University of Southern California Dissertations and Theses

The role of cues in the arousal of anxiety

(USC Thesis Other)

The role of cues in the arousal of anxiety

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE ROLE OF CUES IN THE AROUSAL OF ANXIETY by Tamar Ilanit A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Psychology) August 1959 UNIVERSITY OF SOUTHERN CALIFORNIA GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES 7, CALIFORNIA This dissertation, written by ---- ------- -- ---, __ -·-·-----. T.amar. .. llanlt ____ - ·----. ----· ·----- ---- under the direction of har.. ... Dissertation Com mittee, and approved by all its membe,·s, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of DOCTOR OF PHILOSOPHY Date ............. J.une ... 29., .... l.95.9 ..................... . ACKNOWLEDGMENT Foremost I want to take this opportunity to express my gratitude to my Chairman, Dr. Alfred Jacobs, for his per sistent and unceasing support. Especially for retaining his good humor throughout many crises, both real and imagined. I also want to thank Dr. Grings for his much needed technical assistance; Dr. Klein for his patient smoothing out of the rough edges, and Dr. Locke for his ready co-oper ation, despite being .on sabbatical leave this year. Robert Williamson, .Samuel Harman, Paul Fischer, Robert Ruhl, and Charles Locks, of the Los Angeles City College, gave up valuable class time so that I could gather the data for this study. To them I am sincerely grateful. To my cousin, Jacqueline Berman, who took time to read the manuscript and offer many helpful suggestions, my special thanks. Finally, I want to thank my two co-workers, my friend Zena Malek and my husband, Nathan Ilanit. In particular, my deepest appreciation to my husband for his patience with me throughout many hours of insistent requests for help. ii TABLE OF CONTENTS ACKNOWLEDGMENT Chapter I. The Problem Anxiety as a Drive Anxiety Arousal Anxiety and Performance Anxiety and Cues The Aim of the Study Hypothesis One Hypothesis Two Hypothesis Three Hypothesis Four II. Background of the Problem Anxiety and Drive The Relation of Drive to Performance The Relationship Between Anxiety and Intelligence Cues Transfer of Training Primary Generalization and Discrimination Cues and Language Cues in Human Learning and Performance III. Procedure The Subjects The Controlled Variables The Measure of Anxiety The Measure of Perceptual Speed The Measure of Performance Procedure IV. Results . . . . . . . . . . . . . . . . . V. Hypothesis One Hypothesis Two Hypothesis Three Hypothesis Four Discussion ..... Hypothesis One Hypothesis Two iii Page ii 1 12 37 48 64 Chapter Hypothesis Three The Level to Which Anxiety Has Been Aroused The Experimental Variables--the Cues Hypothesis Four Suggestions for Further Research VI. Summary and Conclusions Purpose of the Study Method Results Conclusions REFERENCES APPENDIXES Appendix A: Statistical Data Appendix B: Test Materials iv Page 73 79 89 123 LIST OF TABLES Table Page 1. Original Mean Scores on Anxiety Test Question naire, Visual Speed and Accuracy, and Symbol Symbol Substitution Tests . . . . . . . 49 2. Correlations Among Variables . . . . . . . 51 3. Comparisons of Mean Scores of SSST of Matched Groups . . . . . . . . . . . . . . . . . . . 53 4. Comparisons of Mean Scores of SSST of Matched Groups: Both Semesters Combined . . . . . . 54 5. Comparisons of Mean Scores of SSST of Matched Groups: Exp. A vs. Exp. B. . . . . . . . . 60 6. Comparisons of Mean Scores of SSST of Matched Groups: The Control Group Versus Each of the Four Experimental Groups Tested in the Second Semester . . . . . . . . . . 62 7. Raw Scores of the First Class on the: Sarason Anxiety Test Questionnaire, SSST, and Per ceptual-Speed Test, From Which Pairs Were Matched, Clerk (T)*, Same . . . . . . . 89 8. Raw Scores of the Second Class ... , Clerk (T)*, Different, (Z)* • • . . . 90 9. Raw Scores of the Third Class ( T) -x-, Same . . . . . . . . ., Professor 10. Raw Scores of the Fourth Class ... , Professor 91 (T)I, Different (Z)* . . . . . • . . . . . . 92 11. 12. Raw Scores of the Fifth Class Raw Scores of the Sixth Class Same . . . . . . . • . ., Control 93 . , Clerk ( Z) -x-, . . . . . . . . . 94 13. Raw Scores of the Seventh Class .•. , Clerk (Z)*, Different (T)* . . . . . . . . . . . . 95 14. Raw Scores of the Eighth Class ... , Professor ( z) -x-, Same . . . . . . • . • . . . . . . 96 15. Raw Scores of the Ninth Class (Z)*, Different (T)* ... V ., Professor 97 Table 16. Raw Scores of the Tenth Class . . . , Control Group ...... . . . 17. Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual Speed Scores: Clerk, Same, vs. Professor, vi Page Different, Distribution A . . . 99 18. Perceptual-Speed ... Scores: Clerk, Same, vs. Professor, Same, Distribution B . . . . 100 19. Perceptual-Speed ... Scores: Clerk, Differ- ent, vs. Clerk, Same, Distribution C . . 101 20. Perceptual-Speed .. . Scores: Clerk, Same, vs. Clerk, Same ... 21. Perceptual-Speed ... Scores: Over-all Z* (Clerk, Same and Professor, Same), vs. Over all T* (Clerk, Same and Professor, Same), 102 Distribution E. . . . . . . . . . . . . 103 22. Perceptual-Speed ... Scores: Clerk, Differ- ent, vs. Professor, Different, Distribution F . . . . . . . . . . . . . . . . . l 05 23. Perceptual-Speed ... Scores: Professor, Same, vs. Clerk, Different, Distribution G . . 106 24. Perceptual-Speed ... Scores: Professor, Same, vs. Professor, Same, Distribution H . . . . 107 25. Perceptual-Speed ... Scores: Professor, Same, vs. Professor, Different, Distribution I . . 108 26. Perceptual-Speed ... Scores: Professor, Same, vs. Clerk, Different, Distribution U . . 109 27. Perceptual-Speed ... Scores: Clerk, Same, vs. Professor, Same, Distribution V 110 28. Perceptual-Speed ... Scores: Clerk, Same, vs. Clerk, Different, Distribution W. . 111 29. Perceptual-Speed ... Scores: Clerk, Differ- ent, vs. Professor, Different, Distribution X . . . . . . . . . . . . . 112 Table 30. Perceptual-Speed ... Scores: Professor, Same, vs. Professor, Difrerent, Distribu- vii Page tion Y . . . . . . . . . . . . . . . . . 113 31. Perceptual-Speed ... Scores: Clerk, Same, vs. Professor, Different, Distribution Z 114 32. Perceptual-Speed ... Scores: Control, vs. Clerk, Same* . . . . . . . . . . . . 115 33. Perceptual-Speed .. . Scores: Control, vs. Clerk, Different* 34. Perceptual-Speed ... Scores: Control, vs. Professor, Same*. . . . . ..... 35. Perceptual-Speed ... Scores: Control, vs. Professor, Different* ........ . 36. Raw Scores Where the Median SSST Fell . 116 117 118 119 LIST OF ILLUSTRATIONS Figure 1. Composite Curve of Sensory Generalization . 2. 3. First Semester Testing, Exp. A and B Second Semester Testing, Exp.Band A . Graph 1. Correlation Between Anxiety Test Scores and Page 26 39 39 SSST . . . . . . . . . . . . . . . . . . . . 120 2. Correlation Between SSST and Perceptual-Speed Scores, Control Group . . . . . . . . . . . 121 viii CHAPTER I The Problem Much attention has been devoted in recent years to the concept of anxiety. Personality theorists beginning with Freud (1936), later the so-called neo-Freudians (Horney, 1945; Rank, 1936; Sullivan, 1947), a~d currently authors such as McClelland (1951), have attributed a key position to anxiety in their formulations. Anxiety is an important con cept for such theorists because they have regarded it as a significant etiological factor in the development of psycho logical symptomatology. For example, Freud (1936) stated that symptoms develop as an unconscious means of repressing anxiety. Similarly, Dollard and Miller (1950) suggest that symptoms develop as a way of dealing with anxiety, and that such symptoms as phobias comprise strong habits which have been reinforced because these symptoms were successful in avoiding anxiety. According to Dollard and Miller, the de velopment of such symptoms can be understood in terms of the relationship between anxiety, cues, and habits or perform ance. Broadly speaking, the present study is concerned with certain relationships between anxiety, cues, and performance. Anxiety as a Drive Dollard and Miller (1950), as well as many other pres ent-day theorists, conceive of anxiety as a drive. Miller 1 (1948) has demonstrated experimentally that anxiety reduc tion is reinforcing like the reduction of any other drive. Consequently, the laws of learning which apply to other drives apply to anxiety too. Anxiety Arousal 2 Some attempt to put order in the vast and often contra dictory field of anxiety has taken the course of fractionat ing anxiety into specific varieties and linking them to par ticular tasks. An example is the Sarason Test Anxiety Ques tionnaire, which restricts its range of concern to anxiety over test-taking (Mandler and Sarason, 1952) and is not highly related to measures of general anxiety (Mandler, Gardner and Crouch, 1957; Sarason, 1957), and shows signs of predicting performance on academic tests better than general anxiety measures (Martin and McGowan, 1955; Sarason, 1957). Anxiety is frequently aroused by direct means, such as by applying a noxious stimulus to the organism. This is a method easily applied to animals but usually not welcomed by human subjects, so that other means of generating anxiety are necessary. For human subjects words are often substi tuted for actions and special threatening instructions, varying in content but with a common aim, have been used to arouse anxiety. In his chapter on Motivation, Underwood (1949) discusses Ego-involvement as one aspect of the topic of motivation. In that connection he says that "our present 3 problem is to examine ways in which E(experimenter) can vary motivation. Making S(subjects) ego-involved is presumably one of these ways 11 (1949, p. 192). He further says that this conception which an individual has of his role or status will be called the ego. When an individual is said to be ego-involved experimentally we mean that he is placed in a situation which is a threat to his status or role; it is a.threat to his prestige or self-esteem. These threats instigate feelings of anxiety. Anxiety, be ing a painful condition, seems as a motivating factor [Mowrer, 1939, 1940] by energizing the organism. (Under wood, 1949, pp. 192-193) The condition of reading threatening instructions to students is frequently labeled 11 stress si tuation 11 (Berry and Martin, 1957; Farber and Spence, 1956; Schlaff, 1957) prob ably to distinguish the anxiety aroused by the instructions from the original level of anxiety of the subjects. Al though the method of arousing anxiety by threatening in structions is not always successful (Farber and Spence, 1956), it is on most occasions. Eriksen (1957), reviewing recent studies on anxiety and stress, says in this connec tion: There has been little progress in sorting out the differ ent factors in stress and its effects upon performance. While the different kinds of stress and methods of induc tion may all be assumed to produce anxiety, it is likely that the nature of the performance effects would differ depending, for example, upon whether the subject was threatened with electric shocks while actually performing the task or whether the stress was produced by circum stances entirely unrelated to the experimental situation. (1957, p. 97) Anxiety and Performance According to the original hypothesis of Hull (1943), 4 who influenced Dollard and Miller, among other learning the orists, performance is a function of the interaction between the existing habits and the level of drive at any particular time, or: D x sHr = sEr. More specifically, the drive level interacts with the existing habits to determine the proba bility of the occurrence of a particular response. Since this probability is a multiplicative function of drive level and habit, it is clear that according to this formulation the higher the drive-level the more likely is the response to occur. Since the number of times a certain response oc curs is often used as a means of measuring performance, e.g., number of bar-pressings in an instrumental conditioning ex periment, then it can be said that the higher the D level the 11 higher 11 is the performance. When anxiety is the drive in question then the higher the anxiety level--the higher the performance. Taylor (1951) has experimentally demon strated that when the habit of eyelid response was condi tioned, that high-anxiety subjects conditioned better than low-anxiety subjects, as predicted by the above formula. Subsequent studies (Ramond, 1953; Stevenson and Iscoe, 1956) involving repetition of the Taylor experiment, using habits other than eyelid response, frequently obtained dif ferent results. These later studies often showed that high anxiety subjects performed worse~ rather than better, than low-anxiety subjects. The concept of family-habit hierarchy was then invoked to explain the apparent contradiction. For 5 every situation there is usually a group of related habits that have been developed to deal with that situation, re ferred to as a habit-family. In this habit-family some hab its are stronger than others, so that in effect, a hierarchy of strengths exists. If in a given situation the correct or appropriate habit is the strongest one, then increasing the drive-level will cause an increase in performance, as stated in the formula of D x sHr = sEr. If, however, an incorrect habit is stronger, then it is obvious that increasing the drive level will cause an increase in the incorrect perform ance, at least until the time when the correct habit is learned and becomes stronger. Therefore, in situations in volving performance of a complex task, increasing the drive level ought in many instances result in increase of incor rect performance. When anxiety is the drive involved then raising the level of anxiety in a situation requiring the performance of a complex task is more likely to have a dis ruptive effect (increase incorrect responses) than a con structive effect. It is not surprising, then, that differ ent studies (Cox, 1956; Davidson and Andrews, 1956; Farber and Spence, 1953, 1956; Fortier, 1956; Schultz and Calvin, 1955) have shown that the performance of high-anxiety sub jects is poorer, better, more variable, or is more unpredict able, than the performance of low-anxiety subjects. The most common experimental finding has been, as stated before, that high-anxiety subjects perform less efficiently than 6 low-anxiety subjects, probably because the competition of several habits is often involved in the task used to measure performance. Anxiety and Cues There are many kinds of cues, varying along different dimensions. A cue may be inherent in the stimulus, like color or size, or it may be extrinsic to it, like a verbal label given to the stimulus. Applying the same verbal label to different stimuli gives them a certain learned equivalence, like calling two different pieces of metal "money." On the other hand, applying different labels to similar stimuli gives them a certain learned distinction, like calling one shiny coin a "nickel" and a similar one "candy money." Every situation is made up of many stimuli, and hence of a large variety of cues. Two situations can vary both in terms of the number of cues present, like a bare room versus a room full of objects, and the kind of cues, like a room full of furniture versus a room full of people. The Aim of the Study The present study aims to investigate two aspects of the relationship between anxiety, cues, and performance: A. The relationship between the disruptive. effects of anxiety and the number of anxiety cues present in a situation, and B. The relationship between the disruptive effects of 7 anxiety and the kind of anxiety cues present in a situation. As was discussed earlier, anxiety can be assumed to be aroused by reading threatening instructions intended to make subjects ego-involved. In the present study anxiety will be presumed to have been aroused by reading such instructions. Clearly, this anxiety is only one kind of anxiety, which is defined by the particular conditions of this experiment. Specifically, anxiety here is defined in terms of of the presence of a certain condition (threatening instructions) and the occurrence of impaired performance on a specific test. Schlaff (1957), who also aroused anxiety by instruc tions, found differences between subjects whose anxiety had been aroused and those whose anxiety had not been aroused .