‘J..|.-.V 2 O I 'l 2 .‘ ‘4' l 11": . .3!) .II‘J‘ , u... . n. .11; ' ' A a 8.3“ _:. ,, 1 1:53:11 ’0 o a. div. .21., g . . ind 1.5 :u1 4.”! . o auntA . Rm. . . 4n." o u . ’ .31 inn? .4! t ‘4 . l Ty’v’ .' .Vfi. LWA «Emmi. lg'f‘ p.» . . .... I: “aweflyrwhfiofldfrd mus ‘0 .(I h...» DUI}?! This is to certify that the thesis entitled AN EXPLORATORY STUDY OF THE MODIFICATION OF STUDENT ANXIETY LEVELS UTILIZING COGNITIVE STYLE MATCHING presented by William R. Terrell has been accepted towards fulfillment of the requirements for Ph . D . degree in Education W Major professor Date August 9, 1974 ’ 0-7639 "\ 32-0 . 2.? r ' fl — . I ,~ . A. .’ 7 ’ " A-Lvn‘tfitl: f“ ‘1 -\ " \. "7“ i ‘ ‘ 'u—"r) I: f ‘ 1"" ‘éfi'x‘lr. ABSTRACT AN EXPLORATORY STUDY OF THE MODIFICATION OF STUDENT ANXIETY LEVELS UTILIZING COGNITIVE STYLE MATCHING BY William R. Terrell Purpose of the Study The purpose of this study was to explore whether the matching of student cognitive styles to the cognitive style of the instructional mode would reduce student anxiety levels. A review of the literature pointed out that, whereas, mild anxiety might serve as a stimulus to learn- ing, high anxiety is debilitating. This study is founded on the idea that by matching the cognitive style of the student to the cognitive style of the instructional mode it is possible to place students in instructional situa- tions with which they are able to cope and thus reduce their anxiety level. Procedures The subjects of this study were fifty-one fresh- men students enrolled in an audio-tutorial Life Science course at Oakland Community College. Cognitive style maps U /" William R. Terrell h ;\’ {} toc’for each of the students were examined to determine whether they matched the cognitive style of the instructional mode. A-Trait anxiety data was collected during the class orien- tation period. A-State anxiety data was collected at the end of each audio-tutorial session. Three null hypotheses were tested to determine whether the matching of student cognitive styles to the instructional mode would reduce anxiety levels. Two ad- ditional hypotheses were tested to support the predictive validity of the instructional mode cognitive style map and to examine whether there was a difference in the achieve- ment levels between students with high A—Trait and those with low A—Trait. Each of the null hypotheses was tested using the Kolmogorov-Smirnov two sample analysis of vari- ance. The significance of the findings were tested against the two-tailed chi-square with two degrees of freedom at the .05 alpha level. Conclusions 1. High A-Trait students with cognitive styles matched to the cognitive style of the instructional mode tend to have greater reduction of A-State anxiety levels than non-matched students. 2. Students with cognitive styles matched to the cognitive style of the instructional mode tend to achieve higher grades than students that are non-matched. William R. Terrell 3. Low A-Trait students with cognitive styles non- matched to the cognitive style of the instructional mode tend to have greater reduction of A—State anxiety levels than matched students. 4. It is essential to control A-Trait as one of the variables in the study of the reduction of A-State anxiety levels. 5. It is essential to control for cognitive proc- esses, through cognitive style matching or some other equally effective method, in the study of the effects of anxiety on achievement. AN EXPLORATORY STUDY OF THE MODIFICATION OF STUDENT ANXIETY LEVELS UTILIZING COGNITIVE STYLE MATCHING BY William Rijerrell A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1974 Llfllsrll‘. (\ \l Ixf v‘ \I. ACKNOWLEDGMENTS Many individuals contributed to the development of this study. This writer wishes to express his appreciation to the following persons: To Dr. Kent Gustafson for his professional guidance as chairman and for his support and understanding; To Drs. James Nord, Normal Bell, and Stephen Yelon who as members of the advisory committee offered questions and support; To Dr. Joseph Hill, President, and Mr. William McNaughton, faculty member, of Oakland Community College for their professional guidance and practical support in data collection; To Mrs. Joyce Minnett for her service as my long distance trouble shooter and friend; To my wife, Ann, and son, Jean-Paul, for their assistance with the study and their patience. W.R.T. ii ll TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . Chapter I. INTRODUCTION . . . . . . . . . . Purpose of the Study . . . . . . II. Importance of the Study . . . . . . Definition of Key Terms . . . . . . Background of the Study . . . . . . Individualized Instruction . . . . Cognitive Styles of Individuals . . Modification of Anxiety . . . . . Hypotheses . . . . . . . . . . Assumption . . . . . . . . . . Summary . . . . . . . . . . . REVIEW OF THE LITERATURE . . . . . . Introduction . . . . . . . . . Anxiety . . . . . . . . . State-Trait Anxiety . . . . . . Cognitive Style . . . . . . . . Aptitude Treatment Interaction . . . Cognitive Style--Background . . . . Cognitive Style-~Studies . . . Cognitive Style--Relationship to Anxiety . . Studies Related to the Attempt to Modify Anxiety Levels . . . . . . . . Cognitive Style Matching Process . . . Cognitive Mapping . . . . . . . Symbols and Their Meanings . . . Cultural Determinants of the Meaning of Symbols . . . . . . . . . Modalities of Inference . . . . . Summary . . . . . . . . . . . iii Page vi 17 17 17 28 31 32 36 38 41 43 44 46 48 54 55 57 Chapter Page III. DESIGN OF THE STUDY . . . . . . . . 59 Sample . . . . . . . . 59 Limitations of the Study . . . . . 61 Instruments . . . 63 Oakland Community College Cognitive Style Diagnostic Test Battery . . 63 State-Trait Anxiety Inventory . . . 64 Research and Design Procedures . . . 65 Statistical Hypotheses . . . . . . 66 Analysis of the Data . . . . . . . 69 Summary . . . . . . . . . . . 69 IV. ANALYSIS OF THE DATA . . . . . . . 71 Findings of the Study . . . . . . 72 Secondary Findings . . . . . . . 82 Discussion of the Findings . . . . . 87 V. SUMMARY, CONCLUSIONS AND IMPLICATIONS FOR FUTURE RESEARCH . . . . . . . . . 92 Summary . . . . . . . . . . . 92 Discussion . . . . . . . . . . 94 Conclusions . . . . . 97 Implications for Future Research . . . 98 APPENDICES . . . . . . . . . . . . . 100 BIBLIOGRAPHY . . . . . . . . . . . . . lll iv Table 1. 11. 12. 13. 14. 15. 16. LIST OF TABLES Report of the Data . . . . . . . . Kolmogorov-Smirnov Analysis of Hypothesis Kolmogorov-Smirnov Analysis of Hypothesis Kolmogorov-Smirnov Analysis of Hypothesis Kolmogorov-Smirnov Analysis of Hypothesis Kolmogorov-Smirnov Analysis of Hypothesis Kolmogorov-Smirnov Analysis of Hypothesis Summary of Research Hypothesis Testing . Summary of Secondary Hypothesis Testing . Test-Retest Reliability--STAI . . . . STAI Means, Standard Deviations, and Alpha Reliabilities . . . . . . . . . . Item Remainder Correlation Coefficients --STAI O O O O O O O O O O O O Correlations--STAI . . . . . . . . Means, Standard Deviations, and Alpha Re- liabilities--STAI . . . . . . . . Significance of Incomplete Data—~Matched . Significance of Incomplete Data—-Non- Matched . . . . . . . . . . . . [.1 Page 75 78 80 82 84 84 86 88 88 102 103 104 106 106 109 110 Figure 1. 2. 3. 4. LIST OF FIGURES Anxiety Curve . . . . . . Aptitude Treatment Interaction . Sample Cognitive Style Map Research and Design Procedures vi Page 20 33 57 67 CHAPTER I INTRODUCTION Purpose of the Study Primary concerns of the educator include the safety and well-being of students as well as ensuring their suc- cess in achieving educational objectives. While many studies in achievement motivation indicate mild anxiety has a motivating effect, high anxiety generally has a debilitating effect on learning. High anxiety may also be maladaptive to psychological development. A general goal of the educator, then, should be to attempt to reduce the anxiety levels of students to levels which are produc- tive rather than destructive. Those students with high anxiety levels are fre- quently confronted with situations they interpret as stress- ful. They may react to stress situations with either de- fense mechanisms or coping behavior. Students who have encountered stressful situations unsuccessfully with de- fense mechanisms are conditioned to avoid those situations. Those students who have encountered stressful situations successfully with coping behavior are conditioned to re- act to those situations with reduced anxiety levels. This conditioning is called habituation. 1 The purpose of this study is to explore the possi- bility of employing cognitive style to match students to learning situations in which habituation has occurred. The hypothesis is that the matching of student cognitive styles to the cognitive style of the instructional mode will result in the placement of students in instructional situations or settings with which they have learned to cope and, thus, reduce anxiety levels. It is hypothesized that the degree of reduction of A—State anxiety levels among students with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of students with cognitive styles that are non-matched. Importance of the Study Educators have long recognized that there are in- dividual differences among students. Examples of individual differences include: ability level, attitude, self-concept, past experiences, cognitive style, interests, and anxiety level. Student anxiety is an especially important indi- vidual difference because it has potential for being con- trolled by the teacher. Many other individual differ- ences, i.e., ability level and previous experience, may not be controlled. This study is important in that it explores a possible method for teachers to improve the potential for individual students to learn by controlling levels of stu- dent anxiety toward instruction. High levels of student anxiety toward instruction are common and are important factors for consideration in the attempt to individualize instruction to meet individual student needs. This study is also important in that while there are several reports on the debilitating effects of high anxiety levels on student learning, a search of the lit- erature has revealed only two studies which report attempts to modify or control high levels of student anxiety toward instruction. This study, then, may contribute to the knowledge of an area which requires a great deal more investigation. Definition of Key Terms l. Audio-Tutorial Instruction--"Recorded explana- tions, descriptions, directions, and their use by indi- viduals or groups of learners as they seek knowledge or develop skills."1 2. Cognitive Style2--A concept for describing an individual's mode of behavior in searching for meaning. It is identified by an individual's disposition to use lPaul Wittich and Charles Schuller, Instructional Technology: It's Nature and Use (New York: Harper & Rowe, Publishers, 1973), P. 355. 2A more detailed definition and explanation is presented in Chapter II. certain types of symbolic forms versus others; the deri- vation of meaning of symbols from roles the individual has found most satisfying; and the manner in which he reasons. An individual's cognitive style is determined by the way he takes notice of his total surroundings--how he seeks meaning--how he becomes informed. Is he a lise. tener or a reader? Is he influenced in decision-making by his family or by his group of associates? 3. Matched Cognitive Style—-The student's cogni- tive style is mapped and the cognitive style of the in- structional setting is mapped. A student's cognitive style is considered matched with the cognitive style of the instructional setting when there is either a major or minor orientation. A major orientation is 76 to 100 percent congruence, a minor orientation is 26 to 75 percent congruence, and a negligible orientation is 0 to 25 per- cent congruence. 4. Non-Matched Cognitive Style--The student's cognitive style is considered non-matched with the cog- nitive style of the instructional setting when there is a neglibible orientation. 5. State Anxiety (A—State)--State anxiety may be conceptualized as a transitory emotional state or condition of the human organism that varies in intensity and fluctuates over time. This condition is characterized by subjective, consciously perceived feelings of tension and apprehension, and activation of the autonomic nervous system. Level of A—State anxiety will be high in circum- stances that are perceived by an individual to be threat- ening, irrespective of the objective danger; A-State should be relatively low in objectively non-stressful situations, or under circumstances in which an existing danger is not perceived as threatening.3 6. Trait Anxiety (A-Trait)--Trait anxiety refers to relatively stable individual differences in anxiety proneness. Persons who are high in A-Trait tend to re- spond to threatening situations with A-State elevations of greater intensity than those who are low in A-Trait.4 Background of the Study The background of this study is concerned with three basic ideas: (1) individualized instruction, (2) cognitive style matching, and (3) anxiety. Individualized instruction is considered as the effort to meet student needs by allowing for individual differences. Cognitive style matching is employed as a method to allow for indi- vidual differences by matching the student cognitive style to the instructional mode cognitive style. Anxiety is the 3Charles D. Speilberger, The Effects of Anxiety on Computer Assisted Learning (Tallahassee, Florida: Florida State University, 1970), p. 7. 4 Ibid., p. 7. student characteristic that this study seeks to control. Attempts by other researchers to control anxiety levels are reviewed. Also presented is support for the notion that anxiety levels can be modified. Individualized Instruction Efforts to match student cognitive styles in an attempt to reduce student anxiety levels should be con- sidered within the total framework of individual differ- ences and the attempt to meet student needs through indi- vidualized instruction. Individualized instruction is defined as "learning experiences specifically designed for individual students on the basis of diagnostic pro- cedures employed to determine individual interests and needs. . . ."5 Individualizing instruction in an attempt to meet individual needs has taken many forms. Schools have altered the class day with flexible or modular scheduling. Elective and mini-courses have been developed. Team teaching, large group, small group and individual study are frequently employed. Students requiring more time for achievement may be allowed to control the de- livery rate of instruction through self—pacing units and opportunities for repetition. A wide range of instruc- tional equipment and materials are available for adoption in individualized instruction programs. SWittich and Schuller, p. 604. All these techniques may be utilized in the indi- vidualization of instruction. Individualization, however, has generally been limited to allowances for self-pacing and repetition. A situation in which all students are given their individual cassette tape—slide program may not truly individualize instruction, inasmuch as the pack- ages may not meet individual student needs. Allowances should be made for student characteristics requiring such elements as external pacing, external-intrinsic motiva- tion, exposition-inquiry, small group-large group, induc- tive-deductive, formal-informal structure, visual, aural, and/or tactile stimuli. One means of understanding individual differences is that of recognizing that each individual possesses a cognitive style, and that differing cognitive styles, and the different experiences underlying them, contributes to the uniqueness of each person. Accordingly, no single instructional approach, or format, will completely satisfy every student's needs. An individualized instruction course of study or learning experience employing the most systematic planning techniques, most relevant content material, and latest instructional equipment and materials will not be optimally effective and efficient if a single format, i.e., audio-tutorial, large group, small group, individual tutoring programed instruction, etc., is expected to meet all student's cognitive styles. Student characteristics such as level of ability, attitudes, in- terests, anxiety level, manner of deriving meaning, and method of reaching decisions form part of the need struc- ture of students. Any serious attempt to individualize instruction to meet student needs must consider these characteristics. The Personalized Education Program at Oakland Community College employs the educational sciences to "personalize" instruction. At Oakland Community College each student carrying more than two classes, or who is enrolled in a class using the "burst"6 approach is tested to determine his cognitive style. "Prescriptions" that match the student's cognitive style to a considerable degree are employed to conduct the instructional program. The educational sciences, as deveIOped at Oakland Commu- nity College, represent an important attempt to truly individualize instruction. Once cognitive styles of stu- dents are determined any or all of the strategies of in- struction may be employed to develop an individualized program for the student. The form of instruction: in- dividual, small group or large group may be prescribed. The media may be selected utilizing the characteristics k 6Burst--Oakland Community Colle e uses this term to describe the dividing of a regular class into indi- vidualized study groups. of the various media, learning theory, communication theory, and cognitive styles of learners. The necessary schedule may be organized for Optimum use of student and teacher time. This study is particularly concerned with the audio-tutorial approaches to individualized instruction. Audio-tutorial instruction is a specific term employed by Oakland Community College to identify an audio-cued approach which is defined as any ". . . recorded explana- tions, descriptions, directions, and their use by indi- viduals or groups of learners as they seek knowledge or deveIOp skills."7 Audio—Tutorial approaches to indi- vidualizing instruction, because they are so different from traditional instruction patterns, may generate high levels of student anxiety in some students. Thus, one of the major characteristics of students that must be considered when prescribing the audio-tutorial approach is anxiety level. Cognitive Styles of Individuals The individual's cognitive style is his unique manner of seeking meaning. It is his total pattern of receiving and processing information. Cognitive style is made up basically of three sets. Set 1, symbols and 7Wittich and Schuller, p. 355. 