. and Intelligence had done for the males in the study. Again. those scales that ceased being significant had the most significant relationships with the various WAIS subscales. Comparison of the scores obtained by better and poorer readers on the Wechsler scales in the present study indicated that a rethinking of methodology may be in order. The findings clearly showed that the better-reading college student. whether male or female. scored significantly higher on practically all the WAIS subscales but Digit Symbol. Further. among this sample at least. no useful information was gained by totaling and averaging the scale scores and comparing that mean to each scale score to see whether it was lower or higher than the total mean. ‘The WAIS scale scores for the combined group. males. and females. are shown in Table»21. The scale scores are compared with the means of the scale scores. For the purposes of this study. a difference of one scale score indicates whether the group did better or worse on a particular scale than it did on the test. 'This will replicate the situation where any variation among subtest scores was considered significant. From Table 21 it is seen that the poorer readers had some of the reported patterns of P10 higher than VIQ; Low Information. High 89 mm.O. mO.N. 4O... N4.m. IN.O. mm.N. Imp: pamm m O..O. m m..N. m .N... m N4.N. m .N.O. m ON.N. ..IEOmp<.up.IO c080mcm.. .I.. .0 Io. N m NN.O 4 OO.O. 4 .N.m 4 .N.N. 4 NO.O 4 NM... O 0.3...” IO.I . .pgmm-N m mN.O. m N4.N. I ON.N. m OO.m. O O.... m OO.m. CO.mpO ouo.O :0 0 a 0 pamm m O0.0. m mN.N. O OO... O Nm.m. O O. .. m OO.N. .w.h.”.w asp. . ..OIO.I N I mm.m. I OO.I. m Nm... 4 .N.N. I mI.N. m mm.m. .OOEIm ..O.O cm.382. co 304 4 .s.m m m..m. 4 .N.m m ON.4. 4 44.m m mm.m. ..p.O.muo> 02mm m mO.O. m m..N. m N4... m N4.N. m Nm.O. m mN.N. Imam 4.O.O I-N .O-. Oox.I I .O... m O..m. O NO... m .N... I mN... m Nm.m. mp322.5; 4-N .O-. Omx.z 4 N0.0 m NN... m O..O. m mm.m. 4 Os.m m mm.N. O..OEI...< Oo.mOpIp.aeoO :0 IO.I I NN .. I Om.s. I ms.N. I ms.m. I II... I II... c222.838 co..ms.o.c. co 304 4 Nm.O O NO... 4 Nm.m m Om.m. 4 m..m Om.N. co..pe.o.c. .4.NO. .N.N.. NO.OO. 4N.O.. NO.NO. .N.N.. O.. O_> I... .OIO.I O_I . . .I .O. IN O.. OI.IO. .N.NN. MO.NO. MN.I.. O.> NI.IO. 4m.N.. OO.OO. N4.NN. NN.mO. .O.m.. O.. .c :00 m.0um0x m.0000m m.0000¢ n.0000m n.0cm0x m.0000¢ . o z .0304 .0aa: .0304 .0qa: .0304 .0co: ECLu— >50) O—MUW w_<3 m.0000m .00. m0_mE0u m0_mz 002.4504 .m0_mE0. 0:0 .m0_ms .o:0.m 00c_4eou .0» m0.0um 0.00m m_<3uu._N 0.40k 9O Comprehension. Low Arithmetic (females). High Similarities (females). Low Vocabulary. High Block Design (males). and Low Picture Arrangement (males). Table 21 appears to be similar to the table provided by Spache (1976. p. 139). but it is also evident from Table 21 that better readers scored higher in Comprehension and lower in Picture Arrangement and that. apart from Information and Vocabulary. little additional information can be gained from an analysis of poorer readers' subtest patterns. Even though PIQ was higher than VIQ for the poorer readers. the better-reading females had a PIQ higher than their VIQ. In any event. the 15 points required for significance were not there any more than was the required difference between subtests according to Newland and Smith's (1967) formula. One fact might be gleaned from this study of a community col- 1ege sample: The IQ profiles of the better readers were significantly better on those subscales relating to acquired knowledge. interest in the environment. and reasoning than those of their poorer-reading counterparts. Information. Arithmetic. Vocabulary. and Similarities are all skills Cohen (1959) placed in his Verbal Comprehension I com- ponent. All are necessary for school achievement. especially at the college level. It seems unlikely that an IQ profile can be developed for the open door community college as a result of this study. But as a result of this research it can be said that. in community colleges with similar characteristics. poorer readers will have average 105. with deficits in the language and reasoning areas. In relation to their 91 better-reading counterparts. poorer readers will have less information to draw on and a poorer vocabulary with