• using a learning task involving a symbol-symbol substitution test, adapted from the Wechsler digit-symbol subtest. Since the present study will be carried out in a situation almost identical to that of Schlaff's, and use the same learning task that he used, the disruptive effects of anxiety on the performance of a symbol-symbol substitution test (SSST) will be taken for granted. At the end of the study, however, this assumption, that anxiety has been aroused by the in structions, will be tested. Ordinarily the effects of anxiety, usually disruptive effects if the task is a complex learning task, are measured immediately after anxiety is aroused and no change is made 8 in the situation. In this case, then, all cues which were associated with anxiety-arousal, such as the object which generated the anxiety, the physical surroundings, etc., will remain unchanged in the situation. If, however, certain as pects of the situation are changed after anxiety is aroused but before the effects of it are measured, then one might expect that this change in the number of anxiety-cues will cause a difference in the effects of anxiety. That is, since some of the cues associated with anxiety are removed, the disruptive effects of anxiety will be lessened. Accord ingly, four hypotheses have been selected for investigation. These will be considered in order. Hypothesis One The disruptive effects of anxiety (as inferred from differences in performance in an SSST udder two experimental conditions) will be lessened if certain cues, associated with the arousal of anxiety, are removed before the effects of anxiety are measured. Therefore, if anxiety is aroused by verbal instructions by a certain person, its effects will be reduced if this person is removed from the situation, thus removing some anxiety-cues, and the effects of anxiety measured by another person. As stated before, applying different labels to similar stimuli gives them a certain learned distinction. The two 9 different labels may arouse different sets of associations and thus have different effects in similar situations. This study is particularly interested in comparing the effects of the label 11 professor" versus the label II clerk" in an anxiety situation involving college students. Common sense experi ence suggests that students might be more attentive and more anxious when a test is administered to them by a "professor 11 than by a "clerk." Accordingly, Hy:i;othesis Two Statements aimed at arousing anxiety in students in a testing situation will be more effective in arousing anxiety (as inferred from the differences in performance in an SSST under two experimental conditions) when read by a person labeled 11 professor 11 than when identical statements are read by a person labeled 11 clerk. 11 Therefore, if anxiety is aroused in relation to stu dents' school performance, this anxiety will be greater if aroused by a person labeled 11 professor of psychology making an evaluation 11 than if aroused by the same person labeled as 11 a clerk aiding in the evaluation. 11 A combination of the two preceding variables should be more effective in increasing or decreasing anxiety in a sit uation than either variable alone. 1I'hat is, the label "pro fessor" combined with a situation where all anxiety-cues re main unchanged should be more effective in arousing anxiety 10 than the label 11 clerk 11 combined with a situation where some anxiety-cues are removed. Accordingly, Hypothesis Three The disruptive effects of anxiety (as inferred from differences in performance in an SSST under two experimental conditions) on students in a testing situat_on should be greater if aroused and measured by the same person labeled 11 professor 11 than if aroused by a person labeled 11 clerk 11 and measured by another person altogether. Therefore, if anxiety is aroused by verbal instructions by a person labeled 11 professor of psychology making an eval uation," who also remains to test the effects of' anxiety, then the effects of' the latter will be greater than if' aroused by a person labeled "clerk from the Office of Admis sions 11 and this person is later removed from the situation, and the effects of' anxiety measured by another person. Since Hypotheses One and Three of' thj_s study involve the use of' two experimenters, each in the same two roles, it is interesting to compare the results obtained by the two people in the same role. Studies (Bernstein, 1956; Cleve land, 1951; Eron and Ritter, 1951; Sanders, 1951; Van Kreve len, 1954) using projective tests and conducted under iden tical conditions, except for having different examiners, ob tained different test protocols from subjects under differ ent examiners. Furthermore, some examiners obtained records 11 characterized by specific tendencies, such as hostile rec ords, anxious records, and so forth. One might therefore expect that the two experimenters in the present study may obtain different results under similar conditions. The label "professor" may arouse more anxiety in students when one ex perimenter is the "professor" than when the other is the "professor," or vice versa, and similarly with the label "clerk." Accordingly, Hypothesis Four A certain label may be more effective as an anxiety arousing cue (as inferred from differences of performance in an SSST under two experimental conditions) when applied to one individual than when applied to another individual. Therefore, if two people, each designated by two labels, arouse anxiety in students, then one will be more effective in a given label in arousing anxiety than the other, and vice versa. CHAPTER II Background of the Problem The development of psychopathology is an area in psy chology which is little understood. Very little is known about the etiology of psychopathology in human beings. AnaJ: ysis of theories which attempt to explain the development of symptoms suggests that most, if not all of them, use the concept of anxiety. Freud, as May (1951) points out, was the first to di rect attention to anxiety as the key for the understanding of emotional problems. In one of his last works he indi cated that anxiety is the 11 fundamental phenomenon and the central problem of neurosis" (Freud, 1936, p. 111). His last formulation of anxiety stated that anxiety activates the defense mechanism of repression, that, in order to es cape anxiety which results from a traumatic experience, an individual may subconsciously resort to repressing the anxi ety-arousing event. As a result of this process, Freud held, symptoms may develop. Other personality theorists also emphasized the impor tance of anxiety. One of the criteria differentiating among these theorists was the manner in which they handled the con cept of anxiety. For Rank (1936) anxiety first originated from the birth-trauma and this anxiety served as the proto type for all later anxiety. Later situations which dealt 12 13 with separation were therefore also anxiety-producing. Adler (1940), on the other hand, dealt with this concept in directly, using the term 11 inferiority feelings 11 in much the same way as other writers have used anxiety. Horney (1945), being interested in the social aspect of all behavior, sees "basic a.nxiety 11 as the mainspring of neurotic symptoms. By this term she means 11 the feeling a child has of being iso lated and helpless in a potentially hostile world" (1945, p. 41). Sullivan (1947), following a similar vein of thought, saw anxiety as stemming first from an insecure relationship with the significant persons in childhood and later from a relationship with society at large. Anxiety and Drive In recent years much attention and study have been de voted to the problem of motivation, and the related concept of drive. Some writers have advocated that the terms drive and motivation were synonymous, while others argued that a clear distinction between the two concepts should be made. One of the writers holding the iatter view is Brown (1953). According to him there are three functional properties of drives. The first is the energizing or activating property of drives, that is, "drives function in combination with ex isting reaction tendencies to produce overt behavior" (1953, p. 3). The second property is that under certain special conditions a reduction in the drive will increase the 14 probability of recurrence of the response associated with drive-reduction. The final property is that a failure of a response to r·educe a drive will reduce the probability of the recurrence of that response. Drives, accordingly, do not have the functional property of directing behavior, but rather, they operate to start the machine or to impel behav ior. Brown's position is that 11 every case of directed be havior is to be ascribed not to drives or motives, but to the capacities of stimulus cues, whether innate or acquired, to elicit reactions 11 ( 1953, p. 6) . Mowrer (1950), another learning theorist, realized the wide application that anxiety, if conceived as a drive, could have for learning theory. He was among the first to make explicit comments about the drive properties of anxiety. He conceived anxiety to be a conditioned response which fol lows the anticipation of danger. He was able to show that anxiety reduction is rewarding and is correlated with learn ing, and that: just as a reduction in hunger, reduction in thirst, reduc tion in sex drive, reduction in fatigue, reduction in oxy ~en lack, or reduction in any other organic 11 need 11 or 1 discomfort 11 tends to reinforce the behavior which brings it about (or is at least temporally contiguous with) such reduction or state of 11 relief, 11 so likewise is a reduction in the particular form of discomfort called anxiety is ef fective in fixating behavior that is associated wherewith. (Mowrer, 1950, p. 66) Miller (1948) tested Mowrer's hypothesis that an origi nally neutral stimulus which is presented simultaneously with a noxious stimulus will eventually acquire the ability 15 to arouse a state called fear or anxiety. He shocked rats in a white box, permitting them to escape the shock by flee ing into a black box. The animals continued to run into the black box even when there was no shock administered in the white box. The animals were then prevented from running in to the black box to escape the shock. They had to learn to turn a wheel that would remove the barrier that separated the two boxes. From this experiment Miller concluded that a drive state was, in fact, aroused because the animals were able to learn a new response that was associated with the acquired drive. Maier and his associates (1951) rejected Miller's ex planation and offered an alternative hypothesis in terms of the frustrating effect of blocking a running response. How eve_r, Brown and Jacobs ( 1949) later verified Miller's re sults by an experiment which precluded Maier's explanation. Farber (1948) has shown that anxiety-reduction was r>einforc ing and produced fixated behavior in animals as measured by a resistance to extinction. Others (Farber and Spence, 1953; Maltzman and Morrisett, 1953) have shown a similar relation of anxiety to resistance to extinction in human subjects. It appears now that sufficient evidence exists to justi fy conceptualizing anxiety as a drive, according to the above definition of drive given by Mowrer. 