10 their meanings, concerns the way a person acquires knowl- edge and meaning through the utilization of symbols. Set 2, cultural determinants, is concerned with the way the individual's meaning of symbols is influenced by cul- tural background. Set 3, modalities of inference, is concerned with the type of inference process the individual employs in arriving at logical conclusions. The cognitive style of an individual is deter- mined by measuring the cognitive characteristics through the administration of a comprehensive cognitive style test battery. The results of the test battery are dis- played in a cognitive style map. The cognitive style map displays the test results for each of the three cognitive style sets in the form of a Cartesian product. cultural {modalities G = {Symbols} X {determinants of inference The cognitive map may then be employed as one of the tools for the diagnosis of the individual's cognitive style. Prescription of an instructional mode for an in- dividual will often include cognitive style matching. In this case the individual's cognitive style map will be compared with the cognitive style maps of the instructional modes available, i.e., programed instruction. The instruc- tional mode matching the cognitive style map of the indi- vidual is considered to offer the student Optimum compara- tibility among the alternatives. 11 Modification of Anxiety A survey of the literature, while disclosing sev- eral studies attesting to the debilitating effects of high anxiety, yielded only two studies attempting to modify anxiety levels. Gustafson's8 attempt to employ simulation to reduce anxiety levels among student teachers produced no significant reduction in anxiety. Sieber and Kameya9 reported success in reducing the undeSirable effects of anxiety in the solution of a puzzle by providing memory support. Though the Sieber-Kameya findings relate to only a single cognitive process, they do provide support for the idea that anxiety levels can be modified. Further support for the feasibility of controlling high anxiety levels is found in observations of studies in instructional experiences which have revealed that certain instructional practices do alleviate the effects of anxiety vis-a-vis student performance. Tobias states, there is a strong rationale for expecting an inter- action between anxiety and program sequence. The high ratio of reinforcement present in programed instruction, the hiearchial and sequential 8Kent L. Gustafson, "Simulation of Anxiety Situa- tions and Its Resultant Effect on Anxiety and Classroom Interaction of Student Teachers," (Ph.D. dissertation, Michigan State University, 1969). 9J. E. Sieber, "A Paradigm for Experimental Modi— fication of the Effects of Test Anxiety on Cognitive Processes," American Educational Research Journal 6, l (1969): 46-62. 12 organization of the materials, and the low level of uncertainty experienced by students working on the instructional materials are such that pro- gramed devices ought to be especially beneficial for highly anxious individuals.lo Phillipsll supports Tobias concerning the use of programed instruction to present material in a non—stressful manner. He adds, however, that programed instruction may be used for self-evaluation prior to teacher evaluation to help build the student's confidence in the mastery of the ma- terial. Sarason cites teacher response to student behavior as an important factor in the control of anxiety levels, In some classrooms failure or lack of progress by a child is responded to by the teacher in a way that increases the child's feeling of inade- quacy. In other classrooms such a child is responded to in a way that, while it recognizes the child's failure or rate of progress, does not make him feel that the teacher is rejecting or derogating him.12 The anxious student will tend to generalize ciriticism as further evidence of his personal inadequacy. Evalua- tion must be conducted in such a manner that it is clear to the student that it is his performance on a given task that is being evaluated-~not his personal adequacy. 10$. Tobias, The Effect of Sequence and Famili- arity with Subject Matter in Achievement from Programed Instruction. Final Report (New York: City University, 1971), p. 7. 11 Phillips, p. 69. 128. Sarason, et. al., Anxiety in Elementary School Children (New York: Wiley, 1960), p. 272. 13 Thus, sequencing of instruction, formative evalua- tion, and student-teacher relationships are suggested as areas for study in the development of methods to control anxiety levels. This study purports to employ cognitive style matching to control anxiety levels. A student whose cognitive style is properly matched to that of the in- structional mode would have the appropriate instructional sequence in a compatible instructor. Thus, if the hypothe- ses advanced in this study are supported, the potential is greater for the control of student anxiety levels. Hypotheses Three hypotheses were advanced to explore whether the matching of student cognitive styles to the instruc- tional mode would result in the reduction of student anxiety levels. The first hypothesis does not discrimin- ate between high and low A-Trait anxiety levels. The second hypothesis employs high A-Trait and the third hypothesis low A-Trait as controls. The secondary hypothe- ses are advanced for additional analysis related to achievement. Cognitive style matching is employed as a variable in the fourth hypothesis and A-Trait anxiety in the fifth. The hypotheses advanced for testing include: Hypotheses l: The degree of reduction of A-State anxiety levels among students with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of students with cognitive styles that are non-matched. l4 Hypothesis 2: The degree of reduction of A- State anxiety levels among high A-Trait stu- dents with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of high A-Trait students with cognitive styles that are non- matched. Hypothesis 3: The degree of reduction of A- State anxiety levels among low A-Trait students with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of low A-Trait students with cognitive styles that are non-matched. Hypothesis 4: The achievement levels of students with cognitive styles that are matched to the cognitive style of the instructional mode will be greater than those of students with cognitive styles that are non-matched. Hypothesis 5: The achievement level of low A— Trait students will be greater than that of high A-Trait students. The hypotheses were analyzed using the Kolmogorov- Smirnov analysis of variance. The significance of the results were tested against the two-tailed chi-square at the .05 alpha level. Assumption The major assumption of this study is that expressed opinions in response to a questionnaire reflect an exist- ing state of an individual. The Oakland Community College Cognitive Style Diagnostic Test Battery and the State-Trait Anxiety Inventory each rely heavily on self reporting techniques. It must be recognized that an individual tak- ing a test or responding to a questionnaire can distort 15 his responses to form a pattern that does not represent the true or existing state of that individual. Though no control mechanism has been used to allow for lying, there is also no probable reason that students would choose to distort their responses. It will be assumed that the self reporting in response to the measures em- ployed in this study represent the true or existing state of the individuals responding. Summary This chapter has related the deSirability of lower- ing anxiety levels of students to the general trend toward individualized instruction. It has also been illustrated that lowering anxiety levels is only part of the overall study of achievement motivation. Within the framework of achievement motivation it must be recognized that it might be desirable to raise the anxiety levels of indi- viduals operating on the boredom level. In demonstrating these relationships it has been shown that, whereas, at- tempts to individualize instruction have taken into con- sideration limited student characteristics, a more com- prehensive diagnosis of student characteristics should result in a more truly individualized approach. The Personalized Education Program at Oakland Community College, employing cognitive styles, has been suggested as a method for comprehensive diagnosis of 16 student characteristics. Anxiety has been singled out as a student characteristic with potential for harmful effects on student performance and mental health. It is thought that anxiety levels of students can be controlled through cognitive style matching. CHAPTER II REVIEW OF THE LITERATURE Introduction This study is founded on the research conducted in the areas of anxiety and cognitive style matching. The review of the literature concerning this research ex- amines the relationship of anxiety to student performance and the relationship of cognitive style matching to anxiety. Also included is a detailed explanation of the cognitive style matching process. The chapter is divided into three major segments, (1) anxiety, (2) cognitive style, and (3) the cognitive style matching process. Anxiety This study is concerned with the exploration of a method to control student anxiety levels as one of the characteristics within the total framework of individual differences. It is necessary, therefore, (1) to examine the relationship of anxiety to student performance, (2) to investigate whether there is any evidence that control- ling anxiety will affect student performance, and (3) to establish a procedure for determining the nature and degree of anxiety levels. The relationship of anxiety 17 18 to student performance will be discussed under the general heading anxiety. This section will include discussion of the relationship of anxiety to achievement motivation and the debilitating affects of anxiety. The investiga- tion of whether there is any evidence that controlling anxiety will affect student performance is discussed under the heading Classroom Treatment of Anxiety. The establish- ment of a procedure for determining the nature and degree of anxiety levels will be discussed under the heading State-Trait Anxiety. The relationship of anxiety to student perform- ance has been confused by what seems to be conflicting reports concerning the motivating effects and the debili- tating effects of anxiety. This relationship may be clarified by viewing anxiety within the framework of 13 affect arousal model of achievement moti- McClelland's vation. McClelland describes motivation as a change in affect. The absence of a change in affect may be de- scribed as a static state or boredom. Mild change in af- fect is motivating, however, excessive change in affect is debilitating. The affect of concern in this study is anxiety, though the other emotions will have the same ef- fect. 13D. C. McClelland, The Achievement Motive (New York: Appleton-Century-Crofts, Inc., 1953). 19 As a more precise description of change in affect the physiological changes in relation to anxiety may be considered. Anxiety inducing stimuli elicit either coping behavior or defense mechanisms and activates a sympathic autonomic nervous reaction. The sympathic autonomic nervous reaction may take the form of unusual motor activity, increased pulse rate and blood pressure, glandular secre- tions, and restricted bowel activity. The subject reacts to a mild form of anxiety by seeking to avoid or overcome the sympathic nervous reaction with coping behavior which may be described as motivation to accomplish a task about which he feels anxious. High anxiety results in an ex- cessive sympathic autonomic reaction which is debilitat- ing because the subject responds to the stress stimulus instead of the instructional stimulus. Very low anxiety elicits a parasympathic autonomic nervous response which is a complete relaxation of the symptoms described above. Low anxious reactions are debilitating to achievement in that there is insufficient stimulation resulting in bore- dom and lack of motivation. The effects of anxiety may be described as a bell shaped curve. One tail of the curve represents low anxiety which is debilitating. The second tail represents high anxiety which is debilitating. The apex of the curve or middle range indicates a state in which there is suffi- cient anxiety to be motivating but not so much as to be 20 debilitating. It is assumed that this state would afford the optimum anxiety conditions for achievement motivation. optimum anxiety level low anxiety- boredom high anxiety- debilitating Figure l.--Anxiety Curve As has been previously stated, our attention is directed in this study to the reduction of high levels of anxiety. Positive findings regarding the use of cogni- tive style matching to reduce high anxiety levels would indicate the need for further study of the potential of using cognitive style matching for the identification of the optimum anxiety level. The motivating effects of anxiety are defined in Drive Theory. HA (high anxious) students tend to learn certain kinds of tasks faster and more easily than LA (low anxious) students. However, the tasks for which anxiety appears to be motivating are simple non-compete- tive tasks. O'Neil discusses Drive Theory in these terms, 21 In Drive Theory, there are three major assump- tions concerning the learning process. First,; it is assumed that both correct and erroneous response tendencies are evoked by a learning task, and that the latter continue to be elicited even as the correct response is learned. Second, Drive Theory posits that both correct responses and competing error tendencies are multiplied by H (anxiety). Third, it is assumed that perform- ance is jointly determined by level of H and the relative strengths of correct and competing re- sponse tendencies.l4 Symonds offers support for the motivating effects of anxiety, adding the thought that ". . . a person learns only when he has an unsatisfied need that can be met 15 17 while only by learning." McKinneyl6 and Phillips supporting the motivating effects of anxiety, strongly warn of the debilitating effects of high anxiety. O'Neil offers clarification of the relationship between Drive Theory and debilitating levels of anxiety when he states, 14Harold F. O'Neil, Effects of State Anxiety and Programming Variables on the Computer Assisted Learning of College Students. Final Report (Tallahassee, Florida: Florida State University, 1970), p. 13. 15P. M. Symonds, What Education Has to Learn from P§ycholo y (New York: Teachers College, Columbia Univers- ity, 1958), p0 64. 16Fred McKinney, "Emotional Factors in Learning," Journal of General Psychology 9:101-16 (1933): 109. l7Beeman N. Phillips, Anxiety and School Related Interventions: A Selective Review Synthesis of the Psy- chological Literature (New York: New York State Educa- tion Department, 1971), p. 25. 22 Drive Theory predicts that the performance of high anxious students will be inferior to that of low anxious students on complex or difficult learning tasks in which competing error tendencies are stronger than correct responses. In contrast, on simple learning tasks, in which correct re- sponses are dominant relative to incorrect re- sponse tendencies, it would be expected that the performance of high anxious students would be superior to that of low anxious students.18 Spence19 indicates that when tasks are complex and highly competetive, high anxiety is debilitating to performance. Knight and Sassenrath20 found that high anxious subjects performed poorer than low anxious subjects on difficult tasks. Spielberger states, ". . . performance of HA subjects was inferior to that of LA subjects for the more difficult elements, and superior for the less difficult elements, Of a complex laboratory task."21 18O'Neil, p. 13. 19J. T. Spence, and K. W. Spence, "The Motivational Components of Manifest Anxiety: Drive and Drive Stimuli," in Anxiety and Behavior, ed: C. D. Spielberger (New York: Academic Press, 1966. 20H. R. Knight and J. M. Sassenrath, "Relationship of Achievement Motivation and Test Anxiety to Performance in Programed Instruction," Journal of Educational Psychol- ggy 14 (1966): 185-191. 21C. D. Spielberger, ed., "The Effect of Anxiety on Complex Learning and Academic Achievement," in Anx1ety and Behavior (New York: Academic Press, 1966): p. 376. 23 Cattell22 warns that inducing anxiety to utilize Drive Theory to improve student performance would not be appropriate. The educator is further cautioned not to induce anxiety in order to improve learning by Symonds who remarks, ". . . real learning situations hold suffi- cient frustration and anxiety--it should not be necessary to create them artificially and extraneously.23 "In general, anxiety in the classroom interferes with learn- ing, and whatever can be done to reduce anxiety should serve as a spur to learning."2 The debilitating effects of high anxiety on stud- ent learning and performance are widely reported in the literature. Gaier offers a clear description of the ef- fects of high anxiety in a study in which he found support for the following two hypotheses: First, if the anxiety is above individual threshold, it leads to an impairment in the ability to impro- vise in an unstructured and/or new situation. This results in stereotyped, habitual, and familiar ap- proaches that may be maladaptive in the situation. 2 . . . 2Raymond B. Cattell, "Anx1ety and Mot1vat1on: Theory and Crucial Experiments," in Anxiety and Behavior, ed: C. D. Spielberger (New York: Academic Press, 1966), p. 45. 23Symonds, p. 64. 24Symonds, p. 70. \( \rl‘l|\,|ll\{"l . [ (It! ’I' f )1 ll? .Ill ('5'? I'll. {willy lllll...’ . 'l { (I? 24 Second, if anxiety is above individual threshold, the individual becomes self-engrossed and con- cerned with his personal adequacy.25 Meyersz6 found that high anxiety appears to inhibit the utilization of memory span in the solution of concept problems. Further findings indicating the debilitating effects of anxiety are reported in studies by Palermo,27 Sarason, Hill, and Zimbardo,28 Davidson,29 Chansky,30 and 31 Zwiebelson32 reported that the negative rela- Ripple. tionship between intelligence and anxiety was greater when an intelligence test was administered in a test-like 25R. L. Gaier, "Selected Personality Variables and the Learning Process," Psychological Monoggaphs 66 (1952): 11. 26Joel Meyers and Jack Dunham, "Effects of Anxiety on Aptitude by Treatment Interactions," Paper presented at the American Educational Research Association, New York, February, 1971. 27D. S. Palermo, A. Castenda, and B. R. McCandless, "The Relationship of Anxiety in Children to Performance in a Complex Learning Task," Child Development 27 (1956): 336. 28Phillips, p. 50. 29Phillips, p. 49. 