which to work. They will also need to learn to reason and calculate. The personality profile of this population would reflect the sexual differences evident in this study. The better-reading male student would be more intelligent (IQ = 120). more imaginative. and more liberal and experimental in his thinking and would be decisive and resourceful. He would come to college equipped with a good vocabulary and reasoning abilities. The better-reading female student would be more intelligent (IQ = 115). more assertive and headstrong. more quick and alert. lack- ing self-insight. secure and self-confident. more liberal and experi- mental in her thinking. decisive and resourceful. restless and excitement seeking. have complacent attitudes toward anti-social behavior of self or others. and not be hurt by criticism. W The present study was not compared with research done at the elementary-school level. Maturational and developmental factors make comparisons between college and elementary-school students tenuous. at best. 'Therefore. this study was compared only with other research conducted at the college level. Anderson's (1961) correlational study indicated consistent relationships between reading achievement and personality factors of Intelligence. Conformity. Sensitivity. Imagination. Radicalism. and Self-Sufficiency. 'Total reading scores of the males were significantly 92 related to the personality factors of Intelligence. Conformity. Sensitivity. Imagination. Radicalism. and Self-Sufficiency. For the females. Anderson recorded significant relationships between reading achievement and the personality factors of Intelligence and Self- Sufficiency. In the present study. group reading achievement correlated with the personality factors of Dominance. Impulsivity. Imagination. Inse- curity. Radicalism. Self-Sufficiency. Intelligence. Agitation. and Anxious Depression. Anderson did not measure the clinical factors. but among the 16 PFs measured. Intelligence. Imagination. Radicalism. and Self-Sufficiency were in agreement between the two very different populations. In the present study. males' reading achievement corre- lated with the personality factors of Imagination. Radicalism. Self- Sufficiency. and Intelligence. By contrast. Anderson found relation- ships between males' neading achievement and the personality factors of Intelligence. Conformity. Sensitivity. Imagination. Radicalism. and Self-Sufficiency. Again there was partial agreement between the two study populations. Intelligence. Imagination. Radicalism. and Self- Sufficiency had statistically significant relationships with reading achievement in both studies. The correlation of females' reading achievement with their personality factors showed little agreement between the two studies. 'The present study found nine correlations of females' reading achievement with personality factors. Anderson's study listed only two. Intelligence and Self-Sufficiency. Both of these factors were related to reading achievement in the present study. 93 but 5 of the 16 personality factors found no comparison in the Anderson study. The Lockhart and Menary (1979) pilot study divided the study group by sex and then used Spearman correlations to detect significant differences. Among males. correlations were found between total reading achievement and the personality factors of Warmth. Intelligence. and Self-Sufficiency. The number of statistically significant relation- ships reported for this small sample (N = 22) differed from those in Anderson's research and in the present study but still included two of the descriptors. Intelligence and Self-Sufficiency. It is interesting that correlations for the better readers in the Lockhart and Menary study were reported in the direction of the correlations in the Ander- son study on the factors of Conformity. Sensitivity. and Radicalism. but they never reached significance with reading achievement. In the present study. the same indication existed for the males. ‘The small number of subjects in the pilot study might have resulted in these factors' not reaching statistical significance. Among the female subjects in Lockhart and Menary's pilot study. statistically significant correlations existed between