16 The Relation of Drive to Performance Hull (1943), a learning theorist who inspired the above stated formulations of Farber, Mowrer, Miller, and others, assumed that the response strength is a multiplicative func tion of drive strength and habit strength (R = f(sHr x D)). Therefore, in simple conditioning, where only one habit is learned, the response, or performance, will be stronger when the drive level is higher. Since drive multiplies indis criminately the habit strengths of all responses in complex situations which involve a hierarchy of competing responses the effect of higher drive level will not be a simple one. It will depend upon both the number of responses in this hierarchy, and the relative strength of the correct response as compared with the strengths of the incorrect responses. Where the correct response is higher to begin with, increas ing the drive level will cause better performance. Where, however, the incorrect response is higher, increasing the drive level will cause performance to be poorer until the correct response is learned. All these relationships have been demonstrated experi mentally. Taylor and Spence (1951; Taylor, 1951) undertook a series of experiments involving the simple task of eyelid conditioning where only one habit was involved. Using groups of high anxiety and low anxiety subjects, they showed that, as anticipated, the anxious people were superior in condi tioning. Most human learning, however, involves complex 17 rather than simple relationships, where more than one habit is involved. While some studies (Sarason, 1956; Schultz and Calvin, 1955) reported, as anticipated, that low anxiety persons perform better than high anxiety people in complex situations, others (Farber and Spence, 1953, 1956) reported what seem to be contradictory and confusing results. For example, Fortier (1956) found that anxious subjects per formed better, rather than poorer, than non-anxious subjects under stress conditions in difficult serial learning. David son, Andrews, and Ross ( 1956) reported that 11 manifest anxie ty does not produce differential effects on performance," when the task was continuous high speed color naming, while Cox (1956) concluded that "high anxious individuals tend to perform more variably than low anxious subjects on a variety, of tasks." What appears to be arbitrary and unlawful in the rela tionship between anxiety and learning in complex situations may actually mask lawful relationships which are not immedi ately apparent. Farber and Spence (1953) and Taylor and Spence (1954) call attention to the fact that experiments do not always control the number and strength of the competing response tendencies evoked by the heightened drive level. Consequently, it is not surprising that various experiments report different results. Fredenburg (1956) controlled the number and strength of competing responses in a paired-asso ciate learning. He presented subjects with two lists which involved different degrees of competition. List I was de signed to minimize competition among responses and list II was designed to maximize competition among responses. Ac cordingly, he reports that "the high anxiety subjects per formed at a higher level on list I; for list II, however, 18 ... the low anxiety subjects showed a higher level of per formance." Similar results are reported by Taylor and Chap man (1955) and by Spence, Taylor and Ketchel (1956), also using the paired-associate learning task, and by Denny and Reisman (1956) and Maher (1956) using another task, all ex plaining their results in terms of competing response ten dencies. Other authors deal with the relationship between drive level and performance in terms of the task involved. These authors contend that the reason studies differ in results is that anxiety exerts differential effects as the nature of the tasks is varied. Following this line of thought, Sieg man (1956) found that subjects who scored high on the Taylor Anxiety Scale obtained significantly lower scores on the timed than on the untimed subtests of the Wechsler Adult In telligence Scale. He therefore suggested that anxiety has a disruptive effect on tasks like abstraction, incidental learning, and timed intelligence tests. These results are in line with the investigation of Witt (1956) who found that high anxious subjects do poorer in a story recall task but not in a vocabulary task. Interestingly enough, Witt also 19 found that "high anxious persons tend to do poorer in a fail ure testing situation than in a neutral situation, while the low anxious persons tend to perform better in a failure sit uation than in a neutral situation. 11 These findings may be interpreted to mean that adding anxiety (a failure situation) to an already existing high level of anxiety (high anxious people) disrupts performance, while adding anxiety (failure situation) to a low level of anxiety (low anxious people) improves performance. The latter is surely expected on the basis of the hypothesis that sEr = f(D x sHr), while the first finding, that adding anxiety to high drive level makes for poorer performance, may be more difficult to explain. The possibility of there being a curvilinear rather than a linear relationship between performance and anxiety, as seems to be suggested by the above analysis, was tested by Matarazzo and Phillips (1955). Using three rather than two levels of anxiety, they report that the middle anxious groups performed significantly better than the low anxious group, and better, though not significantly so, than the high anxious group. In another study, Matarazzo, Ulett and Saslow (1954) found that when time rather than number of trials was used as the criterion of learning, those persons with medium anxiety performed best to a statistically sig nificant degree when compared with both extreme anxiety groups. In other words, up to a certain point anxiety may facilitate learning but beyond this point it may disrupt it. 20 At the same time these authors caution that it can be predicted from the studies reported so far that further research on anxiety and learning will reveal some tasks for which there will be found no relationship to anxiety level and still others showing various types of relationship including a reversal of the curves found in this study. Related here is a study by Sarason (1956) who was interested in the question of how different instructions would influ ence people with different degrees of anxiety. He found that 11 high motivational instructions were detrimental for high anxiety subjects and facilitating for low and medium groups." One interpretation which he offers is consistent with what was suggested here previously, namely, that anxie ty (or high motivational instructions) up to a point facili tate learning. Since many studies testing the relationship between anxiety and other variables are done with college students, Mandler and Sarason (1952) developed a questionnaire which is specifically intended to measure anxiety in the situation of being tested, rather than to measure general anxiety. Child (1954) has regarded this questionnaire as especially appropriate for verifying hypotheses about the effects of anxiety when the hypotheses predict performance in a situa tion in which the subject is being tested. Martin and Mc Gowan suggest that the Sarason Questionnaire is a measure of general anxiety rather than specific anxiety because in their study ''there was no differential tendency for high as compared with low test anxiety Ss to be more anxious right 21 bef'ore a test than otherwise 11 (1955, p. 468). Even though this may be true, Child's point, stated above, is still ap plicable. In the studies that have been done with the Test Anxie ty Questionnaire (Mandler and Sarason, 1952; Sarason and Mandler, 1952; Sarason, Mandler and Craighill, 1952), the research indicated the superiority of' low anxiety people in novel testing situations. However, they f'ound that high anxiety people did better in college in terms of' grade point average. The authors hypothesize that, in time, the disad vantage that is attributable to anxiety occurring in novel situations may, in fact, be turned into a motivational ad vantage so that the high anxiety individual learns to re spond to his anxiety with task relevant responses. This is particularly interesting in view of the f'ollowing discussion about anxiety and intelligence. The Relationship Between Anxiety and Intelligence In recent years the relationship between intelligence and anxiety has been investigated partly as an attempt to find out whether the inverse relationship, of'ten found be tween anxiety and performance, may be due to an inverse re lationship between anxiety and intelligence. It was re ported earlier that Siegman found that anxiety disrupts per formance on certain subtests of' the WAIS. Negative correla tions between anxiety and various measures of' intelligence 22 have been reported in several recent studies (Calvin, Koons, Bingham and Fink, 1955; Duvall, 1956; Grice, 1955; Matarazzo, Ulett, Guze and Saslow, 1954; Matarazzo, Ulett and Saslow, 1955; McCandless and Castenda, 1956). These studies suggest that the inferior performance on a complex task by high anx iety subjects might be attributed to intellectual differ ences as readily as to differences in motivation. On the other hand, Farber and Spence (1955) state that they have been unable, over a period of years, to find any relation ship between the Anxiety Scale scores of college students and conventional measures of intellectual ability such as entrance examination scores and grade-point averages. (1955, p. 10) Klugh and Bendig (1955), Mayzner, Sersen and Tresselt (1955), Taylor (1955), and Sarason (1956), mentioned earlier, have also reported essentially zero correlations between anxiety and various measures of intelligence. Schultz and Calvin (1955), who also failed to find a relationship between anxi ety and intelligence, suggested that the contradictory find ings cited above might be attributed to variations in selec tion procedures at different institutions. Similarly Spiel berger (1958) in a recent study failed to find a relation ship between anxiety and intelligence except for people who are low in IQ, where a positive relationship with anxiety was found. He calls attention to the possibility, suggested by the distribution of his subjects, that a small negative correlation between anxiety and intelligence may be found if one samples a wide range of intelligence, especially if the 23 sample contains a number of subjects from the lower part of the range. This relationship, he proceeds, is often ob scured in college students because of selection procedures which usually curtail the range of intelligence and make students more homogeneous as far as intelligence quotients are concerned. The suggested interpretation, that a rela tionship between intelligence and anxiety will be found in groups which are more heterogeneous, is yet open to discus sion since in a previous study Dana (1957) controlled the heterogeneity of intelligence and failed to demonstrate the presumed relationship. Cues A cue may be defined as that characteristic of a stimu lus which makes it distinct and different from all other stimuli. It may be a difference in kind, such as between a sound and a color, or a difference in amount, such as be tween two sounds of different intensities. As mentioned be fore, while drive may be regarded as the condition which im pels the organism to respond, cue may be considered as the aspect which provides direction to a response and will de termine when and where the organism will respond, and which response he will make. A single cue may sometimes suffice to determine a response, as the sound of a bell at the end of the hour will determine that the class will be dismissed. Or a pattern of cues together may be necessary, such as the 24 ringing of' a bell, the smell of' smoke, and a shout of' 11 f'ire" will cause the class to disband hurriedly bef'ore the end of' the hour. The ringing of' a bell, the smell of' f'ire, and the shout, are all examples of' cues external to the organism, but cues can also be internal. A drive state stemming f'rom lack of' f'ood will cause a diff'erent response f'rom that caused by a drive state stemming f'rom lack of' water. Ani mals can learn to distinguish between the two inner drive states and use them as cues, as has been shown by Amsel (1949). Cues do not have to be verbalized in order to be ef'f'ective, as attested to by all animal experiments in this area, and by Bouthillet (1948) also f'or human beings. Transf'er of' Training According to McGeoch and Irion (1952), "transf'er of' training occurs whenever the existence of' a previously es tablished habit has an inf'luence upon the acquisition, per f'ormance, or relearning of' a second habit" (1952, p. 299). Transf'er ef'fects may be: (1) positive, when training in one activity f'acilitates the perf'ormance or acquisition of' a second activity, (2) negative~ when the training in one in hibits or retards the learning or perf'ormance of' another, and (3) zero or indeterminate, when training has no observ able inf'luence on the perf'ormance of' a second activity. The huge literature on transf'er of' training is not reviewed here, except f'or the f'ollowing relevant subjects. 25 Primary Generalization and Discrimination ·The simplest form of transfer of training is the phe nomenon of stimulus generalization. The empirical phenome non of stimulus generalization is defined by the observation that a response which has been learned to a particular pat tern of cues tends to occur when a similar pattern of cues is presented. This innate tendency for transfer to occur is usually called primary stimulus generalization. As it is usually measured, stimulus generalization represents posi tive transfer in the sense that the generalized response oc curs without, or with less, prior training. The less simi lar the pattern of cues to the original pattern, the less the tendency for generalization to occur. This variation in transfer is referred to as a gradient of generalization. Figure 1 illustrates the gradient of stimulus generalization for the conditioned galvanic skin response. Similar gradi ents may be obtained in other situations. If a generalized response is not rewarded, the tendency to perform that response is weakened. By rewarding the re sponse to one pattern and not rewarding or punishing the re sponse to a somewhat different pattern of cues, a discrimina tion may gradually be established. The process of discrimi nation tends to correct maladaptive generalizations. It in creases the specificity of the cue-response connection. Thus behavior becomes more specific as the organism learns to make one response to one cue or set of cues, and another 19- 18- 17- 16- 15- 14- 13- 0 12- 11- 0-1---0 _____ 1 _____ 2 _____ 3 _ __, Fig. 1. Composite Curve of Sensory Generalization. (From Hovland, J. Gen. Psychol., 1937, li, 136.) 26 response to another set of cues, only slightly different from the first set. Cues and Language 27 While smell is an important cue for the dog and will indicate if food is near, if the enemy is approaching, or if a female in heat is nearby, for his human master verbal cues assume more importance than sensory ones. The letters YEN OM may signify only a group of nonsense syllables at first, but when spelled backwards they may become a potent cue capable of arousing many ideas. Use of language in gen eral seems to facilitate discrimination among cues, accord ing to the analysis made by Dollard and Miller (1950). They talk about language as a cue-producing response, defining cue-producing response as "one whose main function is to produce a cue that is part of the stimulus pattern leading to another response' 1 (1950, p. 98). One need not necessari ly accept Dollard and Miller's nomenclature in order to ac cept its underlying content. According to them, ''attaching the same cue-producing response to two distinctive stimulus objects gives them a certain learned equivalence, increas ing the extent to which instrumental and emotional responses will generalize from one to another' 1 (1950, p. 101). This is called acquired equivalence of cues or secondary generali zation. Birge (1941) found that the generalization or posi tive transfer of a manipulative response from one stimulus 28 to another stimulus occurred more readily when young chil dren had previously learned to give the same nonsense sylla ble response to the two stimuli for the manipulative re sponse, i.e., overt verbalization facilitated positive trans fer. Working with adults, Gagne and Baker (1950) have re ported that the amount of positive transfer to a motor dis crimination task involving similar color-position cue combi nations is a direct function of the amount of prior experi ence in distinguishing among these cues by means of differ· ent letter labels. Rossman and Goss (1951) studied the fa cilitative effects of the acquisition of discriminative ver bal responses upon the acquisition of motor response to the same stimuli. They used three groups of subjects and each learned a 12 figure syllable paired associate list and an appropriate motor response to the same list of figures alone. They found that the group which has mastered discrimination between pairs of similar f1gures by means of nonsense sylla ble responses acquired discriminative motor responses to the same list of figur'es significantly more rapidly than groups which received only one and four verbal learning trials. On the other hand, 11 attaching distinctive cue-producing responses to similar stimulus objects tends to increase their distinctiveness" (Dollard and Miller, 1951, p. 101). Each of the previously similar objects will have its own di~ tinctive associated (or generalized) set of unique character istics after a distinct label has been attached to it. For 29 example, if one person is labeled "professor" he would tend to arouse a different attitude in the mind of a student than that aroused by the same person labeled as 11 clerk." Al though no direct experimental evidence bearing on this point is available, one may surmise that the label 11 professor 11 would call forth a more serious and respectful attitude than the label 11 clerk. 