30N. M. Chansky, "Threat, Anxiety and Reading Behavior," Journal of Educational Research 51 (1958): 336. 31R. E. Ripple, et. al., "Learner Characteristics and Instructional Mode: A Search for Disordinal Inter- actions," Journal of Educational Psychology 60 (1969): 117. 32I. Zwiebelson, "Test Anxiety and Intelligence Test Performance," Journal of Consulting Psychology 20 (1956): 479-481. 25 atmosphere with ego-involving instructions rather than in a relaxed, neutral atmosphere. Feldhusen and Klau8e meier33 reported negative correlations between anxiety and achievement and IQ for fifth graders of average in- telligence. Correlations reported were: anxiety corre- lated with WISC IQ .35, Reading -.31, Arithmetic -.20, and Language -.33 (WISC IQ, Reading and Language corre- lations significant at the .05 level). Reasons advanced to explain the debilitating ef- fects of anxiety upon student performance and learning involve the displacement of performance and/or learning activity with anxious activity. The individual learns to respond to anxiety inducing stimuli in ways that have been successful in the past. The response may, in the case of mild or familiar anxiety inducing stimuli, take the form of extra effort to overcome the anxiety. High anxiety or unfamiliar anxiety inducing stimuli may re- sult in avoidance responses to escape the anxiety. When learning or performance activities are introduced with anxiety inducing stimuli the individual responds in a manner appropriate to the anxiety. Mild anxiety might result in the stimulation of the individual to master the 33J. F. Feldhusen and H. J. Klausmeier, "Anxiety, Intelligence, and Achievement in Children of Low, Aver- age, and High Intelligence," Child Development 33 (1962): 407. 26 anxiety and the associated learning or performance activity. High anxiety, however, would result in an avoidance re- sponse to the anxiety and the associated learning or per- formance activity. In either case the learning or per- formance activity is displaced by the anxiety inducing stimulus as the primary response inducing stimulus. Biggs states, . . . when decisions are made under emotional or stressful conditions, immediate memory span capac- ity is effectively lowered. A convenient way of putting this is that the stress signals themselves occupy valuable immediate memory span space.3 There are other considerations regarding the ef- fects of anxiety, perhaps with even more far-reaching consequences than the debilitation of student performance. Alvin Toffler35 warns that technological societies over- stimulate individuals with overchoice and too many de- cisons. He indicates that the communal societies of the hippie movement and the facination with the fabulous fifties are symptoms of the attempt to escape the anxieties of the technological rat race. The educator is faced with the issue of whether, even in the event of acceptable student performance toward achieving the instructional objectives of a unit, it would be acceptable to generate 34J. B. Biggs, Information and Human Learnigg (Glenview, Illinois: Scott, Foresman and Co., 1968), p. 22. 35Alvin Toffler, Future Shock (New York: Random House, 1970). 27 high levels of anxiety if there is a possiblity of harm to the student's general mental health. Sarason36 found high anxious students more self- deprecatory, self-occupied, and generally less content with themselves than low anxious students. Spielberger,37 in a study of high anxious and low anxious students at Duke University revealed that twenty percent of the high anxious students were classified academic failures as compared to six percent of the low anxious students. 38 Zuckerman and Buss report that perceptually defensive participants in a hostile-neutral word recognition task scored higher on the Anxiety Index. Biggs relates the mental health problem to classroom conditions in his statement, The extra stimulation that (usual) classroom re- wards provide--merit marks, pleasing teacher, competition, straight-out punishment for "laziness" --could easily, in the highly neurotic and/or introverted children, cause serious psychological disturbance.39 361. G. Sarason, "Empirical Findings and Theoreti- cal Problems in the Use of Anxiety Scales," Psychologi- cal Bulletin 57 (1960): 401. 37 Spielberger, Anxiety and Behavior, p. 380. 38M. Zuckerman and A. Buss, "Perceptual Defense and Prerecognition Responsivity in Relation to Hostility, Anxiety and Impulsiveness," Journaltmelinical Psychology 16 (1960): 49. 39Biggs, p. 52. 28 Concern about anxiety is not a strictly modern phenomenon. Ali Iban Hazm of Cordova, an 11th century Arab philosopher wrote, I have constantly tried to single out one end in human actions which all men unanimously hold as good, and which they all seek. I have found only this: the aim of escaping anxiety. Not only have I discovered that all humanity considers this end good and desirable but also that no one is moved to act, or resolves to speak a single word, who does not hope by means to this action or word to release anxiety from his spirit.4 State-Trait Anxiety Some individuals will tend to react to more situa- tions with high anxiety while others rarely react to situations with high anxiety. Individuals who tend to react with high anxiety may not react with high anxiety to a given situation, while a person who rarely reacts with high anxiety may react with high anxiety to that same situation. The study of student anxiety levels toward instruction requires measurement of anxiety level in the given instructional situation. The anxiety level present in the instructional situation must be measured rather than the general tendency to react to situationsv with a level of anxiety. Employing instructional materials, methods, and techniques in an effort to control student anxiety makes 40J. Kritzeck, "PhilOSOphers of Anxiety," Common- weal 63 (1955/56): 573. 29 requisite the knowledge of which students under what con- dition will respond to a given instructional mode with an optimum anxiety level. This study makes a distinction between state and trait anxiety. State anxiety is a temporary reaction to a perceived threat which produces feelings of tension and activates the autonomic nervous system. Trait anxiety is a relatively stable personality characteristic which indicates a proneness to react to given situations with a certain level of state anxiety. Some high A-Trait individuals, because they ex- perience certain high anxiety situations frequently, learn to c0pe with those situations. When habituation occurs the individual does not react with high A-State in those situations. Spielberger states, Although STAI (State-Trait Anxiety Inventory) A-Trait scores were relatively good predictors of level of A-State intensity, individual differences in A-Trait were unrelated to performance. For stud- ents whose A—Trait and A-State were consistent, how- ever, high A-State was associated with poorer per- formance on the more difficult part of the CAI learning task, and with better performance on the easier CAI learning materials. For students whose A-Trait and A—State scores were inconsistent, high levels of A-State intensity were especially detri- mental to the performance of low A-Trait/high A- State students. These students have infrequently experienced anxiety states in the past and we may speculate that they have not learned to cope ef- fectively with such states when they occur. In contrast, the high A-Trait/high A-State students made more errors on the first section of the diffi- cult task, but the performance of these students was ultimately superior to that of low A—trait/high A- State students. This finding suggests that high A-Trait students have had more experience than low A-Trait students in adjusting to anxiety states in 30 the past and are therefore better able to cope with such states when they occur.4 Thus, when anxiety level is to be measured for a given situation care must be exercised that A-State anxiety is being measured since a high A-Trait subject may not react with high A-State if this is a situation he has learned to cope with. The debilitating effect of high anxiety on in- struction has been substantiated by research. Exploratory studies are needed to examine possible methods of control- ling levels of student anxiety toward instruction. O'Neil reports that his studies, in general, support the fact that high anxiety disrupts performance. A problem of concern to educators is to discover appropriate means to reduce the disruptive effects of anxiety on 'learning. Although it has been recognized that anxiety can interfere with the learning process, relatively little research has been concerned 2 with reducing anxiety in the learning situation. Cognitive style matching may indicate learning situations in which the high A-Trait student has learned to c0pe. Thus, the high A-Trait student, properly mapped and matched, would have the optimum opportunity for success in an instructional situation with which he has learned 41C. D. Spielberger, The Effects of Anxiety on Computer Assisted Learning (Tallahassee, Florida: Florida State University, 1970), p. 11. 420 'Neil, p. 69. 31 to cope. This study, then, is an attempt to explore one method for reducing anxiety in the learning situation. Cognitive Style Fundamental to the concepts of cognitive style and the proposed relationship between cognitive style and anxiety level is the notion thataniindividual with a set of aptitudes or characteristics will reSpond to a given treatment in a different manner than individuals with different sets of aptitudes or characteristics. Support for this notion is found in a group of studies called Aptitude Treatment Interactions and in the work of isolated individuals concerned with the interaction of student characteristics and instructional treatment. This chap- ter provides a review of the Aptitude Treatment Inter- action literature and the individual works of researchers which this researcher feels are supportive of the con- cept of cognitive style. Certain differences between the Aptitude Treatment Interaction research studies and the cognitive style program will be noted. Studies specifi- cally supportive of the Oakland Community College ap- proach to cognitive style and studies relating cognitive style and anxiety will be cited. Other studies attempt- ing to modify or reduce anxiety levels will be discussed. A fuller, more detailed explanation of cognitive style 32 mapping and the meaning Of the map symbols will be in- cluded in the final portion of this chapter. Hptitude Treatment Interaction The Oakland Community College approach to cogni- tive style is often compared or considered as part of the Aptitude Treatment Interaction (ATI) work. While there are similarities, there are also important differ- ences. Whereas ATI is a body of research conducted under clinical conditions for the purpose Of studying inter- actions Of very specific aptitudes with very specific treatment, the Oakland Community College cognitive style approach is part Of an ongoing educational program at Oakland Community College, Operating with real students, under actual school conditions. Though employing ATI research in the development of the cognitive style ap- proach, Oakland Community College, in contrast, considers a number of characteristics of the individual within their educational setting at a given time, rather than isolate a certain aptitude under clinical conditions. 43 Snow comments that much work in media research has sought a single best method to teach everyone. ATI 43Richard E. Snow, "Research on Media and Aptitudes," eds. G. Soloman and R. E. Snow, "Commentaries on Research in Instructional Media: An Examination of Conceptual Schemes," Viewpoints (Bulletin Of the School of Education, Indiana University, 1970), 46 pp. 63-89. 33 seeks the differentiated instructional methods to teach individuals best. Whereas previous media research sought findings predictive of a wide variety of situations, ATI seeks disordinal findings predictive of narrow, specific situations. Snow describes the aim Of ATI research in this statement, Assuming that a certain set of outcomes from an educational program is desired and considering any particular instructional treatment, in what manner do the characteristics of learners affect the extent to which they attain the outcomes from each of the treatments that might be considered? Or, consider- ing a particular learner, which treatment is best for him?44 ' A sample of the disordinal findings Snow discusses is portrayed in the graph in Table 2. High Anxiety Low Simple Complex Performance Level Figure 2.--Aptitude Treatment Interaction45 The disordinal interaction described in the table indi- cates that persons with high anxiety perform superior to persons with low anxiety on simple tasks, while their 44Ibid., p. 65. 450'Nei1, p. 13. 34 performance is inferior to that Of persons with low anxiety on complex tasks. Such an ATI permits prediction of per- formance by individuals on given tasks and prescription of treatment for Optimum performance. Not all ATI research has provided conclusive and supportive findings. However, Tobias in a review of ATI studies suggested that, . . . among a number Of reasons for not finding ATIs two were most probable: (a) the failure to demon- strate that the variable measured by test score was actually Operative in the experimental situation; (b) the failure to vary subject matter since ATIs may appear in one type of content and not another.4 Ripple47 reports failure to find significant interaction in a study to determine the relationship between selected characteristics of students and the relative process they have in learning from programed versus conventionally structured learning tasks. However, he pointed out that random assignment was used and no attempt was made tO match learner characteristics and design of learning ma- terials. Among the ATI studies reporting conclusive and supportive findings are thOSe by”Doty andDoty,48 and 46S. Tobias and T. Abramson, "The Relationship Of Anxiety, Response Mode, and Content Difficulty to Achievement in Programed Instruction," Journal Of Edu- cational Psychology (1971): 357. 47 Ripple, p. 113. 48B. A. Doty and L. A. Doty, "Programed Instruction ~Effectiveness in Relation to Certain Student Characteris- tics," Journal Of Educational Psychology 62, 4, p. 306. 35 Koran.49 Koran50 reportedtflufl:intern teachers scoring high on a Hidden Figures Test asked superior analytic questions after written-modeling treatment than after video-modeling treatment. Interns scoring low on Hidden Figures asked superior analytic questions after video- modeling treatment than after written-modeling treatment. Soloman51 reported that teachers scoring high on general verbal reasoning (GRE-V) asked more questions about a problem after hypothesis generating training than after one attendance training. Teachers scoring low on general verbal reasoning asked more questions about a problem after cue attendance training than after hypothesis gen- erating training. ATI research, though still in a very early stage of development, holds promise for significant contribu- tions to the quest for a methodology for the individualiza- tion of instruction. Individual interactions which relate to the cognitive processes Of students may be included in 49M. L. Koran, "Differential Response to Inductive and Deductive Sequences of Programed Instruction," Journal of Educational Psychology 62, 4, p. 306. 50M. L. Koran, "The Effects of Individual Differ- ences on Observational Learning in the Acquisition Of a Teaching Skill" (Ph.D. dissertation, Stanford University, 1969). 51G. Soloman, "Interaction of Communication-Medium and Two Procedures of Training for Subjective Response Uncertainty Of Teachers" (Ph.D. dissertation, Stanford University, 1968). 36 the cognitive style battery, thus enhancing the role Of cognitive style matching in the individualization of in- struction. Cognitive Style-- Background The cognitive style Of the individual is a com- posite of the sum of his characteristics related to learn- ing. Often single characteristics Of the individual are conflicting with other characteristics so that a predic- tion Of performance based on a single characteristic, without consideration Of the mitigating effects Of the conflicting characteristics, would result in an erroneous prediction. Cognitive style is an attempt to make pre- dictions of performance based upon all the characteristics in concert and within their framework of Operation. Gardner states, Since a number Of cognitive controls are assumed to coexist within a personality, the possibility arises that the composition Of controls within a person may itself have consequences for behavior that can not be deduced from the individual controls alone. We have referred to such arrange- ments of controls as cognitive styles. The impor- tance of style is that it may provide a basis for predicting behaviors that can not be inferred 52 from the characteristics of individual controls. The modern educator is quite familiar with the concept that learner characteristics affect student 52R. W. Gardner, "Cognitive Styles in Categoriz- ing Behavior," Journal of Personality 22 (1935): 214. 37 performance. Age, intelligence, and previous knowledge of the subject matter are usually considered in the plan- ning of instruction. Cognitive style increases the number Of characteristics studied, considers their relationships, and Offers a methodology for their utilization in in- structional design. The expansion of the consideration of student characteristics to a cognitive style approach receives wide support among educational researchers and writers. Hovland53 refers to audience prejudice and pre- disposition to accept an Opinion as differences in learn- ers. Campeau54 states that decisions concerning choosing and using media cannot be justified without thorough analysis of media, learner, and environment—related characteristics. Biggs in his work in information processing in human learning Offers support for the cognitive style ap- proach as he discusses the shortcomings of intelligence testing. He states, In intelligence or IQ testing, the normal proced- ure is to administer a battery of highly differing 53C. E. Hovland, A. L. Lumsdaine, and F. D. Sheffield, Experiments on Mass Communication, v. 3 (Princeton, N. J.: Princeton University Press, 1949), p. 9. 54Peggie L. Campeau, "Selective Review of Liter- ature on Audiovisual Media of Instruction," in L. J. Biggs, P. L. Campeau, R. M. Gagne, and M. A. May, Instructional Media: A Procedure for the Design Of Multi- Media Instruction, A Critical Review of Research and Suggestions for Future Research (Pittsburgh: American Institutes for Research, December, 1966), p. 138. 