reading achieve- ment and only two of the 16 PF scales. Imagination and Insecurity. Neither of these factors was found to be significant in the Anderson study. and the present study had only Insecurity in common with the pilot study. In an unpublished study. Menary and Lockhart enlarged their sample and used the CA0 to measure the personality variables. Although 94 this was not a correlational study. they held the difference of a sten score to be significant with reading and between high— and low-reading- achievement groups. Males' total reading achievement was significantly related to the personality scales Intelligence. Radicalism. Self-Sufficiency. Low Energy Depression. Boredom and Withdrawal. and Psychopathic Deviation. The Anderson (1961) study agreed with Intelligence. Radicalism. and Self-Sufficiency among the 16 PF normal scales but did not measure the clinical factors. The current study also agreed with the scales Intelligence. Radicalism. and Self-Sufficiency but did not find the clinical scales Agitation. Boredom and Withdrawal. or Psychopathic Deviation to be significant. Among the female subjects. the Menary and Lockhart study found total reading achievement related to the personality scales Intelli- gence. Impulsivity. Sensitivity. Imagination. Shrewdness. Insecurity. Radicalism. Self-Sufficiency. and Psychopathic Deviation. The current study found statistically significant relationships between females' total reading achievement and the CAQ scales Dominance. Impulsivity. Shrewdness. Insecurity. Radicalism. Self-Sufficiency. Intelligence. Agitation. and Psychopathic Deviation--a remarkable fit. The Anderson (1961) study. though. indicated only Intelligence and Self-Sufficiency as significant. whereas in the Lockhart and Menary pilot study. only Imagination. Insecurity. and Tension were found to be significant. Thus among the studies using the same basic instrument. certain personality scales seemed comnmuu Among the males. these were Intelli- 9S gence. Radicalism. and Self-Sufficiency. Among the females. they were Intelligence. Self-Sufficiency. Shrewdness. Insecurity. Radicalism. and Psychopathic Deviation. There was much agreement among the samples of community college students. However. had identical statistical tests been employed. greater agreement might have resulted. Raygor and Wark (1964) found that low-scoring males from the University of Minnesota reading center scored high on MMPI scales 4. 7. and 8. They lacked social skills. were shy and depressed. had vague goals. and were less verbal than the control males. ‘The females in the study were higher on Scale 7 (Psychosthenia) but otherwise were above average on adjustment. The present study supported Raygor and Warkfis findings for males in the areas of being less verbal and more group dependent; otherwise. the poorer-reading male in this study did not resemble his counterpart in the Minnesota group. On the other hand. the present study echoed Raygor and Warkfls findings for females; they indicated that poorer-reading females tended to be emotionally healthier than better-reading females. ‘The current study showed the poorer-reading female to be somewhat less secure than her better-reading counterpart. Aside from that. the better-reading female had significant relation- ships with two clinical scales. Agitation and Psychopathic Deviation. thus resembling the Minnesota women. Brunken and Shen (1968) found efficient and effective readers to be high on self-confidence. dominance. exhibition. autonomy. and SAT Verbal. whereas the poorer reader exhibited the opposite 96 characteristics. The present study agreed with the general findings concerning Dominance. Self-Sufficiency. Center of Attention (exhibi- tionl). Autonomy. and Radicalism. This seemed to be quite a good fit. In the current study. though. clarity was added when personality fac- tors were described according to intrasexual comparisons. From the preceding comparison it is evident that although differences existed among the studies there was also a great deal of agreement. Much of the dissimilarity can be attributed to the differences in methodology. but more discrepancies seemed to exist because of the differing populations being described. The