11 This surmise stems from common sense ex perience and from a group of studies assessing the social prestige value of a group of occupations. According to these studies (Counts, 1925; Deeg and Patterson, 1947; Ruch, 1957), clerical work has less prestige than professional en deavor. Every stimulus situation has many characteristics which can be used as cues. An organism can learn which aspect of the stimulus is the crucial cue, to which a certain response is to be made, or he can learn which response out of several possible ones to make to one cue, or both. This learning, of course, follows the same rules as any other learning. Therefore, a tendency to act, which had been learned in a certain situation, would be associated with many cues present in that situation. When retention of what had been learned is tested, the presence or absence of cues from the original situation will, in part, be related to the amount of reten tion. The relationship between retention and the number of cues present in the testing situation, which had been dupli cated from the learning situation, was demonstrated many 30 years ago by Abernethy (1940). She found better performance when students were tested in the same room where the original learning took place, as compared with poorer retention when tested in a new room. When a new habit has been learned to an adequate degree, then learning of a second habit may be facilitated if certain cues from the old learning situation are present in the new learning situation. Lawrence (1950) trained two groups of animals to respond to a different cue for each group. In a series of test situations he showed that learning of a new instrumental response is faster on a familiar cue than on a non-familiar one. This, of course, has been long known as "transfer of training," mentioned earlier. The transfer effects do not have to be the trans fer of a habit to react, but can be a habit not to react, as in fear and avoidance situations. Amsel and Cole (1949) shocked rats while drinking water and then put them in an other drinking situation where some of the cues from the shock-situation were present. The results were as follows: The greater the similarity of the drinking to the shock sit uation the greater was the drop in water intake. Another example of the transfer or generalization of fear is offered by Bugelski and Woodward (1951). They shocked animals in one situation and then tested the ease of extinction of a habit learned during shock under three conditions, varying in similarity to the original shock situation. The results showed that extinction was most difficult where the 31 similarity to the shock situation was greatest. Cues in Human Learning and Perrormance Up to the present time the major variables in psycho logical testing which were considered as determining ractors or the response were: the test itselr, its speciric in structions, and the examinee's abilities. Recently, much attention has been directed to the role of the situational variables which heretorore have been generally overlooked. One may conceive or these variables as providing different cues ror the examinee, cues which will determine his re sponse. These cues may be overt and obvious, such as the sex of the examiner, or may be internal and presumptive, such as the examinee's attitude. That the examinee brings with him into the testing sit uation certain sets and attitudes which arrect his responses is generally aclmowledged. However, there are only rew ex perimental studies, among them Luchins' (1949), which tackle this elusive problem. Luchins administered the Rorschach to a group or normal subjects and noticed that their responses were quite different from what he considered the established norms. He then re-administered the test to the same group and extended the period or "Testing the Limits" so as to de termine why the original protocols were given. From this inquiry he concluded that "the responses had been influenced by the operation or specific attitudes toward the test and 32 the tester, by previous experiences, and by the educational, occupational, and cultural background" of' the subjects. Weisskopf' and Dieppa (1951) studied experimentally induced f'aking of' the Thematic Apperception Test, the subjects being instructed to make a "good 11 or a 11 bad 11 impression. Compar ing the records obtained under these instructions with those obtained under previous "neutral" instructions, they f'ound that the subjects changed their neutral records more when trying to make a 11 bad 11 impression than when trying to make a "good" impression. From this observation they tentatively conclude that ''individuals taking the TAT may be attempting to make a good impression on the examiner even if' not ex plicitly required to do so'' ( 1951, p. 47 4). Every testee probably has certain attitudes toward testing which will partly determine his pattern of behavior regardless of the particular testing situation he is in. However, some of' his behavior will be modif'ied according to the specific cues af'I'orded by the situation itself. The importance of' the physical cues in the testing environment has been demonstrated by Klatskin (1952), who f'ound that the testing situation (hospital vs. f'actory) aff'ected the Ror schach protocol. Similarly, Phares, Jerry and Rotter (1956) conclude that "such relatively unimportant variables as the physical setting of the testing ... generally have demon strable effects on test results." The sex of the examiner provides another cue which may be related to the elicitation 33 or inhibition of certain responses. While Alden and Benton (1951) failed to find any relationship between the sex of the examiner and the content of Rorschach responses, most authors do find such a relationship. Rabin, Nelson and Clark (1954) showed that a female examiner tends to inhibit the production of sex responses, but not of anatomy re sponses, in male subjects. Comparing male and female exam iners who used similar instructions, Kenny and Bijou (1953) found that the male examiners obtained longer and richer stories on the TAT. They too conclude that "it is apparent that interpersonal stimulations between examiner and subject had differential effects despite efforts to standardize the instructions and the conduct of the examiners. 11 Curtis and Wolf (1951) also report significant differences between male and female examiners in elicitation of sexual responses, but failed to specify what these differences were. Another set of cues is provided by the atmosphere that develops, intentionally or unintentionally, between the ex aminer and the testee. It may be anticipated that cues like a benign and friendly examiner will call forth a different behavior on the part of the testee than a hostile and severe examiner. In a study of the effects of differential treat ment of three groups of subjects Milan (1954) has demon strated that this expectation holds. Subjects treated in a positive friendly manner by the examiner during administra tion of the TAT produced long stories in which the heroes 34 manifested comparatively positive attitudes and low anxiety. Contrary to the hypothesis, subjects treated in a negative hostile manner produced long stories too, but, as was ex pected, manifested comparatively negative attitudes and high anxiety. Similarly, Lord (1950), using three examiners, each treating the testees in three ways, showed that the manner of treatment of the testee, in a negative, positive, or neutral fashion, makes a difference as far as Rorschach performance is concerned. However, in her experiment, an other set of cues, that of the examiner himself, proved to be so much more important to the subject's performance that it overshadowed the effects provided by the differential treatment of the testees. The relationship between testee and examiner in IQ testing, which is a less subjective situation than the pro jective testing mentioned before, is also significant. Sacks (1952) showed gains of 16 IQ points in retesting a group of children with whom she ingratiated herself before, as compared to gains of one or two points when she main tained no previous contact with a similar group of children. The mere presence or absence of an examiner may change the situation enough to make a difference in the testee's ' reactions, although the specific meaning that the testee at- ,tributes to the presence of an examiner is not always known. Stories written to TAT or MAPS test in the absence of the examiner are longer (Van Krevelen, 1954), include more 35 projective material (Sherer, Winne, Page and Lipton, 1952), are generally richer (Bernstein, 1956), and are more humor ous and happy in their tone (Eron and Ritter, 1951). One may tentatively conclude from these studies that the pres ence of an examiner, at least in projective testing, acts as an inhibitory factor. More specifically, the characteristics of each examiner, his outward appearance, his manner, his personality traits which are manifested in testing, all probably provide a unique combination of cues for the testee. Although Berger (1954) concluded from his study of Rorschach testing under different examiners that "the test as a whole reflects only insignificant examiner 1 s influence," the majority of the relevant articles arrive at the opposite conclusion. For example, overtly hostile examiners elicit Rorschach records which suggest that their subjects tend to behave passively, intrapunitively, and in a constrained fashion, while covert hostility on the part of the examiner elicits records which suggest less passivity and stereotypy (Sanders, 1951). Highly anxious examiners, as compared to less anxious ones, elicited more general, emotional, and hostile content on the Rorschach (Cleveland, 1951). The preceding two experiments were concerned with at tempts to relate certain responses of the examinee to spe cific factors in the personality of the examiner. More often than not these procedures were not followed by other 36 experimenters. Nevertheless, it has been shown repeatedly (Baughman, 1951; Gibby, 1952, 1953; Van Krevelen, 1954) that different examiners, with standardized instructions, obtain different Rorschach records from similar examinees. CHAPTER III Procedure Ten classes were used in this study, the first five in the Spring semester and the other five in the Fall semester. Because two experimenters were used for two of the four ex perimental conditions, the procedure of the first semester was duplicated in the second semester, except that the duties of the two experimenters were interchanged, as specified be low. The two experimenters are designated hereafter as Exp. A and Exp. B. Two of the ten classes were used as control groups while the other eight were treated with two experi mental conditions, namely, varying the kind and/or the !:U:ill! ber of experimental cues. The kind of cues was varied by applying the label 11 clerk 11 to the experimenter who read the threatening in structions to the Symbol Symbol Substitution Test (SSST) (used here as the measure of performance), to four classes, and applying the label 11 professor 11 to the experimenter who read the threatening instructions to the SSST to the other four classes. The number of cues was changed by having the person who read the threatening instructions, twice labeled as II clerk" and twice labeled as 11 professor, 11 also administer the SSST to four classes, while having another person admin ister the SSST to the other four classes. Thus, comparing the performance of the classes treated by a 11 clerk 11 versus 37 38 those treated by a "professor" afford the testing of Hypoth esis II, while comparing the performance of the classes which had the same person read the threatening instructions and administer the SSST versus those who had another person administer the SSST afford the testing of Hypothesis I. Figures 2 and 3 represent the experimental design schemati cally. The Subjects The subjects were obtained from 10 evening classes at the Los Angeles City College during two consecutive semes ters. Most of these classes were introductory psychology classes except for the first control group (later elimi nated), and the class assigned to the condition of 11 clerk" all anxiety cues present, which were social psychology classes. In the few cases where there was an overlap of subjects from one class to another these cases were used on ly in their first class and eliminated from the second class. It was assumed that the population was homogeneous in so far as age and economic background were concerned, and that there was essentially no difference between the kind of stu dents in the first and second semesters. The sex of the 'subjects was noted only for purposes of establishing norms for the Sarason scale. The number of subjects used from each class ranged between 12 and 21. Kind of Cues Clerk vs. Professor Number all Clerk Professor all anxiety all anxiety of cues cues present cues present fewer Clerk Professor cues fewer anxiety fewer anxiety cues cues present cues present Fig. 2. First semester testing, Exp. A and B. Kind of Cues Clerk vs. Professor Number all Clerk Professor all anxiety all anxiety of cues cues present cues present fewer Clerk Professor cues fewer anxiety fewer anxiety cues cues present cues Present Fig. 3. Second semester testing, Exp. B and A. 39 40 The Controlled Variables The relationship of two variables to performance on the SSST, original level of anxiety and original level of per ceptual-speed, was investigated. Both of these variables appeared on a priori thinking to be related to performance on the SSST. The Measure of Anxiety To measure the original level of anxiety, the Sarason Test Anxiety Questionnaire was administered with a slight modification, in that those items where specific reference