38 tests of a cognitive kind and to extract from the scores, by a method of factor analysis, certain basic underlying factors of the intellect. . . . These factors refer to content, not to process.55 Items that affect intellectual performance, but which are affective or emotional in nature, are not only not represented; they are considered to interfere with the purity of the intellectual measures themselves, and are therefore deliberately excluded as being irrelevant. . . . This procedure amounts to throwing out the baby with the bath water, because non-intellective measures are just as relevant as intellectual measures themselves.56 While it may be useful for the teacher to screen his classes according to the IQ of the child, it might be even more useful if he screened according to more basic and analytic process factors, for example by having the stable extroverts in one track, the neurotic introverts in another.57 The processes Biggs discusses are very much a part Of the cognitive style approach of Oakland Community College. The concern, within cognitive style for informa- tion processing, is evidenced by the three sets of the cognitive style: (1) ability and manner of deriving mean- ing from symbols, (2) cultural determinants of the mean- ing of symbols, and (3) modes of inference employed in the process of deriving meanings. Cognitive Style--Studies The Oakland Community College Personalized Edu- cation Program utilizing cognitive style has stimulated 55J. B. Biggs, Information and Human Learning (Glenview, Illinois: Scott, Foresman and Co., 1968), p.88. 56 Ibid., p. 94. 57Ibid., p. 89. 39 a vast array of research projects. At least sixty doc- toral studies have been conducted on various aspects and applications of the Oakland model. Among the reports Offering support for the concept of cognitive style itself is a study of one hundred and eighteen high school English students by Schroeder resulting in findings that indi- cated, students rating the teacher high in effectiveness tended to a statistically significant degree (.05) to be students having cognitive styles "similar" to that of the teacher. Students having "similar" cognitive styles also tended to be the students who received the highest grades. Wyett59 found that instructors who were placed in teaching situations which were not in keeping with their cognitive style did not perform as well as those who were appropriately matched. Hoogasian60 and Fragale61 58A. V. Schroeder, "A Study Of the Relationship Between Student and Teacher Cognitive Styles and Student Derived Teacher Evaluations" (Ph.D. dissertation, Wayne State University, 1968), p. 49. 59J. L. Wyett, "A Pilot Study to Analyze Cogni- tive Style and Teaching Style with References to Selected Strata of the Defined Educational Sciences" (Ph.D. disser- tation, Wayne State University, 1967). 60V. Hoogasian, "An Examination of Cognitive Style Profiles as Indicators of Performance Associated with a Selected Discipline" (Ph.D. dissertation, Wayne State University, 1970). 61Marvin J. Fragale, "A Pilot Study of Cognitive Styles of Selected Faculty Members and Students in a Community College Setting" (Ph.D. dissertation, Wayne State University, 1969). 40 found that the cognitive styles of students successful in a given course Of study could be composited to form a collective cognitive style. The collective cognitive style could be used to predict the success or failure Of future students in that course Of study. Lange62 in a study involving nursing students concluded that students that have cognitive styles and preferred teaching styles that match the instructor's cognitive style and teaching style have lower failure-withdrawal rates than nursing students that are non-matched, and that matched students tended to receive higher grades. Wasser63 reported that elementary teachers tended to give higher grades to stud- ents with cognitive styles that were similar to their own. 64 Shuert was able to identify cognitive style character- istics unique to students that were successful in 62C. M. Lange, "A Study of the Effects on Learn— ing of Matching the Cognitive Styles of Students and In- structors in Nursing Education" (Ph.D. dissertation, Michigan State University, 1972). 63Laurence Wasser, "An Investigation into Cogni- tive Style as a Facet Of Teachers' Systems Of Appraisal" (Ph.D. dissertation, University of Michigan, 1969). 64Keith L. Shuert, "A Study to Determine Whether a Selected Type of Cognitive Style Predisposes One to DO Well in Mathematics" (Ph.D. dissertation, Wayne State University, 1966). 41 mathematics. De Loach65 found that administrators tended to give a higher rating to instructors with cognitive styles similar to their own. Cognitive Style--Rela- tionshipfito Anxiety Anxiety is a single characteristic Of the indi- vidual, whereas, cognitive style attempts to describe the sum Of the cognitive characteristics Of the individual. One inherent difficulty in the study of the effects of a single characteristic is whether that characteristic or another characteristic or set of characteristics actually resulted in the Observed effects. Taylor in a clarifica- tion of studies involving the Taylor Manifest Anxiety Scale points out, No attempt has ever been made to claim that the only difference between individuals receiving different scores on the MAS is in drive level or that all performance differences could be ex- plained by drive. Undoubtedly there are many characteristics other than drive level on which anxious and non-anxious subjects differ.6 Endler provides support for the utility of cognitive style in the study Of anxiety when he states, 65Joseph F. DeLoach, "An Analysis of Cognitive Style Disparity as an Antecedent of Cognitive Disson- ance in Instructional Evaluation: An Exploratory Study in the Educational Sciences" (Ph.D. dissertation, Wayne State University, 1970). 66J. A. Taylor, "Drive Theory and Manifest Anxiety," Psychological Bulletin 53 (1956): 303. 42 There is no single major source of behavioral variance, at least so far as the trait of anxiousness is concerned. Human behavior is complex. In order tO describe it, one must take into account not only the main sources Of variance (subjects, situations, and modes of response) but also the various simple inter- actions (subjects with situations, subjects with modes of response, and situations with modes of response) and, where feasible, the triple interaction (subject with situation with modes Of response). Behavior is a 67 function Of all these factors in combination. COgnitive style would be useful in providing a diagnostic tool in the study of subjects with situations, subjects with modes of response, and situations with modes Of re- sponse. Cognitive style matching would provide a method of treatment for the control of anxiety by matching the subject's style with that Of the situation and/or with the modes of response in sets or patterns that would evoke the Optimum levels of anxiety. Through cognitive style matching, individuals with given anxiety traits could be matched to instructional modes that Offer the Optimum Opportunity for success. Sutter and Reid in a study of undergraduate males engaged in problem—solving CAI found that for "individuals with high test anxiety a non-interpersonal condition was more conducive to learning than the interpersonal 67N. S. Endler, and J. Hunt, "Sources of Behavioral Variance as Measured by the S-R Inventory Of Anxiousness," Psychological Bulletin 65 (1966): 344. |l '1) ill: .l’llrl 'l ‘lll 43 condition."68 Thus, the instructional mode prescription for individuals with high test anxiety might not include pairing the individual with a partner. A more comprehen- sive prescription Of the instructional mode would be possible if the complete range Of cognitive characteris- tics were considered as in cognitive style matching. Studies Related to the Attempt to Modify Anxiety Levels The literature regarding anxiety is rich in stud- ies describing the effects of anxiety on learning. How- ever, there is a definite void regarding studies of methods to control anxiety related to learning. Sieber69 comments On this scarcity of research and Offers a para- digm for the modification Of the effects Of test anxiety on cognitive processes. Sieber proposes the study of individual cognitive processes, i.e., memory, evaluative set, and attention, in relation to anxiety. However, this approach neglects the notion that cognitive processes are relative to one another and together form a cognitive style. Sieber and Kameya employed the prOposed paradigm in a study of the effects of anxiety on memory. It was reported that in the solution Of a puzzle in which memory 68E. G. Sutter, and J. B. Reid, "Learner Varia- bles and Interpersonal Conditions in Computer Assisted Instruction," Journal of Educational Psychology 60, 3, pp. 153-157. 69 Sieber, pp. 46-62. IIIIIIII‘I'I.‘I1IIDIII ‘ 1 ll. .[ ill I lt'llfl‘ll‘ [I]. I‘ll: ll. 44 Of error is critical, anxiety impaired performance. When memory support was provided the undesired effects of anxiety on task performance were reduced. Gustafson70 in a study involving the effects of simulation on student teachers at Michigan State Univers- ity was unable to support the hypothesis that simulation would lower the anxiety Of student teachers. The anxiety test employed measured general or A-Trait anxiety, Gustaf- son suggests that different findings may have resulted if a specific anxiety test had been employed. Another variable that might have resulted in different findings would be whether the subjects' cognitive styles matched that Of the simulation treatment. Cognitive Style Matching Process Essentially, one's cognitive style is his unique mode of behavior in the search for meaning. An individu- al's cognitive style is determined by the way he takes note Of his total surroundings--how he seeks meaning, how he becomes informed. Is he a listener or a reader? Is he influenced only with his own viewpoint or is he influenced in decision-making by his family or associates? Does he reason in categories as a mathematician does, or in relationships as social scientists do? 70Gustafson, p. 89. 45 Although individual's cognitive styles are con- stantly changing (a person is not stuck with a style for life), they can be employed for diagnostic purposes. This procedure leads to the formulation of strategies for the matching and augmentation Of "styles" to increase the probability of a student's successful performance in the myriad of educational tasks he faces daily. Cognitive style not only expresses the uniqueness Of students as persons engaged in meaning-seeking behavior, but allows teachers to communicate this uniqueness precisely and accurately within the realm of education.71 Within the framework Of educational science as defined by Hill, the construct of cognitive style is a Cartesian product composed of the following four sets: (1) symbols and their meaning, (2) cultural determinants of the meaning of symbols, (3) modalities of inference, and (4) neurological, electrochemical and biochemical aspects of memory functions. It should be noted, how- ever, that since the body Of information pertaining to the fourth set is highly speculative at this stage Of its development, the Cartesian product representing cognitive style at the present time is limited to the first three sets. Thus, the construct Of cognitive style is expressed 71L. Wasser, "Are Marks Larks?" (Bloomfield Hills, Michigan: Oakland Community College Press, 1971), p. 19. 46 as the Cartesian product G, composed of three sets, S, E, and H. S denotes the set of elements defining symbolic orientations, E indicates the set Of cultural determinants, and H designates the set of modalities of inference.72 G: {s} x {E} x {H} The construct of cognitive style is a vehicle which can be used to diagnose individuals and prescribe activities that provide the high probability of the stud- ent's accomplishing successfully the educational task confronting him. It should be noted that the cognitive style of an individual is a relative concept, and depends not only upon the educational level and cultural back- ground of the individual, but upon the symbolic condition Of the task to be accomplished. In this context, the derivation Of an appropriate style for an individual demands that the diagnostician analyze the student as well as the substance of the educational task to be con- sidered. Cognitive Mapping Classifying the elements Of a given set into two categories included in a second set is a process mathe- maticians call mapping. Thus, cognitive style may be 72J. E. Hill, "Cognitive Style as an Educational Science" (Bloomfield Hills, Michigan: Oakland Community College), p. 2. 47 defined as a Cartesian product and expressed as a cogni— tive map. Mathematical mappings employ only the theoreti- cal symbols Of abstract logic or mathematics. Decisions may be rendered by performing indicated Operations on the logical (mathematical) function used to determine the desired outcome. The educational sciences, however, em- ploy empirical mapping involving persons on a "makes sense" or "does not make sense" basis to classify ele- ments (e.g., persons, processes, and properties of a social system called curriculum) into logical categories. Empirically, the diagnostician must decide which elements in the sets S, E, and H, respectively, are apprOpriate for inclusion in the sub-sets that form the individual's cognitive style. The cognitive style of an individual cannot be empirically mapped without considering: (l) the level of educational development of the person, (2) the general symbolic conditions of educational tasks he will be called upon to accomplish, (3) certain ante- cedents (e.g., family) to his present state of develOp- ment, and (4) the appropriateness Of the elements under consideration for the conditions under which the educational task must be completed.73 73Hill, "Cognitive Style as an Educational Science," p. 6. 48 Hymbols and Their Meaning74 Man acquires knowledge and meaning through the utilization Of symbols. The two kinds Of symbols that man uses are the theoretical which is comprised of words and numbers, and the qualitative which employs sense and code data. Inasmuch as it is impossible to educate an individual without using both of these forms Of symbols, the body of information pertaining to symbols and their meanings is basic to the understanding of the educational sciences. Theoretical Symbols.--The theoretical symbol is that symbol which presents to the awareness or nervous system Of the individual something different from that which the symbol itself is. Theoretical symbols include percepts, private interpretations of sensations, and concepts, shared and agreed upon interpretations of sensations which are capable Of being generalized. Thus, words and numbers are the most common form Of theoretical symbols. Theoretical Visual Linguistic Symbols.--These are written or printed expressions, such as the words on this page, which bring to the awareness of the reader 74L. Wasser, "The Educational Science of Cogni- tive Style: An Introduction" (Bloomfield Hills, Michi— gan: Oakland Community College Press, 1971). 49 imageries different from the printed arrangement of the letters and words involved. Theoretical Auditory Linguistic Symbols.--These are sounds of words which bring to the awareness or nervous system Of the listener imageries different from the sounds themselves. The spoken sentence, "they have a beautiful house," brings an imagery to mind that may include a new brick house or a stately ante bellum man- sion, etc. Theoretical Visual Quantitative Symbols.--These are written numbers-~and Theoretical Auditory Quantita- tive Symbols--these are spoken numbers--also include percepts and concepts which differ from their symbols. For example, students having difficulty distinguishing written symbols from their conceptual meanings might infer that 1/2 of 11 is l or that 1/2 of 1,000 is 10. Since the basic purpose of determining an in— dividual's cognitive style is to reflect his strengths at a given level of educational development, the follow- ing principles apply: 1. If the percentile rank Of an individual's score in a given element occurs in the array Of values ranging from the fiftieth through the ninety-ninth per- centile Of a population of these scores, then the individual 50 is assigned a major orientation, expressed as a capital letter. 2. If the percentile rank of an individual's score in a particular element occurs in the array of values ranging from the twenty-fifth through the forty- ninth percentile of a pOpulation of these scores, the individual is accorded a minor orientation expressed as a small letter. 3. If the percentile rank of an individual's score in a particular element occurs in the array of values ranging from the zero percentile through the twenty-fourth percentile of a population of these scores, then the individual is assigned neither a major nor a minor orientation. Theoretical symbols, then, may be expressed as follows: 1. Theoretical visual linguistic (Tvl'tvl) 2. Theoretical auditory linguistic (Tal'tal) ) 3. Theoretical visual quantitative (T ,t vq vq 4. Theoretical auditory quantitative (Taq’taq) Qualitative Symbols.--The qualitative symbol is that symbol which presents and then represents to the awareness or nervous system of the individual that which the symbol itself is to the individual. These symbols, for example, are employed by infants in the process of 51 acquiring and mediating meaning before they have attained any theoretical linguistic symbolic capabilities. Babies, for instance can feel cold (qualitative tactile) before they can say, "I am cold" (theoretical auditory linguis- tic). Qualitative symbols derive their meanings from three sources: (1) sensory stimuli, (2) humanly con- structed formalisms such as games or codes, and (3) pro- grammatic effects of phenomena or Objects which convey an impression of a definite series Of Operations, images, scenes or events. Sensory Stimuli: l. Qualitative auditory(Qa,qa) presents and represents to the individual perceived sounds, e.g., music, hammering, drilling, whose meanings are attained immediately but which cannot be truly shared or gen- eralized, even by persons who hear them at the same point of time, 2. Qualitative visual (Qv,qv), 3. Qualitative tactile (Qt'qt)' 4. Qualitative savory (Qs,qs), and 5. Qualitative Olfactory_(QO,qO) present and represent to an individual the visual entity Of color (redness), the feel of material, taste of food, and scent of 52 roses, respectively. These symbols through which meanings are attained immediately, are also perceptions which can be only approximately but not truly shared or gen- eralized. Symbolic Codes: 1. Qualitative code empathic (Q qcem) the cem' ability to identify with, or have a vicari- ous experience of, another person's feel- ings, ideas or volitions. Qualitative code esthetic (Q goes) the ces’ ability of an individual to view with en- joyment the beauty and pureness Of a re- sulting product, situation or idea. Qualitative code ethic (Q ) a commit- cet'qcet ment to a set Of values, a group of moral principles, Obligations and/or duties. Qualitative code histrionic (Qch’qch) arti- ficial behavior, staged conduct or a de- liberate exhibition Of emotion or tempera- ment to produce some particular effect in, or evoke responses from, other persons. Qualitatlve code kine51cs (Qck’qck) the ability to communicate by means of non- linguistic functions and body motions, such as blushes, shrugs, gestures and gesticula- tions. 