studies conducted on open door community college students. using similar instruments. had similar results. Certainly. sample size and how the sample is divided both affect the results of a study. Comparing the McDonald study with the follow-up study illustrated this. as did comparing the Lockhart and Menary pilot study with the present study. In both cases. the earlier study indicated differences and direction; the follow-up study added definition. Discussioumfleoomonoatjons One purpose of this study was to explore the development of a personality and achievement profile for open door community college students that could help clinicians understand the strengths and weaknesses of the students with whom they work. A number of person- ality factors were found that had significant relationships with 97 reading achievement and differentiated the upper and lower reading groups. However. an achievement profile that could aid the clinician in developing remedial materials and programs was not discovered. If anything. the findings indicated that. among this group of open door community college students. the poorer-reading group did significantly more poorly on nearly all subscales of the WAIS than did the upper- reading group. In regard to an achievement profile for the poorer-reading groups. the WAIS subscales that were higher among the poorer readers tended also to be higher among the better readers. Thus. the only thing that can be said concerning achievement on the WAIS subscales is that the poorer readers in this study scored lower than the better readers on most measures and that a deficiency in verbal skills seemed most evident. Scores on Information and Vocabulary were uniformly low among the poorer readers. This finding lends support to Ekwall (1976). whose experience led him to question the worth of IQ testing to dis- cover the reading strengths and weaknesses of poorer readers. He wrote. The question we must ultimately ask .. . is: Do we get enough worthwhile information to warrant giving the test? This. of course. has to be answered by each diagnostician as he works with students in various testing situations. From personal experience. after administering over 500 WAIS. WISC and S-B tests to children and adults. the author's opinion is that seldom. if ever. is enough information obtained about the way a student worked or about his subtest scores to have justified taking time to give the test. . . . There are short vocabulary tests available that will give you an estimate of the student's word knowledge. And. might it not be better to observe the way the student works and reacts in a reading task than in an intelligence testing situation? (p. 173) 98 Based on the results of the present study. the writer would have to agree with Ekwall. But at the same time. the open door commu- nity college clinician must be aware that the poorer reader tends to score lower on intelligence tests than his better-reading counterpart and will probably need programs that develop his reasoning capabilities and broaden the base of information on which he can draw. At the same time. programs need to be developed that attend to the reader's personality. This can be accomplished by incorporating counseling programs into reading programs at the open door community colleges. The investigator was involved in a study (Burnside. McHolland. 8. Menary. 1978) that incorporated the preceding recommenda- tion and attempted to assess the programs of students in structured reading groups with human-potential seminars and of students who worked independently. Students who attended the reading lab were randomly assigned to one of four groups. which were then equated for reading leveJ. One group was assigned to the typical open-lab situation with the traditional material; the second group was assigned to an Open lab with motivational material; the third group received motivational mate- rials and a structured study-skills class; and the final group received a human-potential seminar along with a structured study-skills class. In all cases. gains were greater among students who were given the motivational material. The drop-out rate was lower for students in group situations than for those in the open