was made to Yale University were changed for wording to agree with the setting in which the present study was done. This test, developed by Mandler and Sarason" ( 1952), purports to measure anxiety in the situation of being tested rather than general anxiety. Child (1954) has pointed out that this questionnaire should be very appropriate for veri fying hypotheses about the effects of anxiety when perform ance in a situation such as an academic one is being inves tigated. Anxiety was aroused by reading threatening instructions (see Appendix B for full text), aimed at arousing the stu dents' anxiety about their own personal preformance as well as the comparative performance of the college which they were attending and presumably identified with. The students were told that they were going to take an intelligence test, 41 that they were to be compared to university students since it had been previously shown that city college students usu ally did not perform as well as university students, and fi nally, that their scores would be kept on file at the school for further reference regarding their continuation at school. The effects of this anxiety were not measured directly, but only as it affected the students' performance on a test of perceptual speed, the SSST. It should be noted that one of t,he four experimental conditions always intervened between the arousal of anxiety and the measurement of performance on the SSST, so that the anxiety aroused by the instructions as such was not measured directly. The Measure of Perceptual Speed This was the Visual Speed and Accuracy test of the Em ployee Aptitude Survey, which is a battery of factorially pure aptitude tests. The Visual Speed and Accuracy test is made up of pairs of numbers or names and the exarninee indi cates whether the pairs are the same or different. The score is the number of correct minus incorrect identifica tions the examinee made within a time-limit of five minutes. According to the manual, this test 11 measures the ability to see small detail quickly and accurately 11 (see Appendix B). The Measure of Performance This was a complex symbol-symbol substitution test, 42 derived from the Wechsler-Bellevue digit symbol test, and developed by Schlaff (1957). Schlaff substituted symbols in place of the digits in the key. These symbols in the key were paired with different symbols in the code. Some changes in the Schlaff Symbol Symbol Substitution Test were made for the present study. While Schlaff labeled his test Symbol Symbol Test, the present test was labeled The Symbol-Symbol Substitution Test (revised from the Wechsler Adult Intelli gence Scale). The other changes were minor and unintention al, due mainly to clerical errors in copying the code fig ures. They involved first a reversal of one figure in the code, as a result of which this figure was eliminated from the scoring. Then, three changes were made in the order of some pairs of figures, so that the number of particular fig ures in the code was unchanged, though the order of three pairs was reversed. Procedure Two or three weeks prlor to the test situation the Sar ason Anxiety Test Questionnaire was administered along with the Visual Speed and Accuracy test to five classes. These tests were administered by a graduate student in psychology of the University of Southern California, while the two ex perimenters, also graduate students in psychology at the same university, did not appear at this time. The subjects were told that a survey was being conducted on how students 43 feel about examinations, that it was a general survey not related to any course in particular, and that two tests would be given to them in this connection, the first of which would be a timed test. They were further told that the results of these tests would be sent to their instructor and they would get them through him (see Appendix B). After reading the instructions, the examiner answered any questions the subjects had regarding that test, and passed out the materials for the Visual Speed and Accuracy test. After the five minutes allotted for this test had passed, the tester collected the papers and distributed the Sarason scale. The subjects were then told that upon com pletion of this test there would be a recess. This testing took about thirty minutes, and was always done in the last half-hour of class time. No reference was made, either by the tester or the instructor, that there would be any furth er testing, and the examiner never returned to these classes. Each class was then assigned to one of the four experi mental conditions or to the control situation, according to the order in which they had been tested. The actual testing was done within one week. Two classes were tested in one evening at the beginning of the week and the other three classes were tested in one evening at the end of the week. For the four experimental groups the threatening instruc tions for the SSST which were intended to arouse anxiety were read to the students by one of the experimenters. 44 In the first class, where the first experimental condi tion of having the experimenter labeled as "clerk" and hav ing all anxiety cues present in the situation, the first ex perimenter (Exp. A) appeared dressed in simple clothes. The class instructor informed the group that the school was par ticipating in a project comparing the abilities of students who attend local colleges. He then introduced the experi menter, adding that she was a statistical clerk employed by that project and was going to administer a test to them. The threatening instructions for the SSST were then read by Exp. A and the SSST administered by the same experimenter (see Appendix B). In the second class, where the second experimental con dition of having the experimenter labeled as 11 clerk 11 and having fewer anxiety cues present in the situation, experi menter A again appeared in simple clothes and the same pro cedure was followed as in the first class, up to the adminis tration of the SSST. Here, after reading the threatening instructions for the SSST, Exp. A left the room. The second experimenter, Exp. B, dressed also in simple clothes, then entered the room and administered the SSST without any in troduction from the instructor, or any further comments to the class. Therefore, at the time the SSST was administered to this class, the departure of the person reading the threatening instructions removed some of the anxiety cues. In the third class, Exp. A appeared dressed formally. 45 Here too the instructor informed the group that the school was participating in a project but introduced the experi menter as Dr. ---J who was a research associate from the University of California at Berkeley} in charge of the proj ect. The threatening instructions for the SSST were then read by Exp. A and the test administered by the same experi menter (see Appendix B). In the fourth class Exp. A again appeared formally dressed. The same procedure was followed as in the third class up to the administration of the test. Then, as in the second class} after reading the threatening instructions, Exp. A left the room and Exp. B, also dressed formally, came in and administered the SSST. Therefore, as in the second class, at the time the SSST was administered to this class, the departure of the person reading the threatening instruc tions removed some of the anxiety cues. Following the testing, each class was informed of the true purpose of the experiment, and of the relation of the previous administration of the perceptual-speed and anxiety tests to the experiment. The classes were further told that the results of their perceptual speed test would be given to them by their respective instructors, at the end of that hour. The experimenter then answered any questions the stu dents had regarding the experiment. While during the actual testing the students appeared very serious and even anxious (one girl left the room after 46 hearing the threatening instructions, saying that she was afraid of IQ testing), when later asked whether they were aware of the true nature of the experiment before they were told about it some students said that they suspected some thing of the kind. However, it was apparent that even these few students were motivated by the instructions. At the end of the testing and discussion the students were asked not to discuss this experiment with students of other classes. In the class used as control, the procedure was slight ly different. The instructor turned the class over to Exp. A with no introduction, except for asking the students• co operation in what was coming ahead. Exp. A then told the students that a new short test was being developed, the op timal time-limit for which was not yet established, that various time-limits were being tried and that was one of them. She then asked the class to do its best in performing the test, and the SSST was then administered. It turned out that this class was rather small, in addition to which sever al students did not co-operate but left the test blank, or talked throughout the administration of the test, so that when these unco-operative students were eliminated too few subjects remained in that class. Therefore, the results of this group were eliminated from the final calculations. The following semester the same procedure was repeated for another group of five classes. One class was again used as a control group, while the other four classes were 47 assigned to the four experimental conditions. Exp. B was substituted for Exp. A, and vice versa. Otherwise, the man ner in which the classes were handled was identical with that of the previous semester, and the control group of this semester was found to be adequate. Local norms were calculated for both the men and the women, although norms have been published (Sarason and Gor don, 1953) for the individual items on the Sarason Test Anx iety Questionnaire (see Appendix A). Each graphic scale was eighteen centimeters in length. For scoring purposes the scale was divided into ten units, each of which was eighteen millimeters long. Raw scores, therefore, ranged from Oto 9. After the raw scores were obtained, the median was calcu lated for each item. The final score on any question was determined by the relation of the raw score to the median: 0 was scored if the raw score fell on or below the median, and was scored 1 if it fell above the median. An individu al's total score was obtained by adding the number of times he scored above the median. Total scores could, therefore, range from Oto 35, but actually ranged from Oto 34. The Visual Speed and Accuracy tests were then scored according to the manual, using the formula of Right minus Wrong. Finally, the SSST papers were scored by counting the number of correct substitutions, eliminating correct or in correct substitutions for the one item in which, as previous ly mentioned, a clerical error had resulted in a reversal. CHAPTER IV Results In order to test the experimental hypotheses, it was necessary to establish that all groups were comparable with respect to perceptual speed and original level of anxiety. Table 1 shows that there were variations among groups in both perceptual speed and anxiety. The mean scores on the Visual Speed and Accuracy test ranged from 79.50 to 98.00, while the mean scores on the Anxiety Test Questionnaire ranged from 12.61 to 32.50. It is apparent from Table 1 that although groups tested within one semester are similar in their mean anxiety scores, that there is a consistent difference between the two semesters. For some reason the groups tested in the first (Spring) semester have higher mean anxiety scores than the groups tested in the second (Fall) semester. One can only speculate as to the probable reason for the difference between the two semesters and won der whether students are more anxious in the Spring semester after their Fall semester's work is already known. The following statistical calculations were undertaken in order to establish whether variations in perceptual-speed and/or original level of anxiety were related to Symbol Sym bol scores: (1) A scatter diagram of the Visual Speed and Accuracy and Symbol Symbol scores in the control group, as shown in Appendix A, indicates that the Visual Speed aDd 48 Table 1 Original Mean Scores on Anxiety Test Questionnaire, Visual Speed and Accuracy, and Symbol Symbol Substitution Tests Anxiety Visual Symbol Class and Conditions Test Speed Symbol Question- and Substitu- naire Accuracy tion Test First: Clerk, all cues 32.50 98.00 55.69 Second: Clerk, fewer cues 31.70 79.50 55.05 Third: Prof'essor, all cues 28.30 89.40 50.45 Fourth: Prof'essor, fewer cues 32.50 88.29 52.10 Fif'th: Control (omitted) 35.00 91.31 50.69 Sixth: Clerk, all cues 18.83 84.04 50.96 Seventh: Clerk, £'ewer cues 17.90 88.60 61.53 Eighth: Prof'essor, all cues 18.70 84.50 55.00 Ninth: Prof'essor, f'ewer cues 12.61 88.78 57.69 Tenth: Control 18.00 82.14 56.29 49 50 Accuracy and Symbol Symbol scores are linearly related. A Pearson Product-moment correlation between these scores was .434. (2) Similar calculations between Visual Speed and Accuracy and Symbol Symbol scores in all groups were under taken, as shown in Table 2. It may be observed from Table 2 that original level of perceptual-speed is indeed related to Symbol Symbol test performance, as shown by correlations as high as .808, .739, and .680. Original level of anxiety, on the other hand, seems to have no obvious relationship to Symbol Symbol scores. The correlations between anxiety and Symbol Symbol scores ranged between .013 and -.361, but none were statistically significant. That there is not a linear relationship between original level of anxiety and Symbol Symbol test performance is further suggested by Graph 1 in Appendix A. On the basis of the above analyses the subjects used in the experiment were matched for perceptual-speed since it was highly related to the criterion variable, Symbol Symbol. Appendix A shows the original distributions of the groups before matching, and those pairs used for matching. Tables 17 to 35 in Appendix A show the final matched groups. It can be seen in Appendix A that the distribution of the groups on the perceptual-speed test were quite different from each other. This meant that if matching were attempted for all five groups of each semester that many subjects whose perceptual-speed scores were either too high or too Table 2 Correlations Among Variables First Semester 1st 2nd 3rd 4th Variables class class class class N - 26 N - 22 N - 33 N - 21 Anxiety - Perceptual-Speed Anxiety -- Symbol Symbol Perceptual-Speed Symbol Symbol Variables Anxiety - Perceptual-Speed Anxiety -- Symbol Symbol Perceptual-Speed Symbol Symbol -.069 -.181 a .680 Second 6th class N - 23 -.237 -,339 .232 .142 Semester 7th class N - 17 -.361 -.123 aSignificant at the .01 level. bSignificant at the .05 level. cThis class was later eliminated. -.059 -.044 a .520 8th class N - 23 -.040 -.138 -.158 .013 9th class N - 23 -.318 -.081 51 5thC class N - 13 -.461 ,534 10th class N - 21, .061 -.117 52 low would be lost in the process. Since the groups were not too large to begin with, in order not to lose too many sub jects, matching was done for pairs of groups, between groups tested in the same semester only. This method of matching meant that, for each comparison tested, pairs of groups had to be matched anew. Certain limitations on the interpretation of the dif ferences obtained between groups are thus imposed by the fact that only a part of each original group was used for the final comparisons. It is, of course, possible that those subjects with very high or very low scores, who were left out because no mate could be found for them, would have heen affected most by the experimental variables and would have contributed most to the differences between groups. Moreover, as can be seen from Tables 3 and 4, the paired-matching meant that different subjects with somewhat different mean scores were pooled out from the same original groups for the various comparisons. Hence, one cannot make an over-all comparison between the groups, but only paired comparisons. Again, this procedure makes for limitations on the generalization of the present results. One can not readily generalize the results of these groups to a general population of students. Acknowledging these limitations, one has to recognize that these particular data called for some control, such as matching on the perceptual-speed scores, before the differences on the SSST could be Table 3 Comparisons of Mean Score~ of SSST of Matched Groups Clerk, all cues Clerk, fewer cues df t Professor, all cues Professor, fewer cues df t All cues, professor All cues, clerk df t Fewer cues, professor Fewer cues, clerk df t Professor, all cues Clerk, fewer cues df t Clerk, all cues Professor, fewer cues df t First Semester 54.43 59.28 13 1.42 53.01 50.71 16 .376 50.67 52.52 20 .440 52.53 58.13 14 1.91a 48.12 55.35 16 1.83a 52.33 51.66 14 1.38 Second Semester 51.93 59.43 13 1.32 53.94 54.83 17 .263 53.42 51.14 13 .387 60.86 60.43 13 .086 53.67 59.58 11 1.076 49.67 56.53 14 .81 aDifference significant at the .10 level. 