53 Qualltat1ve code k1nesthet1cs (Qckh’qckh) motor skill abilities and bodily reactions such as threshold Of awareness Of differ- ent weights, athletic performances, ballet, and finger dexterity. Qualitative code proxemics (Q Cp,qcp) the ability Of an individual to judge and ef- fect critical physical and social distances between himself and others in the act Of communicating, such as culturally determined permissible physical or social distance people maintain between them during con- versation. Qualitative code gynoetics (ch,qcs) per- sonal knowledge of oneself in all qualities and theoretical symbolic forms which con- stitute the totality Of the individual in relation to his environment. Qualitative code transactional (Qct'qcfi) the ability to maintain a positive communi— cative interaction which significantly influences the goals of the persons in- volved in that interaction, such as ex- emplified by group leaders. 54 Cultural Determinants of the Meaning of Symbols Man's perceptions influence and are influenced by his culture. Relatively stable social relationships are, therefore, significant factors in the development of the perceptions which an individual has of his "world." In this context, the formation and transformation of the meaning Of symbols in both the theoretical and the quali- tative domain are influenced by culturally created rules Of expression and communication. If a man derives mean- ing for symbols not as a totally unique person, but as an individual cast in a social role, a role which has ex- pectations imposed upon it by societal norms and groups with which the individual interacts, then those groups which exert the main influence on him determine, in great part, his perceptions of life. It is in this context that the groups Of family and associates, plus his indi- viduality, considered tO be the main cultural influences at work on the individual throughout this life, are called the cultural determinants Of the meaning of symbols in the educational sciences. Since these determinants seem to be almost omnipresent in an individual's theoretical and qualitative symbolic mediations and influence greatly the meaning he ascribes to them, the determinants are fundamental considerations Of any effort to educate. 55 Cultural determinants (E) are expressed as follows: 1. Family (F,f) 2. Associates (A,a) 3. Individuality (I,i) Modalities Of Inference The educational science of modalities of infer- ence is concerned with the diverse methods Of mediating into meaning theoretical and qualitative symbols whose meanings are: influenced by family, associates and/or individuality cultural determinants. Since, generally speaking there are two principle means of symbolic media- tion, deductive and inductive, modalities Of inference consist of both deductive and inductive processes. Man can draw probability conclusions by utilizing the follow- ing four types of inductive inference processes: 1. Magnitude (M,m) the forming and advancing of hypotheses composed of norms, attitudes and categories considered true by the person in the process Of producing probability con- clusions. 2. Differences (D,d) involving linear or mosaic differential comparisons, on a one-to-One basis, Of traits, measurements or character- istics Of the subject under consideration. 3. Relationship (R,r) considers the association Of two or more variables of the entities in question--a comparison by analogy in multi- variate terms. 56 4. Appraisal (L) comprised of magnitude, dif- ference and relationship hypotheses in deriving probability conclusions Of evalua- tion concerning the phenomena under investi- gation. In addition to the four inductive inference pro- cesses, individuals must also employ certain deductive inferential processes when dealing with information in selected aspects of the natural sciences, the life sciences, and in almost all realms Of mathematics and symbolic logic. Deductive ((:)) indicates deductive inference process. The following is a computer print-out Of the cog- nitive style map of a Michigan State University graduate student. This sample may be read by referring to the pre- vious lists explaining the codes. Since the computer does not print small letters, minors are designated with a prime ('). Characteristics are listed in order Of their strength as measured by the cognitive style battery. Those characteristics not listed are considered negligible. ,The cognitive style map as displayed on the follow- ing page may be used for the prescription Of an appropriate instructional mode. The individual's cognitive style map will be compared with the cognitive style maps Of the in- structional modes available, i.e., programed instruction tests, large group lecture, audio-tutorial, small group instruction, etc. The instructional mode matching the 57 cognitive style map of the individual is considered to offer the student Optimum compatibility among the alterna- tives. T'(AL) T(AQ) T(VL) T(VQ) Q(CET) Q(CS) Q(CES) 'Q(CK)) Q(CEM . a I G = o Q(CKH) ox. F 0x0 R (K) Q(V) Q(T) Q(P) Q(O) Q'(CH) Q'(CP) Q'(CT) Figure 3.--Sample Cognitive Style Map. Summary This chapter has attempted to clarify the relation- ship between aptitude treatment interaction and the Oak- land Community College cognitive styles. It has been pointed out that ATI is a research area and that the Oak- land Community College cognitive styles is part Of an on-going educational program that utilizes some aspects of the ATI research. The need for a cognitive style ap- proach has been demonstrated by the statements of edu- cators expressing that need. Support for the Oakland 58 Community College cognitive style approach has been indi- cated by the reports of studies successfully employing cognitive styles. The relationship between cognitive style matching and anxiety was illustrated. Reports of other attempts to modify or reduce anxiety levels are discussed. A statement of the theory of the Educational Sciences is included, demonstrating the relationships Of its various elements, including cognitive style. Finally, the cognitive style map is explained and the symbols de- fined. CHAPTER III DESIGN OF THE STUDY The relationship of cognitive style matching to student anxiety levels was examined in this study. The anxiety levels Of students enrolled in an audio-tutorial course were compared to determine whether there were dif- ferences in the degree of reduction of anxiety levels between those students that were matched tO the instruc- tional mode and those who were not. Sample This study was conducted at the Orchard Ridge Cam- pus Of Oakland Community College. Oakland Community College is a public two-year institution located in Oak- land County, Michigan, in the suburban area of Detroit. The college enrollment is approximately fifteen thousand students. The subjects of this study were freshmen students enrolled in the fall 1973 course Life Science. Life Science is a course designed for non-Science majors to fulfill the basic science requirement. The subject mat- ter Of the course is primarily biology. The course 59 60 instructor, William McNaughton, provided the following description Of the students enrolled in Life Science: 1. average reading level Of 6.5, 2. average level of educational achievement-- twelfth grade, 3. students express dislike for previous science courses in public schools, 4. students avoided science courses in public schools. It should be noted that the third and fourth character- istics described by Mr. McNaughton are subjective de- scriptions. NO data was collected to support these characteristics. The sampling procedure employed is described by Hill and Kerber75 as purposive or judgement selection. The following criteria were used for the selection of subjects for this study: 1. must be enrolled in Life Science, fall semester, 1973, 2. must have taken the Cognitive Style Diagnos- tic Test Battery. The practice at Oakland Community College is to administer the Cognitive Style Diagnostic Test Battery to only those students enrolled in at least two courses. Of the 65 students enrolled in the fall semester Life Science course 75Joseph E. Hill and August Kerber, Models, Methods, and Analytical Procedures in Education Research (Detroit: Wayne State University Press, 1967), p. 43. 61 fourteen were not enrolled in any other courses leaving a sample of 51. The population to which the findings Of this study may be generalized is technically limited by the selective sample technique to students enrolled in a course very similar to Life Science and who have been mapped for cognitive style. However, it seems reasonable to assume some wider generalizability to other college students with characteristics similar to those described above. The instructional mode Of the segment Of the Life Science course dealt with in this study is audio-tutorial. The students receive all basic instruction through modules or packages which include instruction as well as guidance in laboratory experiences. The instructor for the course is available primarily as a resource person for individual guidance and support. The students were enrolled in Life Science in the traditional manner with no attempt to in- dividualize or personalize through grouping. Thus, the course was made up of a heterogeneous group of students. Some students had cognitive styles which matched the in- structional mode and the cognitive styles Of other stud- ents did not match. Limitations Of the Study The findings Of this study are limited in gen- eralizability by the fact that the cognitive style 62 battery employed is designed specifically for use with students at Oakland Community College. It is highly recommended by Dr. Hill, the College's President, that other institutions wishing to employ cognitive style mapping develop a battery to meet their own specific needs. The findings of this study are further limited by the nature of the institution involved. It is sus- pected that anxiety levels and reactions to treatment might be quite different for the urban commuter-junior college setting as compared to the residential, four year college or university. Another limitation to the findings of this study is the type of instruction involved. Generalizability would be limited to the audio-tutorial science laboratory setting. The findings might be quite different in another subject area or another instructional mode. The inability to determine whether changes in anxiety levels were caused by in-class or external events is a serious limitation of this study. The intent of the study is that the matching of student cognitive styles to the instructional mode cognitive styles will affect anxiety levels. However, there is no assurance that re- sults are not caused by external events. Further re- search, with similar results, would be necessary to help alleviate this doubt. 63 Instruments Oakland Community College COgnitive Style Diagnos— tic Test Battery The Oakland battery is divided into two parts. The first part is a selection Of twelve theoretical tests requiring two and one-half hours to complete, the first six of which represent aptitude testing. The second six tests Of the first part are inventories which give both qualitative and theoretical information about a student. The second section of the test is a qualitative non- written test requiring ninety minutes to complete. This section is composed of performance tests of sensory acuity. The characteristics described by particular in- struments are: Theoretical Test Test NO. l: Verbal Reasoning--TVL Test No. 2: Listening--TAL (Recording) Test NO. 3: Numerical Reasoning--TVQ Test NO. 4: Reading-—TVL (Gates) Test No. 5: Numerical Listening--TAQ (Tape) Test No. 6: Grammar--TVL Qualitative Tests (written) Test NO. 7: Q(CEM). Q(CEs). Q(CET), Q(CH), C(CK) Test NO. 8: Q(CKH), Q(CP). Q(Cs). Q(CT) Inventory NO. 9 and No. 10: Cultural Determinants—-I,A,F 64 Inventory NO. 11: Modalities of Inference-- M,D,R,L Inventory No. 12: Personal Code Inventory Qualitative/Non-Written Test No. 1: Qualitative Auditory Test NO. 2: Qualitative Olfactory COpies Of Cognitive Style Diagnostic Test Battery ma- terials, manuals, and tests are available in the Testing Center library at the Orchard Ridge Campus of Oakland Community College. The Cognitive Style Diagnostic Test Battery is made up of items developed at Oakland Community College and items selected from several other measures. These other measures include the Gates Reading Tests, the Dif- ferential Aptitude Test, the Iowa Tests of Educational Development, the Mueller Auditory Test, the Nottus Pattern Test, the Raven Mattrices Survey, and the Science Re- search Associates Batteries. State-Trait Anxiety Inventory (STAI) The State-Trait Anxiety Inventory is divided into two parts labled Forms X-l, measuring state anxiety, and X-2, measuring trait anxiety. The STAI, copyright 1968, was developed by Charles D. Spielberger, Richard L. Gorsuch, and Robert E. Lushene for use with secondary level students and adults in the isolation and measurement 65 of state and trait anxiety. Reliability and validity for the STAI are reported in Appendix A. Research and Design Procedures Fifty-one freshmen students enrolled in the Life Science course at the Orchard Ridge Campus of Oakland Community College participated in this study. The stud- ents were enrolled in the conventional manner, with no attempt to group the students according to cognitive style for instructional purposes. For the purposes of this study, the cognitive style map for each student was compared to the cognitive style map for the audio-tutorial instructional mode in order to identify those students matched to the instructional mode and those students non- matched. A—Trait anxiety levels for each student were de- termined by administering the State-Trait Anxiety Inven- tory form x-2 during the pre-instruction orientation period for this course. The students were categorized as: Matched-~High A-trait, Matched--Low A-Trait, Non-Matched --High A-Trait, and Non-Matched--Low A-Trait. The State- Trait Anxiety Inventory form X-l was administered at the completion of each of three weekly audio-tutorial sessions. The mean reduction of A-State scores for students matched to the instructionalrmxkawas compared to the mean reduction of A-State scores for students non-matched to 66 the instructional mode. The distribution of mean reduc- tion of A-State scores for each group was displayed in the Kolmogorov-Smirnov Analysis to determine whether there was a greater degree of reduction in A-State scores for students matched to the instructional mode as compared to those students non-matched. The Kolmogorov-Smirnov Analysis will be discussed under AnalySis of the data in this chapter. The following flow chart displays the re- search design procedures. Statistical Hypotheses Three testable hypotheses were generated to ex- amine whether the degree Of reduction in anxiety levels of students with cognitive styles matched to the in- structional mode would be greater than that of students non-matched. The null hypotheses and alternate hypotheses are as follows: H01: The degree of reduction of A-State anxiety levels among students with cognitive styles that are non-matched with the cognitive style of the instructional mode will be no different from that of students with cogni- tive styles that are matched. H 1: The degree of reduction of A-State anxiety levels among students with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of students with cognitive styles that are non-matched. H 2: The degree of reduction of A-State anxiety levels among high A-Trait students with cognitive styles non-matched with the 67 Match Administer Audio- «—i Cognitive Style tutorial Battery Cognitive Form Null and Style? Alternate Hypotheses Administer Non- L+. STAI Match Match. Non- Match High High High Hi h g Match Non- or Low Match A-Trait? ' Low Low Low NO .1 '0' Is this the third Administer administration Instruction -Stat of A—State? A e Yes Reject Null- Accept alternate Hypothesis Significant V Differences in A-State Levels ' Ex1t Between Matched (Report) and Non- Matched? Fail to . Reject Null Hypothesis Figure 4.--Research and Design Procedures. 68 cognitive style of the instructional . mode will be no different from that of high A-Trait students with cognitive styles that are matched. The degree of reduction of A-State anxiety levels among high A—Trait students with cognitive styles matched with the cognitive style of the instructional mode will be greater than that of high A-Trait students with cognitive styles that are non-matched. The degree of reduction of A-State anxiety levels among low A-Trait students with cognitive styles non-matched with the cog- nitive style of the instructional mode will be no different from that of low A-Trait students with cognitive styles that are matched. The degree of reduction of A-State anxiety levels among low-A-Trait students with cognitive styles matched with the cogni- tive style Of the instructional mode will be greater than that of low A-Trait stud- ents with cognitive styles that are non- matched. The achievement levels of students with cognitive styles that are non-matched with the cognitive style of the instruc- tional mode will be no different from those of students with cognitive styles that are matched. The achievement levels of students with cognitive styles that are matched to the cognitive style of the instructional mode will be greater than those of students with cognitive styles that are non-matched. The achievement level of high A-Trait students will be no different from that of low A-Trait students. The achievement level of low A-Trait stud- ents will be greater than that of high A- Trait students. 