lab situations. and the group with the human-potential seminar had the highest gains and the lowest drop-out rate. 99 Programs including a counseling component as a regular feature need to be developed. tested. and refined. The findings of this study justify further research into the role that personality plays in interaction with reading achievement. Particular attention should be focused on homogeneous groupings; socioeconomic and racial composition need to be controlled. This study was based on a white.Iniddle-class population; other studies of this nature are needed for community colleges with different populations. e4}. black. Hispanic. or lower socioeconomic levels. In addition. studies attempting to assess the effects of various remedial treatments need to be undertaken so that attention is focused on outcome rather than symptoms. Based on the findings of the present study. there are some possible implications for counselors. reading clinicians. and instructors at the open door community college. The counselor should be aware that a low reading score suggests a range of personality factors that may have to be dealt with before reading gains may be accomplished by the student. Male students might need help in becoming more imaginative. free thinking. logical. and able to formulate their own decisions. This could most easily be accomplished if a counseling component was built into the reading center. This component. handled by a trained counselor. could use a group-dynamics or human-potential format based on the poorer readerks personality needs. The reading clinician. being aware of the personality needs of the student. could develop strategies for using materials that present 100 subjects in a successful format. as well as materials that enable the student to see and then apply logical outcomes and independent decision making. Development of such materials would offer a challenge to the reading clinician. but the outcome would be worth the effort. Instructionally. the clinician. aware that the poorer readers may be more conservative and reluctant to question established ideas. would try to develop strategies designed to aid the students in critical reading. On the other hand. the clinician could aid the better readers in test-taking and study skills by moderating or channeling their assertive and aggressive natures toward positive outcomes. The instructor at the open door community college needs to be involved with the identification and referral of students to the reading clinician or counselor. Low grades and poor writjng ability may reflect deficits that could best be corrected in a study-skills and/or clinical setting. Prompt attention might even help in retention of the marginal student. APPENDIX NELSON-DENNY TABLE OF RELIABILITIES FOR GRADES 9-16 101 102 wEIcHTeo RAW scone VALUES 1 UNWEIGHTED RAW SCORE VALUES Standard Standard Error of Standard Reilabllltv Test Mean Deviation Measurement Mean Deviatoon 1 Co~ihcvant’ (n = 12453) GRADE 9 (n z 4269) - Vocabulary 17.04 9.88 3.14 16.86 9 52 9O Comprehension times two 28.00 11.02 4.92 28.20 10.90 80 Total (V + 2C) 45.03 18.86 5.85 45.07 18.38 90 Reading Rate 220.60 102 32 - - - - J! (n = 12087) GRADE 10 (n . 4442) 1‘ Vocabulary 20.17 12.47 3.27 20.10 11.70 .93 Comprehension times two 31.12 12.06 5.00 31.46 11.62 .83 I Total (V + 2C) 51.30 22.60 6.03 51.56 21.23 .93 ‘ Reading Rate 224.08 95.08 - - - — (n = 10130) GRADE 11 (n = 3589) Vocabulary 23.51 14.51 3.29 23.83 13.71 .95 Comprehension times two 34.00 12.38 4.92 23.86 13.86 .84 Total (V + 2C) 57.50 25.03 5.94 57.51 23.81 .94 ' Reading Rate 234.89 96.70 - - - — ! (n = 9518) GRADE 12 (n I 3368I Vocabulary 26.61 15.07 3.15 26 57 14.60 .96 f Comprehension times two 36.28 12.70 4.98 36.42 12 2O .85 Total (V + 2Cl 62.89 25.93 5.84 63.00 24.90 .95 1 Reading Rate 1 240.26 94.65 - .- — l - i (n =1942i GRADE 13“ .n . 