53 bFirst semester is matched with groups from first se mester and second semester with second semester. Do not compare first with second semester groups. Table 4 Comparisons or Mean Scores or SSST or Matched Groups: Both Semesters Combineda Comparison A: Clerk, all cues Clerk, rewer cues dr t Comparison B: Proressor, all cues Proressor, rewer cues dr t Comparison C: All cues, proressor All cues, clerk dr t Comparison D: Fewer cues, proressor Fewer cues, clerk dr t Comparison E: Proressor, all cues Clerk, rewer cues dr t Comparison F: Clerk, all cues Proressor, rewer cues dr t 53.17 59.36 27 l.896b 53.51 52.83 34 .203 51.77 51.97 34 .057 56.55 59.24 28 .944 50.41 57 .10 28 2.09c 51.00 54.10 29 .98 54 aCompare only pairs, within each comparison but not be tween comparisons. bsigniricant at the .10 level. cSigniricant at the .05 level. 55 interpreted. Table 3 shows the comparisons of mean scores of the Symbol Symbol test for pairs of groups matched to test the hypotheses of this study, and to test some relationships not specifically stated in the hypotheses. The formula (Guil ford, 1956, p. 220) used to test the difference between groups is: Md t = _ / L x2 7/ ·-N__,(,=N=-------1~)- which is a formula to test differences between correlated pairs of means, where Md is the mean of the N differences of paired observations, and x is the deviation of a difference from the mean of the differences. It can be seen from Table 3 that no statistically significant results were found, al though some of the comparisons approached significance (P .10). It was then decided that combining the data from the two semesters to increase the size of the groups was a legit imate procedure because the exchange of roles between exper imenters led to no statistically significant differences be tween the groups which they tested. Of course, only the groups testing the same variables in the two semesters were combined. Clearly, these combined groups were still matched for perceptual-speed scores. Table 4 depicts the same data as Table 3, except that the differences between experimenters is ignored to increase the size of groups used to test the first three hypotheses. It is apparent that Table 4 presents more comparisons than are called for by the hypotheses of this study, showing 9 comparisons for 4 hypotheses. The additional comparisons were undertaken to reveal possible relationships between the experimental and the dependent variables which were not an ticipated by this study. These additional comparisons will be discussed in Chapter V. Hypothesis One Hypothesis I stated that the disruptive effects of anx iety (as inferred from differences in performance under two experimental conditions), which is supposed to be aroused by verbal instructions, will be reduced if the person giving the anxiety-instructions is removed from the situation and the effects of anxiety are measured by another person. There are two comparisons relevant to Hypothesis I, which are marked as Comparison A and Comparison B, in Table ~-- In comparison A the person reading the anxiety-arousing instructions was labeled 11 clerk. 11 This person remained to test the first class, but left the room when the second ex perimenter entered to test the second class. In comparison B the same procedure was followed except that the person reading the instructions was labeled "professor." Compari son A shows that the performance of the class tested by an other individual than the one who read the anxiety- 57 instructions is better than the first class, tested by the same individual who read the anxiety-instructions. In other words, removing some of the cues appears to be related to better performance on the Symbol Symbol test. However, al though the difference is in the direction predicted by the hypothesis, this difference is statistically significant on ly at the .10 level. In comparison B the difference between the groups is reversed, and the class where anxiety-cues were not removed performed slightly better than the class where anxiety-cues were removed. The difference between the means is also not statistically significant, and moreover, the difference is so small that the reversal in direction is probably due to chance only. Hypothesis Two This hypothesis stated that anxiety (inferred from dif ferences in performance under two experimental conditions) aroused by a person labeled 11 professor 11 will be greater than that aroused by a person labeled "clerk." Again, there are two comparisons relevant to this hypothesis, the first com paring the effect of 11 professor 11 versus "clerk" when all anx iety-cues are present, and the second when some of the cues I are removed. Table 4 shows that in both comparisons C and D the performance of the classes was not significantly differ ent, although the differences that do exist are in the pre dicted direction. Hypothesis Three Hypothesis III stated that a combination of the two ex perimental variables will affect anxiety-arousal (as in ferred from differences in performance under two experimental conditions) more than one variable alone. Therefore, using the label "professor" and leaving all anxiety-cues will arouse more anxiety than using the label "clerk" and remov ing some of the anxiety-cues. Comparison E in Table 4 shows that the performance of the class where a "clerk" read the instructions and later left the room is better than the per formance of the class where a "professor" read the instruc tions and remained to administer the Symbol Symbol test. The mean Symbol Symbol scores are 57.10 and 50.41, respec tively. The difference between these means is significant at the .05 level. Another combination of the variables, not stated in hy pothesis III but related to it, was also undertaken. This comparison involved the class having instructions read by a "clerk" who remained to administer the Symbol Symbol test versus the class where a 11 professor 11 read the instructions and left the room when the Symbol Symbol test was adminis tered. Comparison Fin Table 4 shows that the performance of the first group was poorer than that of the second group, with mean scores of 51.00 and 54.10, respectively. Although the difference between these scores is not significant sta tistically, it may still reflect a real difference which, 59 because of the size of the groups or some other reasons, is statistically insignificant. It is of course impossible to tell from this comparison alone which of the two variables is more effective, the number or the kind of cues, and hence responsible for the difference between the above groups. The results of the first two hypotheses do not point to one variable or the other as the more effective one. Hypothesis Four This hypothesis stated that the labels "clerk" or "pro fessor" are more effective as anxiety arousers (as inferred from differences in performance under two experimental con ditions) when applied to some individuals than to others. Table 5 shows that three comparisons were undertaken between experimenters A and B. The first two comparisons involve the two experimenters in the roles of "clerk'' and 11 profes sor." Only the classes where the experimenter both read the instructions and remained to administer the test were com pared, so as to eliminate the effects of the second variable, that of the number of anxiety cues. The third comparison be tween the two experimenters is not specified in the hypothe sis, which discussed individual differences in terms of roles only, but was undertaken nevertheless to see whether there was a difference between the two experimenters as such, dis regarding the roles they have played. As "clerk," there was only a slight difference between Table 5 Comparisons of Mean Scores ot' SSST of Matched Groups: Exp. A vs. Exp. B Comparison G: Clerk, all cues Exp. A Exp. B df t Comparison H: Professor, all cues Exp. A Exp. B df t Comparison I: Over-all Exp. A Exp. B df t aSignificant at the .12 level. 54.47 56.40 14 .41 48.27 55.55 17 1.70a 50.50 55.44 35 1.60 60 61 the two experimenters. Experimenter A obtained a mean score of 54.47 in her class, and experimenter B obtained a mean score of 56.40 in her class. The difference between these two means is not statistically significant. As 11 professor, 11 experimenter A again obtained a lower score than experimenter Bin her class, the mean scores for the respective classes being 48.27 and 55.55. The difference between these two classes, as compared to the last pair of classes, is larger and in the same direction. The over-all comparison between the two experimenters is consistent with the role comparisons. Namely, experi menter A obtained in her class the mean score of 50.50 and experimenter B again obtained a higher mean score of 55.44, but the difference between the two is not statistically sig nificant. The assumption that the anxiety aroused by the threaten ing instructions had a disruptive effect on the performance of the SSST was also tested. This was done by matching pairs on the Visual Speed and Accuracy test, and comparing the control group (which was obtained in the second semester of testing) with each of the four classes tested in the sec .end semester under one of the four experimental conditions, in terms of the mean scores on the SSST. Table 6 shows that in all four comparisons the control group obtained a higher mean score on the SSST than the respective experimen tal group. None of the differences reached statistical 62 Table 6 Comparisons of Mean Scores of SSST of Matched Groups: The Control Group Versus Each of the Four Experimental Groups Tested in the Second Semestera Control Group Clerk, all cues df t Control Group Clerk, fewer cues df t Control Group Professor, all cues df' t Control Group Professor, fewer cues df' t 60.63 50.19 15 1.812b 63.83 59.16 11 1.142 56.61 51.61 17 1.195 63.71 54.14 13 1.829b aCompare by pairs only and not between pairs. bsignificant at the .10 level. 63 signit'icance although two were beyond the .10 level. How ever, since all t'our experimental groups performed quite poorer than the control group, even though not significantly so, one may reasonably presume that the performance or the experimental groups was disrupted by anxiety. CHAPTER V Discussion Hypothesis One The results of hypothesis I were not statistically sig nificant. Therefore, it has not been unequivocally demon strated that reducing the number of cues associated with anxiety-arousal affects the amount of disruption of perform ance by anxiety. However, one of the two comparisons used to test hypothesis I showed differences in the expected di rection and approached statistical significance, P = .10. Specifically, the class where the 11 clerk 11 left the room after reading the threatening instructions performed better than the class where the 11 clerk 11 remained to administer the SSST. That is, there was less disruption of performance in the first class. Hence, the differences in performance were consistent with the hypothesis. One may speculate that, had the groups been larger, the difference between them might have been statistically more significant. Where it was a 11 professor 11 who read the threatening in structions, there was no difference between the performance of the group with all anxiety-cues present and the group with fewer anxiety-cues present. The question arises as to why the number of anxiety-associated cues affected perform ance when a 11 clerk 11 read the instructions but not when a 11 professor 11 read them. One may wonder whether the label 64 65 "pro:fessor 11 originally generated more anxiety than the label "clerk." Further, that removing a few anxiety-cues in a case where much anxiety existed did not make a real differ ence, while removing the same number of anxiety-cues where less anxiety existed made a more material dif:ference. How ever, hypothesis II, which tested the di:f:ference between the labels "clerk 11 and 11 pro:fessor, 11 did not reveal that the label "professor" aroused considerably more anxiety than the label 11 clerk, 11 or that a study where a "professor" was involved was more ego-involving that even a:fter he left there was more threat remaining. Hypothesis Two This study has not shown that the label "pro:fessor" aroused more anxiety than the label "clerk." However, though the di:fferences between the groups are small, they are in the expected direction. Namely, the group tested by the "clerk" performed slightly better than that tested by the "pro:fessor." The dif:ferences between the groups suggest that i:f the kind o:f label, 11 clerk 11 versus "professor," makes a real difference in disruption due to anxiety-arousal, that it probably is so small that in this study it has been masked by other factors. Some of the conditions which might have