69 Analysis of the Data The data indicating the degree of anxiety level reduction as measured by the X-2 form of the STAI was analyzed using the two-sample Kolmogorov-Smirnov test for the analysis of variance. The significance of the findings were tested against the two-tailed chi-square at the .05 alpha level. Hill and Kerber describe the Kolmogorov—Smirnov as, . . . a statistical test of whether two independ- ent samples have been drawn from the same universe, or from universes with the same distribution. . . . If the two samples have actually been drawn from the same pOpulation or populations having the same distribution, the cumulative distributions of both samples should be reasonably close to each other over the range of values involved. If the two-sample cumulative distributions evidence too much divergence at any point, there is a given probability that the samples might come from dif- ferent populations. If the deviation between the two cumulative distributions at any point is so great that it would occur, according to the appro- priate Kolmogorov-Smirnov probability distribution, less than 5 percent, or 1 percent of the time, due to chance factors alone, the null hypothesis (no difference between the cumulative distribu- tions) is rejected in favor of the statistical alternative hypothesis.74 Summary This study was conducted at the Orchard Ridge Campus Of Oakland Community College, Farmington, Michi- gan. The Cognitive Style Diagnostic Test Battery Joseph E. Hill and A. Kerber, Models, Methods and Analytical Procedures in Education Research (Detroit: Wayne State University Press, 1954), p. 311. 70 and the State—Trait Anxiety Inventory form x-2 were util- ized to categorize the fifty-one students enrolled in the Life Science course into Matched-~High A-Trait, Matched-- Low A-Trait, Non-Matched--High A-Trait, and Non-Matched —-Low A-Trait groups. The State-Trait Anxiety Inventory form X-l was utilized to measure the A-State anxiety level immediately following each audio-tutorial session. The Kolmogorov- Smirnov model for the analysis Of variance was used to determine whether those students with cognitive styles that matched the cognitive style of the instructional mode had a greater degree Of reduction in A-State anxiety levels than did the students with cognitive styles that did not match. The significance of the findings were tested against the two-tailed chi-square with two degrees of freedom at the .05 level. Discussed in this chapter are the limitations of the study. A description of the Cognitive Style Diagnos- tic Test Battery and a description Of the State-Trait Anxiety Inventory with supporting reliability and valid- ity data are included in this chapter. Also included are the statistical hypotheses tested in this study and an introduction to the Kolmogorov-Smirnov model for the analysis of variance with the test of significance. CHAPTER IV ANALYSIS OF THE DATA Included in this chapter are reports of the data and an analysis of the findings. The purpose Of this study is to explore whether the matching of student cog- nitive styles to the instructional mode will result in the reduction of student anxiety levels. The hypotheses advanced for testing in this study include: H1: H2: H3: H4: The degree of reduction of A-State anxiety levels among students with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of students with cognitive styles that are non-matched. The degree of reduction of A-State anxiety levels among high A-Trait students with cognitive styles matched with the cognitive style of the instructional mode will be greater than that of high A-Trait students with cognitive styles that are non-matched. The degree of reduction of A-State anxiety levels among low A-Trait students with cognitive styles matched with the cognitive style of the instructional mode will be greater than that of low A-Trait students with cognitive styles that are non-matched. The achievement levels of students with cognitive styles that are matched to the cognitive style of the instructional mode will be greater than those of students with styles that are non-matched. 71 72 H5: The achievement level of low A-Trait stud- ents will be greater than that of high A-Trait students. The hypotheses were analyzed using the Kolmogorov-Smirnov Analysis of variance. The significance of the findings Were tested against the two—tailed chi-square at the .05 alpha level. Findings of the Stugy The raw data collected in this study are reported in Table 12. Under A-Trait are reported the scores from the STAI for X-2. Under A-State are reported the mean reduction of succeeding administrations of STAI form X-l. The column Grade indicates the summative grade reported by the instructor. Under Cognitive Style M indicates the student is matched to the instructional mode, whereas N indicates the student is non-matched. It should be noted that complete data was collected for thirty-nine subjects. All subjects were tested for anxiety levels but the anxiety data for twelve subjects was lost. A student assistant, paid to aid in the admin- istration of the course, was responsible for collecting the data. The student assistant simply could not account for the data. The grades and cognitive styles Of the subjects with incomplete data are reported because they will be included in the examination of whether cognitive style may be used to predict achievement level. The data 73 available would indicate that these subjects were not an extraordinary group. Eight subjects had cognitive styles matching that of the instructional mode, whereas four were non-matched. Grades earned by these subjects include two A's, one B, four X's (failures, one I (incomplete, and four W's (withdrawals). The distribution of the grades of the students with anxiety data was compared to that of students with lost anxiety data to determine whether there was a sig- nificant difference between the distributions. The grades of the matched students with anxiety data was compared to those of the matched with lost anxiety data and the grades of the non-matched students with anxiety data were com- pared to those of non-matched students with lost anxiety data. The two-sample Kolmogorov-Smirnov analysis of variance was used to test whether there was a difference between the distributions. The significance of the find- ings was tested against the two-tailed chi-square at the .05 alpha level. The results indicated that there was not a significant difference in the distributions of the grades between the groups with anxiety data and those with lost data. Therefore, it is assumed that the loss of these subjects did not affect the findings of the study. The relevant tables and computations may be found in Appendix B. 74 It should also be noted that two withdrawals and one incomplete are included among the data for the thirty- nine subjects with complete data. Each of these three subjects completed the portion of the course with which this study is concerned, therefore they may be included in the study. The A-Trait anxiety scores for the subjects in- volved in the study had a mean Of 40.07 with a standard deviation of 7.28. This compares very favorably with the normative data reported in Appendix B. High A-Trait is established at the raw score level of 44 and above. This level conforms to the 80th through the 90th percentile ranks reported in the STAI normative data.77 Subjects ranked in the 80th through 90th percentile are considered to be high anxious. The mean A—Trait score for participants in this study who were rated high anxious is 47. The research hypotheses examine the notion that students with cognitive styles matched to the instruc- tional mode should have greater reduction in A-State anxiety levels than students who are non-matched. The first hypothesis does not take into account differences among A—Trait levels of students. The second hypotheses isolates high A-Trait students and the third hypothesis isolates low A—Trait students. 75 TABLE l.-—Report of the Data. Student A-Trait A-State Grade cogziiive l 24 + 4 C M 2 27 ~14 B M 3 28 - 1 B M 4 29 -14 X N 5 31 - 2 A M 6 33 -19 B M 7 33 6 D M 8 34 1 B M 9 35 - 9 C M 10 35 - 2 X M 11 35 + 1 B M 12 35 - 2 W M 13 35 -14 X M 14 36 -12 D N 15 36 - 4 A M 16 38 - 6 A M 17 39 -15 A M 18 40 ~12 A N 19 40 - 1 B M 20 41 + 7 W N 21 41 -19 I N 22 43 -12 X N 23 43 +18 A M 24 44 + 1 D N 25 44 - 4 C M 26 44 -16 A M 76 TABLE l.--Report Of the Data (Continued). Student A—Trait A-State Grade cogziiive 27 45 +11 B N 28 46 - 6 X N 29 46 + 8 x N 30 46 + A N 31 46 - 9 B M 32 47 -10 A M 33 47 - 9 B M 34 47 + 9 X M 35 48 - 3 B M 36 48 - 2 A M 37 48 -17 X M 38 49 - 4 B M 39 57 - 4 D M 40 I N 41 X N 42 X N 43 w N 44 x N 45 w M 46 w M 47 B M 48 x N 49 W M 50 A M 51 A M 77 Hypothesis 1 Null Form: The degree of reduction of A-State Anxiety levels among students with cognitive styles that are non-matched with the cognitive style of the instructional mode will be no different from that of students who are matched. H01: SARI = SAR2 SARl denotes student anxiety reduction of the D matched group and SAR2 denotes student anxiety reduction of the non-matched group. Alternative Form: The degree of reduction of A- State anxiety levels among students with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of students with cognitive styles that are non-matched. H11: SARl > SAR2 In Table 2 on the following page, the reader will note that the number of subjects (n) is not equal in the two groups. In order to form a common denominator for com- parison purposes the number of subjects in the matched group is multiplied by the number of subjects in the non-matched group, i.e., (n1) (n2) = 308 or (28) (11) = 308. Each event in the interval then forms a cumulative fraction. The matched group has one event in the +20-16 interval which forms the fraction ll/308. Another event in the + 10-6 interval forms the cumulative fraction 33/308. The dif- ferences between the groups are compared by subtracting 78 TABLE 2.--KolmOgorov-Smirnov Analysis of Hypothesis 1. Change in A-State Anxiety Levels Intervals + .. 20-16 15-11 10-6 5-0 0-5 6-10 11-15 16-20 n Groups Cumulative Frequency 1 2 3 ll 5 3 3 28 Match 11 ll 33 66 187 242 275 308 308 308 308 308 308 308 308 308 l 3 l l 4 1 11 Non- 28 112 140 140 168 280 308 Match 308 308 308 308 308 308 308 ll 17 79* 74 47 74 5 Diff. 308 308 308 308 308 308 308 308 *Greatest degree of difference between the distributions. the cumulative fractions found in each interval. The great- est degree of difference is then tested for significance. The greatest degree of difference shown in Table 2 is 79/308 or .256. The significance of this difference is tested against the two—tailed chi-square at the .05 alpha level. The formula for testing is: D indicates the greatest degree of difference taken from Table 2. The number of subjects in each group are denoted 79 by n1 and n2. The formula is computed in the following ’ manner: (I! ) (n ) X2 : 4D2 D1 + n2 1 2 .262 308 9 80.696 1 39 ( .11 X2 = 2.069 The computed test value (2.069) is less than the chi-square value with two degrees of freedom at the .05 alpha level (5.99). Therefore, the degree of difference between the distributions of the two groups displayed in Table 2 is not sufficient and the null hypothesis cannot be rejected. Hypothesis 2 Null Form: The degree of reduction of A-State anxiety levels among high A-Trait students with cognitive styles non-matched with the cognitive style of the instructional mode will be no dif- ferent than that of high A-Trait students with cognitive styles that are matched. H02: SARl = SAR2 Alternative Form: The degree of reduction of A- State anxiety levels among high A-Trait students with cognitive styles that are matched with the cognitive style of the instructional mode will be 80 greater than that of high A-Trait students with cognitive styles that are non-matched. H12: SARl > SAR2 TABLE 3.--Kolmogorov-Smirnov Analysis of Hypothesis 2. Change in A-State Anxiety Levels Intervals . + - 20-16 15-11 10-6 5-0 0-5 6-10 11-15 16-20 Groups Cumulative Frequency 1 5 3 2 Match __5_ _5_ .31 4_5 11 11 55 55 55 55 55 55 l 2 l 1 Non- 11 .13. £1. 11 11 ”atCh 55 55 55 55 55 Diff. 11 .21 3_9* 11 _5 _5 _9_ 55 55 55 55 55 55 55 *Greatest degree of difference between the distributions. The greatest degree of difference found in Table 3 is 39/55 or .709. This value was tested with the Kolmogorov- Smirnov two-tailed two sample test. The computed test value (6.902) is greater than the chi-square value with two degrees Of freedom at the .05 level (5.99). There- fore, the null hypothesis may be rejected and support 81 inferred for the alternate hypothesis of a greater degree Of reduction in anxiety levels among high A-Trait students with cognitive styles matched to the instructional mode. Hypothesis 3 Null Form: The degree of reduction of A—State anxiety levels among low A-Trait students with cognitive styles that are non-matched with the cognitive style Of the instructional mode will be no different from that of low A-Trait students with cognitive styles that are matched. H03: SARl = SAR2 Alternative Form: The degree of reduction of A- State anxiety levels among low A-Trait students with cognitive styles that are matched with the cognitive style of the instructional mode will be greater than that of low A-Trait students with cognitive styles that are non-matched. H13: SARl > SAR2 The greatest degree of difference found in Table 4 on the following page is 61/102 or .598. This value was tested with the Kolmogorov-Smirnov two-tailed two sample test. The computed test value (6.332) is greater than the chi-square value with two degrees of freedom at the .05 level (5.99). Therefore, the null hypothesis may be rejected, however, support is not inferred for the al- ternate hypothesis since the direction of the distribution indicates SARl< SARZ. Support is inferred for the greater 82 TABLE 4.--Kolmogorov-Smirnov Analysis of Hypothesis 3. Change in A-State Anxiety Levels Intervals + _ 20-16 15-11 10-6 5-0 0-5 6-10 11-15 16-20 n Groups Cumulative Frequency 1 1 3 6 2 3 1 17 Match 6 6 12 30 66 78 96 102 102 102 102 102 102 102 102 102 l 4 l 6 Non- l7 l7 l7 17 85 102 “atCh 102 102 102 102 102 102 Diff. 6 6 5 13 49 61* 11 0 102 102 102 102 102 102 102 102 *Greatest degree of difference between the distributions. reduction of anxiety levels among low A-Trait students non-matched to the instructional mode which is the exact opposite of the alternative hypothesis. Secondary Findings Since one of the major foundations of this study is cognitive style and one of the claims of cognitive style is that matching student cognitive styles to the instructional mode will affect learning, it was felt that this notion should be tested. Testing this notion would also determine whether the instructional mode cognitive 83 map that student cognitive styles were being matched to had any predictive validity in this study. The following hypothesis was formulated and tested: Null Form: The achievement levels of students with cognitive styles that are non-matched with the cognitive style of the instructional mode will be no different from those of students with cognitive styles that are matched. ALl denotes achievement level of the matched group and AL2 denotes achievement levels of the non-matched group. Alternative Form: The achievement levels of stud- ents with cognitive styles that are matched to the cognitive style of the instructional mode will be greater than those of students with cognitive styles that are non—matched. H14: ALl > AL2 Hypothesis four was tested twice; once with only those students who completed the instruction and second including those students with incompletes and withdrawals. In Tables 5 and 6 X denotes failure, W indicates with- drawaland I designates incomplete. The greatest degree of difference found in Table 5 is 168/352 or .477. This value was tested with the Kolmo- gorov-Smirnov two-tailed two sample test. The computed test value (7.432) is greater than the chi-square value 84 TABLE 5.--Kolmogorov-Smirnov Analysis of Hypothesis 4. Achievement Levels A B C D X n Groups Cumulative Frequency 10 10 4 2 6 32 Matched 110 220 264 286 352 352 352 352 352 352 2 l 2 6 11 Non- 64 96 96 160 352 MatChed 352 352 352 352 352 Diff. 46 124 168* 126 0 352 352 352 352 352 *Greatest degree of difference between the distributions. TABLE 6.--Kolmogorov-Smirnov Analysis of Hypothesis 4. (Incompletes and withdrawals included) Achievement Levels A B C D X W I n Groups Cumulative Frequency 10 10 4 2 6 4 36 Matched 150 300 360 390 480 540 540 540 540 540 540 540 2 1 2 6 2 2 15 Non- 72 108 108 180 396 468 540 MatChed 540 540 540 540 540 540 540 78 192 252* 210 84 72 0 540 540 540 540 540 540 540 *Greatest degree of difference between the distributions. 85 with two degrees of freedom at the .05 level (5.99). There- fore, the null hypothesis may be rejected and support in- ferred for the alternative hypothesis of greater achieve- ment levels for students matched to the instructional mode. The greatest degree of difference found in Table 6 is 252/540 or .466. This value was tested with the Kolmogorov-Smirnov two-tailed two sample test. The com- puted test value (9.190) is greater than the chi-square value with two degrees of freedom at the .05 level (5.99). Further examination revealed that the computed test value (9.190) is greater than the chi-square value with two de- grees of freedom at the .02 level (7.82). Therefore, the null hypothesis may be rejected and support inferred for the alternate hypothesis of greater achievement levels for students matched to the instructional mode. Another major foundation for this study is the notion that anxiety level has debilitating effects on achievement. However, the premises of this study are that matching cognitive styles would in effect reduce the anxiety levels, thus reducing the effects that high anxiety might have on achievement. It was decided to test whether there was a relationship between A-Trait anxeity and achievement in spite of cognitive style matching. The following hypothesis was formulated and tested: 86 Null Form: The achievement level of high A-Trait students will be no different from that of low A- Trait students. H05: ALl = AL2 ALl denotes achievement levels of low A-Trait students and AL2 denotes achievement levels of high A- Trait students. Alternative Form: The achievement level of low A-Trait students will be greater than that of high A-Trait students. H 5: AL > AL 1 1 2 TABLE 7.—-Kolmogorov—Smirnov Analysis of Hypothesis 5. Achievement Levels A B C D X I W n Groups Cumulative Frequency 4 5 l 2 4 16 High 148 333 370 444 592 592 592 592 592 592 6 11 4 4 6 3 3 37 Low 96 272 336 400 496 544 592 A'Tralt 592 592 592 592 592 592 592 Diff. 52 61 34 44 96* 48 o 592 592 592 592 592 592 592 *Greatest degree of difference between the distributions. 