1942': i Vocabulary 7 34.63 15 92 3.10 34.63 15.92 T: 96 Comprehension times two '; 40.74 11.46 4.77 40.74 11 46 l 83 l Total (V + 2C) 75.38 25 09 5.70 75 38 25 09 05 Reading Rota ' 275 C2 - 5 39 —_ - _ | _-__ _ - _- .. _ __ ;n .. .33: GRADE 14" - Vocabulary j 41 TO 7 7 4* 3 13 [.41 I J: I .' Compreher.;-on : 8‘65 1 o 44 14 1' 3O 4 00 44 '4 11 3C; ' .5 Total (V + 2C) 85.84 26.23 5.78 85.84 26 23 .95 Reading Rate 280.61 105.93 - — — - in = 525) GRADE 15” (n = 525) Vocabulary 45.91 17.10 2.91 45.91 17.10 97 Comprehension times two 46.76 11.00 4.63 46.76 11.00 .82 i Total (V + 2C) 92.67 25.42 5.52 92.67 25.42 .95 = Reading Rate 283.21 95.86 — - - - I (n = 397) GRADE 16" (n - 397) Vocabulary 50.92 17.91 2.72 50.92 17.91 98 ‘ Comprehension times two 47.00 10.48 4.73 47.00 10.48 .80 = Total (V + 2C) 97.91 25.58 5.32 97.91 25.58 .96 ! Reading Rate 288.79 107.72 - .. - - ’1 ‘Corrected by the Snowmen-Brown Formule "In Grades 13.16 the Statisrics are the same for both aid. of the table because there was no weightmg the 1.9...“ a. repeated for the convemence ot the user. BIBLIOGRAPHY 103 BIBLIOGRAPHY Ames. Louise 8.. and Walker. Richard N. "Prediction of Later Reading Abi1ity From Kindergarten Rorschach and IQ Scores." Journal of Educational 1351;001ch 55 (1964): 309-13- Anderson. A. W. "Personality Traits in Reading Ability of Western Australian University Freshmen." Journal 21.Edu§atignal.Be§§aL§h 54 (February 1961): 234-37. Bond. Guy L.. and Fay. Leo. "A Comparison of the Performance of Good and Poor Readers on the Individual Items of the Stanford-Binet Scale. Forms L and M." .1gunna1 of Educational Psychology 46 (December 1955): 489-93. Brunkan. R. J.. and Shen. F. "Personality Characteristics of Ineffec- tive. Effective. and Efficient Readers." .Eensgnnel Guidance Journal (1966): 837-42. Burks. Harold F.. and Paul. Bruce. "The Characteristics of Poor and Good Readers as Disclosed by the Wechsler Intelligence Scale for Children." Joannal .01 Educational Embelm 46 (December 1955): 489-93. Burnside. Robert W.; McHolland. James: and Menary. Earl. "Developmen- tal Programming for Improving Skills and Self-Concepts." Unpub- lished paper submitted to the American College Personnel Associa- tion. 1978. Callaway. Byron; Jerrolds. Bob W.; and Tisdale. Louise. "Symposium of Scores on the California Test of Personality With Reading Achieve- ment." .Enucatign 93 (November-December 1972): 160-64. Cattell. R. 8.. et a1. flannmhictbelfiflanaonaliixfiacmnst. Champaign. 111.: Institute for Persona1ity and Ability Testing. 1970. _. 561303111; Analysis of Eenssznaljix. Champaign. 111.: Institute for Personality and Ability Testing. 1970. . and Delhees. Karl. Manual in: the Clinical Analxajs mas; .tipnnaine. Champaign. 111.: Institute for Personality and Ability Testing. 1970. 104 105 Chronister. Glenn M. "Personality and Reading Achievement)‘ Elmantam School Jaumal (February 1964): 254-61. Cohen. Jacob. "A Factor-Ana1ytica11y Based Rationale for the Wechsler Adult Intelligence Scale." .19u:na1.o£ Consulting Psychology 21 (1957): 451-57. Ekwall. Eldon E. DiagnosisandflsmeulatienuitbeDJsabJadBaauer. Boston: Allyn and Bacon. 1976. . "The Use of WISC Subtest Profiles in the Diagnosis of Read- ing Difficulties." Ph.D. dissertation. University of Arizona. 1966. Evans. James D. "The Relationships of Three Personality Scales to Grade Point Average and Verbal Ability in College Freshmen." Ina Juumal of Educational Baseancb 63 (November 1969): 121-25. Graham. E. E.. and Kamans. D. "Reading Failure as a Factor in the WAIS Subtest Patterns of Youthful Offenders." mm 91 mm W 14 (1958): 302-305. Hage. Dean 3.. and Stroud. James. "Reading Proficiency and Inte11i- gence Scores. Verbal and Nonverbal." Jounnal.oi Educational .Beseangh 52 (March 1959). Hake. James M. "Covert Motivations of Good and Poor Readers." .Ihe .Baading.Ieachen (May 1969): 731-38. Harvood. Thomas B. "Some Intellectual Correlates of Schizoid Indi- cators: WAIS and MMPI." .lounnal,gf.Consulting.Esyghglpgy l (1967): 218. Henderson. Butler and Goffney. "Effectiveness of the WISC and Bende Gestalt Test in Predicting Arithmetic and Reading Achievement for White and Non-White Children." .luurnal.91.Qlln1£al.E§x: .ghology 25 (July 1969). Houston. Camille. and Otto. Wayne. "Poor Readers' Functioning on the WISC. Slossen Intelligence Test and Quick Test." qunnal.g£ ‘Edugatignal.fle§gangh 62 (1968): 12-68. 157-59. Jackson. Douglas hL. and Bloomberg. Robert. "Anxiety: Unitas or Multiplex?" .19uzna1.gf.§pnsu111ng Psychology 22 (1958): 225-27. Joseph. Michael P.. and McDonald. Arthur S. "Psychological Needs and Reading Achievement." An exploratory study. Milwaukee: Marquette University. 