obscured the e:ffects o:f the variables investigated in this study are o:f:fered at the end of the discussion o:f hy pothesis III. 66 Hypothesis Three The hypothesis calling for the combination of the ef fects of the two experimental variables, number and kind of cues, was the only hypothesis that had clearly positive re sults. The class where the less potent label was used ( 11 clerk 11 ) and some of the cues associated with anxiety arousal were removed ( 11 clerk 1 ' le.ft the room) performed bet ter, (hence there was less disruption due to anxiety-arousal) than the class where a more potent label was used ("profes sor") and all anxiety-cues were left in the situation ("pro .fessor" remained to test). The difference between the above mentioned two classes was significant at the .05 level. The additional comparison which was undertaken in con nection with hypothesis III involved the classes where the two experimental variables were combined in such a way that in each class one variable was expected to decrease anxiety while the other one was expected to increase anxiety. It turned out that the group where a 11 clerk 11 (less anxiety pro voking) read the instructions and remained to test (more anxiety provoking) performed poorer, though not significant ly so, than the class where a 11 professor 11 (more anxiety pro voking) read the instructions and left the room (less anxie ty provoking). One may dismiss the difference between these classes as incidental, or one may accept the possibility that a real difference exists between the groups which, be .cause of the small size of the groups or some other reasons, 67 is not large enough to be statistically significant. If one accepts the latter possibility then the question arises: which of the two variables, number or kind of cues, is re sponsible for the difference between the groups and is hence the more effective variable in arousing anxiety. It is im possible to tell from the results of the present study alone which of the two variables is the more effective one. Several possibilities and speculations are offered here as to why the results of the first two hypotheses of this study were statistically insignificant. The Level to Which Anxiety Has Been Aroused One may wonder whether the threatening instructions to the SSST were too severe for testing the particular vari ables of this study. The students' level of anxiety regard ing their performance on the SSST might have been so high that they did not notice the label applied to the experi menter nor that the experimenter left the room after reading the instructions. In other words, making changes of number and kind of cues required by the present hypotheses might have constituted changes too small to be effective in light of the amount of anxiety generated by the threatening in structions. However, looking over the data which might offer some clues as to the amount of anxiety generated by the threaten ing instructions suggests that the instructions have not 68 been too severe. The comparison between the classes tested in the first semester with Exp. A in the major roles, versus the classes tested the second semester with Exp.Bin the major roles, shows no differences as far as the testing of the hypotheses is concerned, as can be seen in Table 3 of Chapter IV. Moreover, the differences between the control group, where the threatening instructions have not been used, and the experimental groups, where the threatening instruc tions have been used, were rather small, although in the ex pected direction. One must therefore conclude that the threatening instructions probably did not arouse such severe anxiety as to obscure the effects of the experimental vari ables. The Experimental Variables--the Cues Removing some of the cues associated with anxiety was accomplished in this study by removing the person who read the threatening instructions, which were intended to arouse anxiety. Obviously, only two variations of "number of cues" were compared--all cues versus some cues. No attempt was made to effect differences in terms of more gradations in the number of cues. There is no way of measuring the ratio of some cues to all cues, or even to estimate it roughly. One might justifiably wonder whether the number of cues re moved from the situation (accomplished by removing the per son reading the threatening instructions) was so small as to 69 make little or no difference. A priori thinking suggested that removing the person who read the anxiety instructions is removing many important anxiety-cues. Similarly, that the label "professor" when related to a testing situation in college students would arouse more anxiety than the label "clerk. 11 Within the con text of the present study, neither was the case, when tested alone. In this connection it is interesting to compare rel evant experiments using animal subjects to investigate prob lems of anxiety involving the use of cues. Miller (1948), for example, discussed more fully, earlier, showed that vari ables such as the color of the walls, although not directly related to the source of anxiety (usually shock), often prove to be sufficiently strong to make a difference in the amount of anxiety aroused in the subjects. Contrastingly, the students of this study probably recognized that it was the threatening instructions themselves which were the source of anxiety. Moreover, the anxiety had to do with the stu dents' performance rather than with an outside factor, such as the person reading the instructions. The person testing them, whether the same person read the threatening instruc tions or not, had really little to do with the content of the anxiety-arousing instructions. Therefore, college stu dents perceive more readily than animals the real source of anxiety and what cues are directly related or merely associ ated with the anxiety, and react accordingly. 70 Because hypothesis III, which combined the effects of the two variables, was statistically significant, and be cause the other two hypotheses showed some differences be tween the groups in the expected direction, one might con clude that changing the number or kind of cues probably ac counts for a small percent of the anxiety. This small meas ure of difference may at times be too small to be detected, if tested in small groups, as was probably the case in this study. Hypothesis Four This study has not unequivocally shown that one label, such as 11 professor, 11 is more effective when applied to one individual than to another. It should be noted, however, that as 11 professor, 11 Exp. A obtained poorer performance on the SSST in his class than Exp.Bin his class, although the difference is significant only at the .12 level. One might speculate again that the difference between the groups points to a real difference, but that because the groups were small, or for other reasons, it is statistically insig nificant. Hence, Exp. A was probably more effective as a 11 professor 11 than Exp. B in arousing anxiety. At the same time one might realize that the difference between the grol.JP:> treated by Exp. A and Exp. B, both as a 11 professor, 11 is only one of three similar differences between Exp. A and Exp. B. In all three comparisons, as 11 clerks, 11 as 11 professors, 11 and 71 on over-all comparison, Exp. A obtained poorer performance than Exp.Bin their respective classes, suggesting that Exp. A was possibly more effective in both roles, as "clerk" and as 11 professor, 11 and was possibly more effective in general in arousing anxiety. The possibility that the students of this study reacted somewhat differently to both experimenters is consistent with findings of other studies, already mentioned in Chapter II, that different experimenters often obtain different test records under similar conditions. In this particular study one probable reason for the difference between Exp. A and Exp. B was the latter's attitude toward the threatening in structions. Exp. B told Exp. A that she did not like the threatening instructions and thought them too severe. Al though she was encouraged to read the instructions very se riously, as did Exp. A, she read them somewhat hurriedly and in a perfunctory tone. It is not surprising then, that her classes were less threatened and performed somewhat better than those of Exp. A. Suggestions for Further Research Foremost, because the results of the present study are inconclusive, this experiment should be repeated with ap propriate modifications. It may be advisable to take into account the original level of anxiety and to divide the groups into high and low anxiety groups. Of course larger 72 groups should be obtained so that small differences, if they point to real differences, may come out statistically sig nificant. Of particular interest is the topic of labels as cues and their relationship to arousal of anxiety. Although one would probably agree that pertinent verbal symbols can in fluence one's reaction, there are few experimental studies related directly to the question of the relationship between anxiety, or other drives, and language. This study has not successfully shown that labels can materially change stu dents' reactions. Yet a relationship between verbal labels and behavior ~ay have important implications for the field of education. If certain characteristics of a teacher or an examiner, such as his professional title or standing, have a bearing on the amount learned or the efficiency of test per formance of students, then these relationships should be recognized. It might therefore prove profitable to explore more specifically the effects of labels on the performance of students. CHAPTER VI Summary and Conclusions Purpose of the Study The present study was undertaken to investigate the re lationship between the number and kind of cues present in an anxiety-arousing situation and the disruptive effects of that anxiety on performance. Specifically, the first three hypotheses stated that the disruptive effects of anxiety will be lessened if: (1) certain cues associated with anxiety arousal are removed from the situation by removing the per son arousing the anxiety, or (2) if the cue associated with the person arousing the anxiety is the label 11 clerk 11 rather than the label 11 professor, 11 or (3) if the cue associated with the person arousing the anxiety is the label 11 clerk 11 rather than 11 professor 11 and those cues associated with that person are later removed from the situation by removing that person. A fourth and last hypothesis stated that the label 11 professor' 1 or "clerk" will be more effective as an anxiety arousing cue when applied to one individual than when ap plied to another individual, under similar conditions. Method Ten classes of college students were tested and divided into 5 groups, called A, B, C, D, E. They were all given a perceptual-speed test and an anxiety-test as preliminary 73 74 tests. A Symbol Symbol Substitution Test (SSST) was then administered to all groups, each under a different situation. Group A was used as a control group and the SSST was admin istered to it with no special instructions. The other four groups, B, C, D, E, received the SSST after being warned that their performance on that test was very important to their scholastic status at school, both as individuals and as a group. The following variations were related to the administration of the SSST in these four groups. The person reading the threatening instructions to the SSST was intro duced to groups Band Casa "clerk 11 and to groups D and E as a "professor." After reading the threatening instruc tions, that person administered the SSST to groups Band D only. Groups C and E received the SSST from another experi-, menter. The performance of the groups on the perceptual speed test and the SSST was compared, as well as their performance on the anxiety test and the SSST, to see whether a relation ship exists between initial perceptual speed ability and level of anxiety and SSST performance. These comparisons showed that initial level of anxiety had no relationship to performance on SSST, but that perceptual speed ability was linearly related to SSST performance. Therefore, before the performance of the groups on the SSST could be compared to each other, subjects were matched by pairs according to their perceptual-speed scores. 75 The performance of the groups on the SSST, after match ing, was compared in four ways: (1) Between the groups where a "clerk 11 read the threatening instructions and the groups where a 11 professor 11 read the threatening instructions (groups B versus D, groups C versus E). (2) Between the groups where the same experimenter read the threatening in structions and remained in the room to administer the SSST and the group where the first experimenter read the threat ening instructions and then left the room and the second ex perimenter came in to administer the SSST (group B versus C, group D versus E). (3) Between the group where a 11 professorn read the threatening instructions and remained to administer the SSST and the group where a 11 clerk 11 read the instructions and left the room and the second experimenter came in to ad minister the SSST (group D versus group C). (4) Between that part of the group where the first experimenter was the 11 clerk 11 and the other part of that group where the second experimenter was the 11 c lerkn ( group B, comprised of two classes); between that part of the group where the first ex perimenter was the 11 professor 11 and the other part of that group where the second experimenter was the 11 professor 11 (group D, comprised of two classes); between those parts of the groups where the first experimenter read the instructions and administered the SSST and the other parts of the same groups where the second experimenter read the instructions and administered the SSST (groups B, D). Results There were no statistically significant differences as far as SSST performance was concerned between groups where: (1) a "professor" read the threatening instructions versus where a 11 clerk 11 read the threatening instructions, and (2) where one person read the threatening instructions and ad ministered the SSST versus when one person read the threat ening instructions but the second person administered the SSST. However, when the effects of the two experimental vari ables, number and kind of cues together were combined, dif ferences in group performance on the SSST appeared. That is, the group where a 11 professor" read the instructions and ad ministered the SSST received poorer scores on the test than the group where a "clerk" read the instructions and then left the room and the SSST was administered by another per son. There were also consistent differences between the groups handled by one experimenter as compared to the groups handled by another experimenter, under similar conditions, but these differences were not statistically significant. Conclusions The most apparent and important conclusion of this study is that the relationship between anxiety, cues, and performance, where human subjects are concerned, is not a 77 simple one. College students probably perceive the source of anxiety better than animals and recognize better what cues are relevant to the anxiety-arousal itself and what cues are associated with it but not directly related to it. Although much work has been done using animals as subjects in this area, similar work with human subjects is meager. This particular study has suggested that the changes effected in anxiety, by varying the number and kind of cues associated with the arousal of anxiety but are not directly related to it, are small if the variations in the cues are minimal. R E F E R E N C E S •'• REFERENCES Abernethy, E. M. The effect of changed environmental condi tions upon the results of college examinations. J. Psychol., 1940, 10, 293-301. Adler, Alfred. Understanding human nature. New York: Perma books, 1940. Alden, P., and Benton, A. L. Relationship of sex of examiner to incidents of Rorschach responses with sexual content. J. Pro,i . Tech. , 1951, 15., 231-234. Amsel, A. Selective association and the anticipatory goal response mechanisms as explanatory concepts in learning theory. J. exp. Psychol., 1949, 32., 785-799. Amsel, A., and Cole, K. F. Generalization of fear-motivated interference with water intake. J. exp. Psychol., 1953, 46, 243-247. Baughman, E. E. Rorschach scores as a function of examiner difference. J. Proj. Tech., 1951, 15., 243-249, Berger, D. Examiner influence on the Rorschach. J. clin. Psychol., 1954, 10, 245-248. Bernstein, L. The examiner as an inhibitory factor in clin ical testing. J. Consult. Psychol., 1956, 20, 287-290. Berry, J. L., and Martin, G. GSR reactivity as a function of anxiety, instructions, and sex. J. abn. §.2.