87 The greatest degree of difference found in Table 7 is 96/592 or .162. This value was tested with the Kolmogorov- Smirnov two-tailed two sample test. The computed test value (1.172) is less than the chi-square value with two degrees of freedom at the .05 alpha level (5.99). There- fore, the null hypothesis cannot be rejected. Discussion of the Findings Three research hypotheses were generated and tested to examine the theory that students with cognitive styles matched to the cognitive style of the instructional mode would have a greater reduction in A-State anxiety levels than students with non-matched cognitive styles. Two secondary hypotheses were generated and tested to examine the relationship of cognitive style matching to achievement and the relationship of A-Trait anxiety levels to achieve- ment. A summary of the results of the hypothesis testing is reported in Tables 19 and 20. The first hypothesis does not take into account differences among A-Trait levels of students. Findings of no significant difference conform to the experiences of other researchers in the field of anxiety when they have not distinguished between A-Trait anxiety and A-State 78 anxiety. The necessity of distinguishing between kinds 78Spielberger, The Effects of Anxiety on Computer Assisted Learning, p. 30. 88 TABLE 8.-—Summary of Research Hypothesis Testing. Research Hypothesis Results l. The degree of reduction of A-State anxiety levels among students with cog- nitive styles that are matched with the cognitive style of the instruc- tional mode will be greater than that of students with cognitive styles than are non-matched. 2. The degree of reduction of A-State anxiety levels among high A-Trait students with cognitive styles matched with the cognitive style of the in- structional mode will be greater than that of high A—Trait students with cognitive styles that are non—matched. 3. The degree of reduction Of A-State anxuyarlevels among low A-Trait students with cognitive styles matched with the cognitive style of the instructional mode will be greater than that of low A-Trait students with cognitive styles that are non-matched. Null could not be rejected. Null rejected at the .05 level. Null rejected at the .05 level. Support not in- ferred for alter- nate hypothesis. TABLE 9.-—Summary of Secondary Hypothesis Testing. Additional Hypotheses Results 1. The achievement levels of students with cognitive styles that are matched to the cognitive style of the instruc- tional mode will be greater than those of students with cognitive styles that are non-matched. 2. The achievement levels of low A- Trait students will be greater than those of high A-Trait students. Null rejected at the .05 and the .02 levels. Null could not be rejected. 89 of anxiety is supported in the testing of the second and third hypotheses. The second hypothesis tested allowed for the con- trol of A-Trait anxiety by limiting the group to only those subjects with high A-Trait anxiety. Findings of signifi- cant difference in the degree of A-State anxiety levels between students with cognitive styles matched to the cog- nitive style of the instructional mode and students with non-matched cognitive styles supports the basic notion of this study. Cognitive style matching can be utilized as an objective means to control anxiety levels in the class- room. The third hypothesis tested allowed for the con- trol of A-Trait anxiety by limiting the group to only those subjects with low A-Trait anxiety. The null hypothe- sis was rejected, however, the findings of significant difference were inthe direction of a greater degree of reduction of A—State anxiety levels among students non- matched to the cognitive style of the instructional mode. This finding is not viewed by this researcher to be con- tradictory to the findings of the second hypothesis. Only one of the five low A-Trait subjects with a greater reduction in A—State anxiety level received a grade above D. This would suggest that low A-Trait students non- matched to the cognitive style of the instructional mode 90 are not sufficiently stimulated to achieve. This inter- pretation is supported by the reports of researchers con- cerning Drive Theory.79 Further research is required in this area. Under secondary findings the results of the fourth hypothesis tested is that there are significant differ- ences in the level of achievement between students with cognitive styles matched to the instructional mode and those students non-matched. This finding supports the predictive capability of cognitive style matching in re- gard to achievement levels. Significant findings for this hypothesis also established the validity of the in- structional mode cognitive style map for use in this study for the purpose of matching students to the instructional mode. The fifth hypothesis tested examined the possi- bility that differences in A-Trait were related to achieve- ment level in this study. Findings of no significant dif- ference in the achievement levels between high A-Trait and low A-Trait students coupled with the findings of the fourth hypothesis suggests that matching student cogni- tive styles to the instructionalmode nullified any debili- tating effects that high A-State anxiety might have had 790'Neil, p. 13. 91 on achievement. Thus, findings of no significant differ- ence in the fifth hypothesis are supportive of the basic theory of the study that matching cognitive styles would control anxiety levels. CHAPTER V SUMMARY, CONCLUSIONS AND IMPLICATIONS FOR FURTHER RESEARCH Summary The purpose of this study was to explore whether the matching of student cognitive styles to the cogni- tive style of the instructional mode would reduce student anxiety levels. A review of the literature pointed out that, whereas, mild anxiety might serve as a stimulus to learning, high anxiety is debilitating. The literature suggests that it is essential to distinguish between a person's tendency to react with anxiety (trait) and a person's anxiety reaction in a given situation (state). It was also observed in the literature that persons learn to cope with anxiety in given situations or settings. This study is founded onthe idea that by matching the cognitive style of the student to the cognitive style of the instructional mode it is possible to place students in instructional situations with which they are able to copy and thus reduce their anxiety level. The subjects of this study were fifty-one fresh- men students enrolled in an audio-tutorial Life Science 92 93 course at Oakland Community College. Cognitive style maps for each of the students were examined to determine whether they matched the cognitive style of the instruc- tional mode. A-Trait anxiety data was collected during the class orientation period for thirty-nine of the stud- ents. A-State anxiety data was collected at the end of each audio-tutorial session. Anxiety data is not avail- able for twelve students due to administrative error. Three hypotheses were tested to determine whether the matching of student cognitive styles to the instruc— tional mode would reduce anxiety levels. Two secondary hypotheses were tested to support the predictive validity of the instructional mode cognitive style map and to examine whether there was a difference in the achievement levels between students with high A-Trait and those with low A-Trait. The data for each of the null hypotheses was tested using the Kolmogorov-Smirnov two sample analy- sis of variance. The significance of the findings were tested against the two-tailed chi-square with two degrees of freedom at the .05 level. The findings of this study supported the basic idea of the study, that high A-Trait students with cogni- tive styles matched to the instructional mode Will have greater reduction of A—State anxiety levels than those students with cognitive styles that are non-matched. 94 Support was also found for the predictive validity of the instructional mode cognitive style map. This support was founded on the successful prediction of the achievement level of students with cognitive style maps matching the cognitive style of the instructional mode. Support was not found for a hypothesis which attempted to indicate reduction in anxiety levels, but which eliminated discrimi- nation between trait and state anxiety as a control. Discussion The initial direction taken by this researcher was to use cognitive style matching as the only control intimastudy of the modification of anxiety levels. The review of the literature, however, indicated that it is necessary to discriminate between trait and state anxiety in order to reveal changes in anxiety levels. The findings of this study supported the necessity of discriminating between trait and state anxiety. The first null hypothe- sis, which employed cognitive style matching as the only control, could not be rejected. The second null hypothe- sis, which employed both A—Trait anxiety and cognitive style matching as controls, was rejected. The fifth null hypothesis, which employed high and low A-Trait anxiety as the only control, could not be rejected. This re- searcher feels these findings suggest that anxiety studies 95 which fail to discriminate between trait anxiety and state anxiety and which do not control for cognitive style are unlikely to return significant findings. The findings of the third hypothesis, that low A-Trait non-matched students have greater reduction in anxiety levels, in conjunction with the findings of the fourth hypothesis, that non—matched students tend to earn lower grades, suggest that cognitive style should be explored as a means of raising student anxiety levels where appropriate. The literature discusses the role of anxiety as a motivating force. Moderate anxiety acts in Drive Theory as a means of stimulating low A-Trait stud- ents to achieve. The literature warns against artifi- cially creating anxiety to stimulate students because of the dangers of inducing high anxiety. However, if a student is properly matched to the instructional mode sufficient anxiety should be present to afford the appro- priate amount of stimulation. The findings related to the fourth hypothesis of this study suggest that cognitive style matching has poten- tial to optimize the student's chances of success by match- ing him to the instructional mode best suited to his cogni- tive style. Cognitive style matching could be employed as an objective, easily managed method for the individuali- zation of instruction. The outcome of using cognitive 96 style matching to individualize instruction would be to realize the goal of reducing high anxiety levels of stud— ents and the Optimizing of a student's potential success in achievement. Persons with the responsibility of develOping and designing instructional materials are faced with the problem of selecting appropriate instructional strategies and media. Frequently the content and the instructional milieu are definite considerations in the selection. How- ever, individualization of instruction to meet individual student needs requires consideration of instructional strategies and media which consider the characteristics of the student. Student-instructional mode cognitive congruence across the full range of learner characteris- tics should be considered for the individualization of instruction. Cognitive style matching affords the in- structional developer-designer the Opportunity to consider a wide range of learner characteristics for the selection of strategies and media. This study indicates the modifi- cation Of anxiety level as a dimension in the utilization of cognitive style matching and, thus, illustrates the value of employing cognitive style matching for individuali- zation of instruction. 97 Conclusions The conclusions of this study are drawn in full recognition of the limitations of the study. The research was conducted in a community college setting, employing the audio-tutorial instructional mode in a science class. The results might be quite different if a different set- ting, instructional mode, or subject area was employed. The conclusions drawn in this study include: 1. High A—Trait students with cognitive styles matched to the cognitive style of the instructional mode tend to have greater reduction of A-State anxiety levels than non-matched students. 2. Students with cognitive styles matched to the cognitive style of the instructional mode tend to achieve higher grades than students that are non-matched. 3. Low A—Trait students with cognitive styles non-matched to the cognitive style of the instructional mode tend to have greater reduction of A-State anxiety levels than matched students. 4. It is essential to control A-Trait as one of the variables in the study of the reduction of A—State anxiety levels. 5. It is essential to control for cognitive processes, through cognitive style matching or some Other equally effective method, in the study of the effects of anxiety on achievement. 98 Implications for Future Research This research is limited by the relatively small number of subjects involved. Similar results in future studies would lend support to the present findings. This research is further limited by the restriction of the sub- jects to urban community college students enrolled in a freshman science course. Replication of this research in different educational settings, with students of varied age levels, and in varied instructional modes might re- sult in quite different findings. Similar findings in varied situations would enhance generalizability. An- other limitation to this study is the inability to deter- mine whether changes in anxiety level were caused by in- class or external events. It is recommended that future research employ some means of assessing anxiety levels of students prior to each time they enter the class setting. The reduction Of anxiety levels was achieved in this study by employing cognitive style matching. Similar results might occur using other diagnostic and prescrip- tive systems. Researchers working with other systems should investigate the effects on anxiety level. This study employed the full Cognitive Style Diag- nostic Test Battery which requires four hours to adminis- ter. Many institutions do not provide such testing services for system—wide use. In such a setting the 99 individual instructor might be reluctant to use four hours of class time for diagnostic testing. It might be possi- ble to develop mini-batteries which relate to specific content areas or situations and require less time for ad- ministration. Research should be conducted to investigate the feasibility of developing and using such mini-batteries. Finally, it is recommended that future research investigate the implications of the findings of hypothesis three in this study. The findings of hypothesis three suggest that non-matched low A-Trait students are not suffi- ciently stimulated to achieve. Future research should ex- amine the notion that matching students with an appropriate instructional mode would stimulate those students to achieve. Such future research should be conducted within the frame- work of achievement motivation. The affect arousal model of achievement motivation theorizes that motivation occurs as a result of a change in affect. Absence of change in affect results in a static condition which may be labeled in some cases as boredom. Mild change in affect results in motivation, however, ex- cessive change in affect results in learning debilitation or maladjustment. Future research should investigate the potential of using cognitive style matching to place stud- ents in learning situations with sufficient stimulation to induce motivation but not excessive enough to be de- bilitating. APPENDICES 100 APPENDIX A STATE-TRAIT ANXIETY INVENTORY: RELIABILITY AND VALIDITY 101 APPENDIX A STATE-TRAIT ANXIETY INVENTORY: RELIABILITY AND VALIDITY Reliability Test-retest reliability data for the STAI are presented in Table 10. TABLE 10.--Test-Retest Reliability--STAI.a Time Lapse: 1 hour T/R 20 day T/R 104 day T/R N r N r N r A-Trait Males 88 184 38 186 25 .73 Females 109 .76 75 .76 22 .77 A-State Males 88 .33 38 .54 25 .33 Females 109 .16 75 .27 22 .31 aCharles D. Spielberger, R. L. Gorsuch, and R. E. Lushene, STAI Manual (Palo Alto, Calif.: Consult- ing Psychologists Press, Inc., 1970), P. 9. The relatively low r's for the A-State scale may be ex- pected because the level of A-State would fluctuate ac- cording to situations. Alpha reliability coefficients for the STAI scales were cOmputed by formula K-R 2080 80Hill and Kerber, p. 290. 102 103 for the normative samples. These reliability coefficients show good internal consistency and are presented in Table 11. TABLE ll.--STAI Means, Standard Deviations, and Alpha Reliabilities.a Freshmen Undergraduates HS Students 3W Male Female Male Female Male Female E A-Trait 1 Mean 38.07 38.22 37.68 38.25 39.37 41.61 N 332 644 253 231 190 187 SD 8.20 8.20 9.69 9.14 9.40 11.29 Alpha .86 .86 .90 .89 .89 .92 A-State Mean 40.01 39.39 36.35 35.12 36.99 37.57 N 334 648 253 231 190 187 SD 7.85 8.62 9.67 9.25 9.07 11.76 Alpha .83 .86 .89 .89 .86 .92 aSpielberger, STAI Manual, p. 8. Additional evidence of the internal consistency of the STAI scales is provided by item-remainder correlations. These are reported in Table 12 on the following page. The test—retest reliability of the STAI A-Trait scale is high. Reliability coefficients for the STAI A-State scale tend to be low, as would be expected for a scale intended to measure transitory levels of anxiety. However, both the A-Trait and A-State anxiety scales have a high degree of internal consistency.81 81Spielberger, STAI Manual, p. 10. 104 TABLE 12.--Item Remainder Correlation Coefficients--STAI.a A—State Scale A—Trait Scale Item Male Female Male Female 1 2 3 l 2 3 l 2 3 1 2 3 l 50 65 58 54 58 68 39 49 51 47 40 61 2 50 61 51 56 6O 66 30 41 44 36 35 35 3 41 58 56 46 56 7O 23 37 37 36 50 64 4 22 39 26 27 32 51 50 50 52 46 58 64 5 51 65 60 60 58 67 44 37 46 34 39 48 6 46 40 46 51 38 67 36 50 47 38 49 57 7 47 47 47 45 41 52 55 48 64 52 60 63 8 21 30 42 23 36 44 46 61 61 51 55 78 9 13 45 17 22 48 27 50 51 55 49 56 48 10 39 35 38 44 50 67 58 55 55 58 62 68 ll 43 44 40 49 50 57 43 56 4O 48 48 53 12 48 56 58 52 63 72 55 59 6O 46 53 48 13 44 55 57 46 51 67 63 66 63 58 61 67 14 39 52 30 36 58 52 34 39 23 26 31 19 15 58 64 61 57 7O 72 52 55 51 52 51 69 16 44 65 41 57 56 62 48 58 56 55 59 67 17 54 60 58 6O 54 69 43 55 46 38 57 52 18 40 6O 35 29 28 47 58 61 57 47 54 54 19 30 37 39 42 34 51 43 56 52 42 47 68 20 39 59 53 53 53 62 40 64 56 47 59 54 Sample 1: Freshmen: 334 M, 648 F 2: Undergraduates: 253 M, 231 F 3: H. S. students: 190 M, 187 F aSpielberger, STAI Manual, p. 24. 