1964. 106 Ladd. Clayton L. "WAIS Performance of Brain Damaged and Neurotic Patients." .LQuLnal of Clinical Esucnojccx 20 (1964): 114-17. Lockhart. W. E.. and Menary. E. L. "Personality Patterns of Good and Poor Readers as Measured by the IPAT 16 PF Personality Test." Yearbook of the North Central Reading Association. 1979. McDonald. Arthur. "Intellectual Characteristics of Disabled Readers at the High School and College Levels" Journal of Dexelgpmental Bead; ing 7 (Winter 1964): 97-101. Menary. E. L. "Personality Problems of Readers Using the Clinical Analysis Questionnaire." Unpublished paper presented to the North Central Readers' Association. 1979. Neville. Donald. "A Comparison of WISC Patterns of Retarded and Non-retarded Readers." .lcunna1.c£.Educaticn 1 Research 54 (January 1961). ________t "The Relationship Between Reading Ski1ls and Intelligence Test Scores." Ihe Beading Ieacnen 18 (1965): 257-62. ; Profst. Philip; and Dobbs. Virginia. "The Relationship Between Test Anxiety and Silent Reading Gain." .AEBJ 4 (January 1967): 45-49. Newland. T. Ernest. and Smith. Patricia A. "Statistically Signifi- cant Differences Between Subtest Scaled Scores on the WISC and the WAIS." l9urnal.9£.§ChQCl.EfixChClcgu 5 (Winter 1967). Payne. David A.. and Lehmann. Irving J. "A Brief WAIS Item Analysis." lounnal cf Clinical Esucnclccu 22 (1966): 296-97. Raygor. Alton L.. and Wark. David. "Personality Patterns of Poor Readers Compared With College Freshmen." .lcunnal.ci.Bcading (October 1964): 40-45. Sarason. Irwin G.. and Minard. James. "Test Anxiety. Experimental Instructions. and the Wechsler Adult Intelligence Scale." .Lcumal cf. Educational £511:thch 53 (1962): 299-302. Schill. Thomas. "The Effects of MMPI Social Introversion on WAIS PA Performance." .lgunnal 91 Clinical Esychglggy 22 (1966): 72-74 e ; Kahn. Malcolm; and Muehleman. Thomas. "Verbal Condition- ability and Wechsler Picture Arrangement Scores." .Jgunnal of Ccnsulting and Clinical Esuchclch 32 (1968): 718-21. 107 . "WAIS PA Performance and Participation in Extracurricular Activities." .Leunnal cf. Clinical Esucheledx 24 (1968): 95-96. Shaw. Dale J. "Factor Analysis of the Collegiate WAIS." .Jgunnal c1 Consulting Esuchelegu 31 (1967): 217- . "Sexual Bias in the WAIS." .lcunnal.91.QQnsuliln9 Efixghglggy 29 (1965): 590-91. Silberberg. Norman. and Feldt. Leonard. "Intellectual and Percep- tual Correlates of Reading Disabilities." .lpunnal 91 School My 6 (Summer 1968). Spache. George D. Inuestidatind the Issues at Reading Disabili: .ties. Boston: Allyn and Bacon. 1976. ________, "Personality Characteristics of Retarded Readers as Measured by the Picture-Frustration Study." .Educatignal and Esxchelcaical Measurement (1954): 186-91- . "Personality Patterns of Retarded Readers." .lgunna1.c£ ‘Edugntipna1,3eseangn (February 1957): 461-69. Spielberger. Charles D. "On the Relationship Between Manifest Anxiety and Intelligence." leunnal cf Consulting Esucheledx 22 (1958): 220-24. Thurston. Eric L. "The Philosophical and Sociological Bases of Reading." Ecurteenth ieanbeek cf. the Natlenal Beading Center; .ence. Edited by Eric Thurston and Lawrence Hafner. Milwaukee. Wis.: National Reading Conference. n.d. Wattenberg. W. W.. and Clare. Clifford. Final Report of Cooperative Research Project No. 377. Contract No. SAE 7789. 1962. Wechsler. David. EAIS Manual: Mechslec Adult Intelligence Scale. New York: Psychological Corporation. 1955. Wilderson. Frank B.. Jr. "An Exploratory Study of Reading Skill Defi- ciencies and Psychiatric Symptoms in Emotionally Disturbed Chil- dren." Reading Research Cuantenlu (Spring 1967): 47-73. Williams. Robert L. "Personality. Ability and Achievement Correlates of Scholastic Attitudes." Ihe lcunnal cf. Educational Research (May-June 1970): 401-403. 108 Zimmerman. Irla Lee. and Allebrand. George N. "Personality Character- istics and Attitudes Toward Achievement of Good and Bad Readers." iounnal of Educational Beseacch 59 (September 1965)- Zimmerman. Irla Lee. and Woo-Sam. James M. Cdinical.lntenonetatien of. the liechslen Adult Intelligence Scale. New York: Grune and Stratton. 1973. lillllulll lllll i5 1 U 1293 O3‘mm 3403 111111111 3