105 Validity Concurrent validity for the A—Trait scale was determined by correlations with the IPAT Anxiety Scale and the Taylor Manifest Anxiety Scale. These correla- tions are reported in Table 13 on the following page. Construct validity for the A-State scale was established by administering the A-State scale under four different conditions to the same group. The conditions included no treatment (normal), relax training (relax), short exam (exam), and after a film depicting accidents in a wood shop (movie). Alpha reliability coefficients for these conditions are presented in Table 14 on the following page. Thus, the A-Trait scale was found to have high concurrent validity and the A-State was determined to have high construct validity. 106 .HH .e .emsemz Heem .Hmmumnememmm mm. Hm.m om.m~ mm. mo.m oe.mm xeqmm Hm. e~.oe am.hm mm. sm.m mm.mm qmzmoz mm. mm.HH mm.me mm. mm.ea Ho.me zexm mm. mm.HH em.om em. mv.me mo.om mH>oz medee am new: meda< am new: . mCOHuflocoo Ammuzv mmHmem Amoalzv mmamz m.Hoo ouooomum .mcmozll.va mqmde .eeae map 6cm Heem map emmSpmn monumemppoo man now men u 26 .OH .8 .Hmsemz Heem .nmmnmnamnemm as. am. mm. mm. mm. mm. Scam 8m. mm. mm. on. mm. om. mesa nee. me. me. Bade eamH Heem mmze same Haem maze (same Hmem oaoom Apmnzv Romuzv Apmenzv somexee mucmflumm mmamz mmHmEmm mz memeeoo memeeoo m.HdBmllmGOHDMHOHHOUII.MH mqmflh APPENDIX B ANALYSIS OF THE SIGNIFICANCE OF INCOMPLETE ANXIETY DATA 107 APPENDIX B ANALYSIS OF THE SIGNIFICANCE OF INCOMPLETE ANXIETY DATA In Table 15 the following null hypothesis was tested: There is no difference in the distribution of grades between those matched students with lost anxiety data (L.D.) and those matched students with anxiety data (W.D.). The hypothesis was tested using the Kolmogorov-Smirnov analysis of variance. The significance of the findings were tested against the two-tailed chi-square with two de- grees of freedom at the .05 alpha level. The greatest degree of difference found in Table 15 (on the following page) is 108/288 or .375. This value was tested with the Kolmogorov-Smirnov two-tailed two sample test. The computed test value (3.68) was less than the chi-square value with two degrees of freedom at the .05 alpha level (5.99). Therefore, the null hypotheses cannot be rejected. There is no significant difference between the distributions. 108 109 TABLE 15.--Significance of Incomplete Data-~Matched. Achievement Levels A B C D X I W n Groups Cumulative Frequency 2 l 2 3 8 L.D. 72 108 108 108 180 180 288 288 288 288 288 288 288 288 36 Diff. 12 52 84 100 76 108* *Greatest degree of difference between distributions. In Table 16 the following null hypothesis was tested: There is no difference in the distribution of grades between those non-matched students with lost anxiety data (L.D.) and those non-matched students with anxiety data (W.D.) The greatest degree of difference found in Table 16 (on the following page) is 20/60 or .334. This value was tested with the Kolmogorov-Smirnov two-tailed two sample test. The computed test value (1.41) was less than the chi-square value with two degrees of freedom at the .05 alpha level (5.99). Therefore, the null hypothesis cannot 110 be rejected. There is no significant difference between the distributions. TABLE 16.--Significance of Incomplete Data-—Non—Matched. Achievement Levels A B C D X I W n Groups Cumulative Frequency 2 l 1 4 L-D 19 11 .6_0 60 60 60 2 l 2 6 2 2 15 M .1 11 11 .2_0 E 11 Q 60 60 60 60 60 60 60 * Diff, .1 11 12. 2_0 11 __7_ _0 60 6O 6O 60 60 60 60 *Greatest degree of difference between distributions. BIBLIOGRAPHY 111 3.. 1... '1“!- ._4_fi I|.l .14., lull- Il.’ Ir BIBLIOGRAPHY Biggs, J. B. Information and Human Learning. Glenview, Illinois: Scott, Foresman and Co., 1968. Campeau, Peggie L. "Selective Review of Literature on Audiovisual Media of Instruction." In Instructional Media: A Procedure for the Design of Multi-Media Instruction, A Critical Review of Research and Suggestions for Future Research. By Leslie L. Briggs, et. a1. Pittsburg: American Institute for Research, 1966. Cattell, Raymond B. "Anxiety and Motivation: Theory and Cricial Experiments." In Anxiety and Behavior, p. 45. Edited by C. D. Speilberger. New York: Academic Press. Chansky, N. M. "Threat, Anxiety, and Reading Behavior." Journal of Educational Research 51 (1958): 333-340. DeLoach, Joseph F. "An Analysis of Cognitive Style Dispar- ity as an Antecedent of Cognitive Dissonance in Instructional Evaluation: An Exploratory Study in the Educational Sciences." Ph.D. dissertation, Wayne State University, 1969. Doty, Barbara A. and Doty, Larry A. "Programed Instruction Effectiveness in Relation to Certain Student Learner Characteristics." Journal of Educational Psychology 55, 6: 334-338. Endler, N. S. and Hunt,J. McV. "Sources of Behavioral Variance as Measured by the S-R Inventory of Anx- iousness." Psychological Bulletin 65 (1966): 336- 346. Feldhusen, J. F. and Klausmeier, H. J. "Anxiety, Intelli- gence, and Achievement in Children of Low, Average, and High Intelligence." Child Development 33 (1962): 403-409. 112 113 Fragale, Marvin J. "A Pilot Study of Cognitive Styles of Selected Faculty Members and Students in a Commu- nity College Setting." Ph.D. dissertation, Wayne State University, 1969. Gaier, R. L. "Selected Personality Variables and the Learning Process." Psychological Monogpaphs 66 (1952): 17. Gardner, R. W., et. al. "Cognitive Styles in Categorizing Behavior." Journal of Personality 22 (1953): 214-233. Gustafson, Kent L. "Simulation of Anxiety Situations and Its Resultant Effect on Anxiety and Classroom Interaction of Student Teachers." Ph.D. disserta- tion, Michigan State University, 1969. Hill, Joseph E. Cognitive Style as an Educational Science. Bloomfield Hills, Michigan: Oakland Community College. , and Kerber, A. Models, Methods, and Analytical Procedures in Education Research. Detroit: Wayne State University Press, 1954. Hoogasian, Vaughn. "An Examination of Cognitive Style Pro- files as Indicators of Performance Associated with a Selected Discipline." Ph.D. dissertation, Wayne State University, 1970. Hovland, C. I., et. a1. Experiments on Mass Communication. v. 3. Princeton, N. J.: Princeton University Press, 1949. Knight, H. R. and Sassenrath, J. M. "Relationship of Achievement Motivation and Test Anxiety to Perform- ance in Programed Instruction." Journal of Educa- tional Psychology 14 (1966): 185-191. Koran, M. L. "Differential Response to Inductive and De- ductive Sequences of Programed Instruction." Journal of Educational Psychology 62, 4: 300-337. . "The Effects of Individual Differences on Ob- servational Learning in the Acquisition of a Teach- ing Skill." Ph.D. dissertation, Stanford University, 1969. . ll \lln ll. Illll: Ill 114 Kritzeck, J. "Philosophers of Anxiety." Commonweal 63 (1955/56): 573. Lange, Crystal M. "A Study of the Effects on Learning of Matching the Cognitive Styles of Students and In- structors in Nursing Education." Ph.D. disserta- tion, Michigan State Universtiy, 1972. McClelland, D. C., et. al. The Achievement Motive. New York: Appleton-Century-Crofts, 1953. McKinney, Fred. "Emotional Factors in Learning." Journal of General Psychology 9 (1933): 101-116. Meyers, Joel and Dunham, J. "Effects of Anxiety on Apti- tude by Treatment Interactions." Paper presented at the American Educational Research Association, February, 1971. O'Neil, Harold F., Jr. Effects of State Anxiety and Pro- graming Variables on the Computer-Assisted Learn- ing of College Students. Final Report. Talla- hassee, Fla.: Florida State University, 1970. Palermo, D. S., et. al. "The Relationship of Anxiety in Children to Performance in a Complex Learning Task." Child Development 27 (1956): 333-337. Phillips, Beeman N. Anxiety and School Related Interven- tions: A Selective Review and Synthesis of the Psychological Literature. New York: New York State Education Department, 1971. Ripple, Richard E., et. a1. "Learner Characteristics and Instructional Mode: A Search for Disordinal Inter- actions." Journal of Educational Psychology 60 (1969): 113-120. Sarason, S. B., et. al. "A Longitudinal Study of the Rela- tion of Test Anxiety to Performance on Intelligence and Achievement Tests." Child Development Mono- graphs 29 (1964): 1-51. . Anxiety in Elementary School Children. New York: Wiley, 1960. . "Empirical Findings and Theoretical Problems in the Use of Anxiety Scales." Psychological Bulletin 57 (1960): 403-415. lilil li1|ll llllllll‘ll ‘1 l 115 Schroeder, Arlen V. "A Study of the Relationship Between Student and Teacher Cognitive Styles and Student Derived Teacher Evaluations." Ph.D. dissertation, Wayne State University, 1969. Shuert, Keith L. "A Study to Determine Whether a Selected Type of Cognitive Style Predisposes One to Do Well in Mathematics." Ph.D. dissertation, Wayne State University, 1970. Sieber, J. E. "A Paradigm for Experimental Modification of the Effects of Test Anxiety on Cognitive Proc— esses." American Educational Research Journal 6 (1969): 46-62. Solomon, G. "Interaction of Communication Medium and Two Procedures of Training for Subjective FEsponse Un- certainty Of Teachers." Ph.D. dissertation, Stan- ford University, 1968. Spence, J. T. and Spence, K. W. "The Motivational Compon- ents of Manifest Anxiety: Drive and Drive Stimuli." In Anxiety and Behavior. Edited by C. D. Spiel- berger. New York: Academic Press, 1966. . The Effects of Anxiety on Computer-Assisted Learning. Tallahassee, Fla.: Florida State Uni- versity, 1970. , et. a1. STAI Manual. Palo Alto, Calif.: Con- sulting Psychologists Press, Inc., 1970. Sutter, E. G. and Reid, J. B. "Learner Variables and Interpersonal Conditions in Computer-Assisted In- struction." Journal of Educational Psychology 60 3: 153-157. Symonds, P. M. What Education Has to Learn from Psychology. New York: Teachers College, Columbia University, 1958. Taylor, J. A. "Drive Theory and Manifest Anxiety." Psy- chological Bulletin 53 (1956): 303-320. Tobias, Sigmund. The Effect of Seqpence and Familiarity with Supject Matter in Achievement from Programed Instruction. Final Report. New York: City Uni- versity, 1971. 116 , and Abramson, T. "The Relationship of Anxiety, Response Mode, and Content Difficulty to Achieve- ment in Programed Instruction." Journal of Educa- tional Psychology. 1971. Toffler, Alvin. Future Shock. New York: Random House, 1970. Laurence. Are Marks Larks? Bloomfield Hills, Mich.: Oakland Community College Press, 1971. The Educational Science of Cognitive Style: Bloomfield Hills,Mich.: Oakland 1971. Wasser, An Introduction. Community College Press, "An Investigation into Cognitive Style as a Ph.D. Facet of Teachers' Systems of Appraisal." dissertation, University of Michigan, 1969. Instructional Tech- Charles. Harper & Wittich, Paul and Schuller, nology: Its Nature and Use. Rowe, Publishers, 1973. "A Pilot Study to Analyze Cognitive Style and Teaching Style with Reference to Selected Ph.D. New York: Wyett, Jerry L. Strata of the Defined Educational Sciences." 1969. dissertation, Wayne State University, Zukerman, M. and Buss, A. "Perceptual Defense and Pre- cognition Responsivity in Relation to Hostility, Anxiety, and Impulsiveness." Journal of Clinical 45-50. Psychology 16 (1960): "Test Anxiety and Intelligence Test Per- Journal of Consulting Psychology 20 Zwiebelson, I. formance." (1956): 479-481. General References "Effect of Three Conditions of Administra- Allen, George J. tion on 'Trait and State' Measures of Anxiety." Journal of Consulting and Clinical Psychology 34 355-359. Personality, A Psychological Interpre- Henry Holt and Co., 1937. (1970): Allport, Gordon W. New York: tation. Princeton, An Introduction to Motivation. Atkinson, J. W. ' N. J.: Van Nostrand, 1964. 117 and Reitman, W. R. "Performance as a Function of Motive Strength and Expectancy of Goal-Attainment." Journal of Abnormal and Social Psychology 53 (1956): 361-366. Barron, F. Creative Person and Creative Process. New York: Holt, 1969. Brookover, Wilber B., et. a1. "Self-Concept of Ability and School Achievement." Sociology of Education 37 (1964): 271-278. Brown, J., et. a1. Applied Psychology. New York: Mac- millan, 1966. Combs, A. W. "The Self in Chaos: A Review of Sylie, R. C., The Self Concept: A Critical Survey of Pertinent Research Literature." Contemporary Psyphology 7 (1962): 52-54. Coller, Allen R. The Assessment of Self-Concept in Early Childhood Education. Urbana, Ill.: ERIC Clear- inghouse on Early Childhood Education, 1971. Coopersmith, R. S. "Conformity and Creative Thinking." In Contemporary Approaches to Creative Thinking by H. Gruder, et. al. New York: Atherton, 1962. Cotter, Jude T. "The Effects of the Educational Science of Cultural Determinants of the Meanings of Symbols on Curricular Choice." Ph.D. dissertation, Wayne State University, 1970. Cronbach, L. J. "The Two Disciplines of Scientific Psy- chology." American Psyghologist 12 (1957): 671- 684. Davidson, K. S. "Interviews of Parents of High-Anxious and Low-Anxious Children." Child Development 30 (1959): 341-351. DeCecco, John P. The Psychology of Learning and Instruc- tion: Educational Psychology. Englewood Cliffs, N. J.: Prentice-Hall, Inc., 1968. DeLoach, Joseph F. Cognitive Styles and Cognitive Disso- nance. Bloomfield Hills, Mich.: Oakland Community College Press, 1970. 118 Denny, J. P. "Effects of Anxiety and Intelligence on Con- cept Formation." Journal of Experimental Psychology 72 (1966): 596-602. Diethelm, Oscar and Jones, M. R. "Influences of Anxiety on Attention, Learning, Retention and Thinking." Archives of Neurology and Psychiatry 58 (1947): 325-336. Felker, Donald W. and Treffinger, Donald J. Self Concept, Divergent Thinking Abilities, and Attitudes about Creativity and Problem Solving. Lafayette, Ind.: Purdue University, 1971. Flanagan, John C. "General Considerations in the Selection of Test Items and a Short Method of Estimating the Product-Moment from the Data at the Tails of the Distribution." Journal of Educational Psychology 32 (1939): 674-680. Forgus, R. H. and Schwartz, R. J. "Efficient Retention and Transfer as Affected by Learning Method." Journal of Psychology 43 (1957): 135-139. Freud, S. An Outline of Psychoanalysis. New York: Norton, 1949. . The Problem of Anxiety. New York: Norton, 1936. Gagne, Robert M. "The Acquisition of Knowledge." Psycho- logical Review 69 (1962): 355—365. Gardner, R. W., et. a1. "Cognitive Control: A Study of Individual Consistencies in Cognitive Behavior." Psychological Issues 1: 4, New York: International University Press, 1959. Gustad, John W. "On Improving College Teaching." NEA Journal 53 (1964): 37-38. Hand, James D. "Matching Programed Instruction Packages and an Instructional Setting to Students, In Terms of Cognitive Style: An Exploratory Study." Ph.D. dissertation, Michigan State University, 1972. Haskell, Roger W. "Effect of Certain Individual Learner Personality Differences on Instructional Methods." AV Communication Review 19 (1971): 287. 119 Hebb, D. O. "Drives and the C.N.S. (Conceptual Nervous System)." Psychological Review 62 (1955): 243- 245. Hill, Joseph E. The Education Sciences. Bloomfield Hills, Mich.: Oakland Community College, 1972. . Symbols and Their Meanings. Bloomfield Hills, Mich.: Oakland Community College Press. . Qualitative Symbols. Bloomfield Hills, Mich.: Oakland Community College Press. Hogan, R. A. "The Development of a Measure of Client Defensiveness in the Counseling Relationship." Journal of Consulting Psychology (1952): 417-424. Hunt, J. McV. "Intrinsic Motivation: Information and Circumstances." In Perponality Theory and Informa- tion Processing. Edited by H. Schroeder and Peter Suefeld. New York: The Ronald Press Co., 1941. Husband, David W. "Analysis of Certain Components of the Audio-Tutorial System of Teaching." Ph.D. disser- tation, Purdue Universtiy, 1970. Johnson, Charles D. Personality Attributes and Preferences for Group vs. Individual Work Situations. Mil- waukee: Wisconsin University, 1971. Kagan, Norman, et. a1. Studies in Human Interaction. East Lansing: Michigan State University, 1967. Klausmeier, Herbert J. and Goodwin, W. Learning and Human Abilities. New York: Harper & Rowe, Publishers, 1966. Klein, G. S. "The Personal World Through Perception." In Perception: An Approach to Personality. Edited by R. R. Blake. New York: Ronald Press, 1951. Koran, M. L. "Varying Instructional Methods to Fit Trainee Characteristics." AV Communication Review 20 (1972): 135. Marshall, H. H. "Learning as a Function of Task Interest Reinforcement, and Social Class Variables." Journal of Educational Psychology 60 (1969): 133-137. 120 Merrill, M. David and Goodman, R. I. Selecting Instruc- tional Strategies and Media: A Place to Begin. Brigham Young University, National Special Media Institutes, 1972. Meumann, Ernst. The Psychology of Learning. New York: D. Appleton and Co., 1913. Ort, Barbara. "An Examination of Relationships Between the Measurable Cognitive Characteristics of a French I Teacher and the Student's Success in That Course." Ph.D. dissertation, Michigan State University, 1971. Poling, E. G. Videotape Recordings in Counseling Practicum. University of South Dakota, 1965. Purkey, William W. The Development of a Pupil Behavior Inventory to Infer Self Concept. Gainesville, Fla.: 1971. Roth, Theodore C. "Expanding the Concept of Individualized Education." The Educational Forum 36 (1971): 61-62. Schroeder, Arlen V. Cognitive Style: The Key to Student Placement? Bloomfield Hills, Mich.: Oakland Commu- nity College Press, 1971. Snow, Richard E. "Research on Media and Aptitudes." In "Commentaries on Research in Instructional Media: An Examination of Conceptual Schemes." Viewpoints (Bulletin of the School of Education). Edited by G. Soloman and R. Snow. Indiana University, 1970. Spence, J. T. and Spence, K. W. "A Theory of Emotionally Based Drive (D) and Its Relationship to Performance in Simple Learning Situations." American Psycholo- gist 13 (1958): 131-141. Thelen, Herbert A. Classroom Grouping for Teachability. New York: John Wiley & Sons, 1967. Thorndike, E. L. The Fundamentals of Learning. New York: Teachers College, Columbia University, 1932. Traweek, Melvin W. "The Relationship Between Certain Per- sonality Variables and Achievement Through Pro- gramed Instruction." California Journal of Edu- cational Research 15 (1964): 215-220. 121 Van De Riet, H. "Effects of Praise and Reproof on Paired Associate Learning in Educationally Retarded Child- ren." Journal of Educational Psychology 55 (1964): 139-143. Weisberg, P. S. and Springer, K. J. "Environmental Factors in Creative Function." Archives of General Psy- chiatry 5 (1961): 554-561. Witkin, H. A., et. a1. Personality Through Perception. New York: Harper and Brothers Pub., 1954. Yerkes, R. M. and Dodson, J. D. "The Relative Strength of Stimulus To Rapidity of Habit Formation." Journal of Comparative Neurology and Psychology 10 (1968): 459. Zussman, Steven P. "A Pilot Study Exploration of Cogni- tive Styles and Administrative Style as Defined in the Education Sciences." Ph.D. dissertation, Wayne State University, 1968.