£• f§;;[ chol., 1957, ~, 9-12. Birge, Jane S. Verbal responses in transfer. Unpublished doctoral dissertation, Yale Univer., New Haven, 1941. Bouthilet, Lorraine. Measurement of intuitive thinking. Univer. of Chicago Press, 194b. Brown, J. s. Problems presented by the concept of acquired drives. In Current theory and research in motivation. Lincoln: Univer. of Nebraska Press, 1953. Brown, J. S., and Jacobs, A. The role of fear in motivation and acquisition of responses. J. exp. Psychol., 1949, 3.2, 747-759. Bugelski, B. R., and Woodward, D. P. The effect of distinc tive cues on the spatial gradients of avoidance. J. comp. physiol. Psychol., 1951, 44, 450-456. 79 Bo Calvin, A. D., Koons, P. B., Bingham, J. L., and Fink, H. H. A further investigation of the relationship between manifest anxiety and intelligence. J. consult. Psychol., 1955, 1.2, 280-282. Child, I. L. Personality. Annu. Rev. Psychol., 1954, 2, 149-170. Cleveland, S. E. The relationship between examiner's anxie ty and subjects' Rorschach scores. Dissertation Ab stracts, 1951. Counts, G. S. The social status of occupations: a problem in vocational guidance. School Review, gs_, 33, 16-27. Cox, F. N. Social stimulation, anxiety level and learning efficiency. Aust. J. of Psychol., 1956, 8, 20-26. Curtis, H. S., Wolf, E. B. The influence of the sex of the examiner on the prediction of sex responses on the Ror schach. Amer. Psychologist, 1951, 6, 345-346. Dana, R.H. ology. Manifest anxiety, intelligence, and psychopath J. consult. Psychol., 1957, 21, 38-40. Davidson, W. z., Andrews, T. G., and Ross, S. Effects of stress and anxiety on continuous high speed color nam ing. J. exp. Psychol., 1956, ;ig, 13-17. Deeg, M. E., and Patterson, D. G. Changes in social status of occupations. Occupations, 1947, gs_, 205-208. Denny, M. R., and Reisman, J.M. Negative transfer as a function of manifest anxiety. Percept. Mot. Skills, 1956, 6, 73-75. Dollard, J., and Miller, N. E. Personality and psychother ~- New York: McGraw-Hill, 1950. Duvall, M. I. The effect of manifest anxiety upon inciden tal and new task learning. Unpublished doctoral dis sertation, George Peabody College, 1956. Eriksen, C. W. Personality. In Annual review of psychology~ · edited by Farnsworth, P. R., and McNemar, Q. Palo Alto, 1957. Eron, L. D., and Ritter, A. M. A comparison of two methods of administration of the TAT. J. consult. Psychol., 1951, 15., 55-61. 81 Farber, I.E. Response fixation under anxiety and non-anxi ety conditions. J. exp. Psychol., 1948, 312, 111-131. Farber, I.E., and Spence, K. W. Complex learning and con ditioning as a function of anxiety. J. exp. Psychol., 1953, 15., 120-125. Farber, I.E., and Spence, K. W. Main and interactive ef fects of several variables on reaction time. U. s. Naval Res. Lab., March 1, 1955, Tech. Rep. No. 3- Con tract N9 onr-93802, State Univer. of Iowa, p. 10. Farber, I.E., and Spence, K. W. Effects of anxiety, stress, and task variables on reaction time. J. Pers., 1956, gs_, 1-18. Fortier, J. J. The effects of test anxiety on difficult se rial learning. Dissertation Abstracts, 1956, 16, 1950. Fredenburg, N. C. Paired associates learning as a function of anxiety level and shock. Dissertation Abstracts, 1956, 16, 1950. Freud, S. The problem of anxiety. New York: W.W. Norton, 1936. Gagne, R. M., and Baker, K. E. Stimulus pre-differentiation as a factor in transfer of training. J. exp. Psychol., 1950, 40, 439-451. Gibby, R. G. Examiner influence on the Rorschach inquiry. J. consult. Psychol., 1952, 16, 449-455. Gibby, R. G., Miller, D.R., and Walker, E. L. er's influence on the Rorschach protocol. Psychol., 1953, ll, 425-428. The examin J. consult. Grice, G. R. Discrimination reaction time as a function of anxiety and intelligence. J. abn. §.2£• Psychol., 1955, 29., 71-74. - - Guilford, J. P. education. Fundamental statistics in bsychology and New York: McGraw-Hill, 195 . Horney, Karen. Our inner conflicts. New York: w. W. Norton, 1945. Hull, C. L. Principles of behavior. New York: D. Appleton Century, 1943. Kenny, D. T., and Bijou, S. W. Ambiguity of pictures and extent of personality factors in fantasy responses. J. 82 consult. Psychol., 1953, 17., 283-288. Klatskin, E. H. An analysis of the effect of the test situ ation upon the Rorschach record: formal scoring charac teristics. J. Proj. Tech., 1952, 16, 193-199- Klugh, H. E., and Bendig, A. W. The manifest anxiety and ACE scales and college achievement. J. consult. f§x_ chol., 1955, 12, 487. Lawrence, D. H. Acquired distinctiveness of cues: II. Selec tive association in a constant stimulus situation. J. exp. Psychol., 1950, 40, 175-188. Lord, E. Experimentally induced variations in Rorschach performance. Psych. Monogr., 1950, 64, 10. Luchins, A. S. Situational and attitudinal influences on Rorschach responses. Amer. J. Psychiatry, 1947, 103, 780-784. Maher, W. B. B. Manifest anxiety, shock, and stimulus-re sponse condition as determinants of performance on a verbal learning task. Dissertation Abstracts, 1956, 16, 2205. Maier, N. R. F., and Ellen, P. Can the anxiety-reduction theory explain abnormal fixation? Psych. Rev., 1951, 5§, 435-455- Maltzman, T., Fox, J., and Morrisett, L. Some effects of manifest anxiety on mental set. J. exp. Psychol., 1953, 46, 50-54. Mandler, G., Gardner, L., and Crouch, R. G. Thematic apper ception test: indices of anxiety in relation to test anxiety. Educ. psychol. Measurements, 1957, .ll, 466- 474. Mandler, G., and Sarason, S. B. A study of anxiety and learning. J. abn. ™· Psychol., 1952, 17_, 166-173. Martin, B., and McGowan, B. Some evidence on the validity of the Sarason Test Anxiety Scale. J. consult.~ chol., 1955, 12, 468. Matarazzo, J. D., Ulett, G. A., Guze, S. B., and Saslow, G. · J. The relationship between anxiety level and several measures of intelligence. J. consult. Psychol., 1954, 18, 201-205. Matarazzo, J. D., Ulett, G. A., and Saslow, G. J. Human maze performance as a function of increasing level of anxiety. J. gen. Psychol., 1955, 2.3., 79-95. 83 Matarazzo, J. D., and Phillips, J. S. Digit symbol perform ance as a function of increasing levels of anxiety. J. consult. Psychol., 1955, 1.2, 131-134. May, R. The meaning of anxiety. New York: Wm. Sloane, 1951. Mayzner, M. S., Sersen, E., and Tresselt, M. E. Manifest Anxiety scale and intelligence. Psychol., 1955, 12., 401-404. The Taylor J. consult. McCandless, B. R., and Castenda, A. Anxiety in children, school achievement, and intelligence. Child Develop., 1956, g:J__, 379-382. McClelland, D. C. Personality. New York: Wm. Sloane, 1951. McGeoch, J. A., and Irion, A. L. The psychology o( human learning. New York: Longmans, Green and Co., 1952. Milan, J. R. Examiner influences on TAT stories. J. Pro,j. Tech., 1954, 18, 221-226. Miller, N. E. Studies of fear as an acquirable drive: I. Fear as motivation and fear-reduction as reinforcement in the learning of new responses. J. exp. Psychol., 1948, 3.§., 89-101. Mowrer, 0. H. A stimulus-response analysis of anxiety and its role as a reinforcing agent. Psychol. Rev., 1939, 46, 553-565. Mowrer, 0. H. Anxiety-reduction and learning. ~-exp.~ chol., 1940, g]_, 497-516. Mowrer, 0. H. Learning theory and personality dynamics. New York: Ronald Press, 1950. Phares, E. Jerry, and Rotter, Julian B. situation on psychological testing. chol., 1956, 20, 291-293. An effect of the J. consult. Psy- Rabin, A., Nelson, W., and Clark, M. Rorschach content as a function of perceptual experience and sex of the exam iner. J. clin. Psychol., 1954, 10, 188-190. Ramond, Charles K. Anxiety and task as determinants of ver bal performance. J. exp. Psychol., 1953, 46, 120-124. 84 Rank, 0. Will therapy. New York: A. A. Knopf, 1936. Rossman, I. L., and Goss, A. E. The acquired distinctive ness of cues: The role of discriminative verbal re sponses in facilitating the acquisition of discrimina tive motor responses. J. exp. Psychol., 1951, 42, 173- 182. Ruch, F. L. Unpublished manuscript, 1957. In Ruch, F. L., Psychology and life. Chicago: Scott, Foresman and Co., 1957. Sacks, Elinor L. Intelligence scores as a function of ex perimentally established social relationships between child and examiner. J. abn. soc. Psychol., 1952, 17., 354-358. Sanders, R. The relationship between examiner's hostility and subjects' Rorschach scores. Dissertation Abstracts, 1951. Sarason, I. Effect of anxiety, motivational instructions, and failure on serial learning. J. exp. Psychol., 1956, 51., 253-260. (a) Sarason, I. G. The relationship of anxiety and "lack of de fensiveness" to intellectual performance. J. consult. Psychol., 1956, 20, 220-222. (b) Sarason, I. G. Test anxiety, general anxiety and intellec tual performance. J. consult. Psychol., 1957, 21, 485- 490. Sarason, S. B., and Gordon, E. M. The Test Anxiety Ques tionnaire: Scoring norms. J. abn. soc. Psychol., 1953, 1953, 48, 447-448. Sarason, S. B., and Mandler, G. Some correlates of test anxiety. J. abn. soc. Psychol., 1952, !±1_, 810-817. Sarason, S. B., Mandler, G., and Craighill, P. G. The ef fect of differential instructions on anxiety and learn ing. J. abn. soc. Psychol., 1952, 47, 561-565. Sherer, I . W. , Winne, J . F . , Page, H. A. , and Lipton, H. Analysis of patient-examiner interaction with the Szondi pictures. J. Pro.-j . Tech., 1952, 16, 225-237. Schlaff, A. An experimental study of the effect of anxiety on performance. Unpublished doctoral dissertation, Univer. of Southern California, 1957. Schultz, R. E., and Calvin, A. D. A failure to replicate the findings of a negative correlation between manifest anxiety and ACE scores. J. consult. Psychol., 1955, 1.2, 223-324. Siegman, A. W. The effect of manifest anxiety on a concept formation task, a nondirected learning task, and on timed and untimed intelligence tests. J. consult.~ chol., 1956, 20, 176-178. Spence, K. W., and Farber, I.E. Conditioning and extinc tion as a function of anxiety. J. exp. Psychol., 1953, ~, 116-119. Spence, K. W., and Taylor, Janet. Anxiety and strength of the UCS as determinants of the amount of eyelid condi tioning. J. exp. Psychol., 1951, 42, 183-188. Spence, K. W., Taylor, Janet, and Ketchel, R. Anxiety (drive) level and degree of competition in paired-asso ciates learning. J. exp. Psychol., 1956, 5,g, 306-310. Spielberger, 0. D. On the relationship between manifest anxiety and intelligence. J. consult. Psychol., 1958, 22, 220-224. Stevenson, H. W., and Iscoe, I. Anxiety and discriminative learning. Amer. J. Psychol., 1956, §2, 113-114. Sullivan, H. s. Conceptions of modern psychiatry. Washing ton: Wm. Alanson White Psychiatric Foundation, 1947. Taylor, Janet A. The relationship of anxiety to the condi tioned eyelid response. J. exp. Psychol., 1951, 41, 81-92. Taylor, Janet A. telligence. The Taylor Manifest Anxiety Scale and in J. abn. so~. Psychol., 1955, 51., 347. Taylor, Janet A., and Spence, K. W. The relationship of anxiety level to performance in serial-learning. J. exp. Psychol., 1952, 44, 61-64. Taylor, Janet A., and Spence, K. W. Conditioning level in the behavior disorders. J. abn. soc. Psychol., 1954, 52, 497-502. Underwood, B. J. Experimental ps*chology. New York: Apple ton-Century-Crofts, Inc., 19 9. Van Krevelen, A. Some effects of subject-examiner interac tion on projective test performance. J. Proj. Tech., 86 1954, 18, 107-109. (a) Van Krevelen, A. A study of examiner influence on responses to MAPS test materials. J. clin. Psychol., 1954, 10, 292-293. (b) Weisskopf, E. A., and Dieppa, J. J. Experimentally induced faking of TAT responses. J. consult. Psychol., 1951, 15., 469-474. Witt, G. The differential effects of stress on task perform ance as a function of anxiety level. Dissertation Ab stracts, 1956, 16, 1949. A P P E N D I X E S APPENDIX A STATISTICAL DATA Table 7 Raw Scores of the First Class on the: Sarason Anxiety Test Questionnaire, SSST, and Perceptual-Speed Test From Which Pairs Were Matched Clerk (T)*, Same Sarason SSST 21 54 16 67 10 64 25 64 33 45 28 50 16 51 21 26 13 57 22 45 21 54 33 72 16 46 2 59 9 54 14 84 24 65 15 56 5 77 25 53 4 53 31 56 14 50 5 53 20 53 21 40 P. Speed 100 96 95 90 71 103 106 73 96 87 109 130 91 129 86 113 120 93 134 112 75 100 101 86 102 50 Groups matched for: A 8 10 11 l 12 9 2 15 3 4 13 14 5 1 B 13 18 10 7 21 15 16 3 19 6 17 8 4 20 9 2 11 12 5 14 1 C 14 11 6 2 4 1 7 3 5 8 10 9 12 13 A= Clerk, same, vs. Professor, different B = Clerk, same, vs. Professor, same C = Clerk, same, vs. Clerk, different D = Clerk, same, vs. Clerk, same E = Over-all, same, vs. Over-all, same * T stands for the first experimenter 89 D 6 7 9 11 5 4 12 3 8 15 1 2 14 10 13 E 11 16 9 7 31 20 5 15 1 21 3 4 28 36 Table 8 Raw Scores of the Second Class on the: Sarason Anxiety Test Questionnaire, SSST, and Perceptual-Speed Test From Which Pairs Were Matched Clerk (T)*, Different, (Z)* Sarason SSST P. Speed 27 72 87 16 39 57 2 56 69 19 34 84 8 44 72 25 50 62 21 So 117 12 56 89 8 24 60 29 58 91 14 60 124 2 49 57 16 50 71 14 57 80 13 65 89 15 63 95 6 50 69 23 57 90 14 56 74 32 70 95 8 56 71 30 65 46 Groups matched for: C 1 3 4 5 6 7 8 2 10 11 12 9 14 13 F 6 7 11 2 4 1 10 8 3 13 5 9 14 15 12 C = Clerk, same, vs. Clerk, different G 16 14 6 13 2 15 12 8 1 3 11 10 9 7 5 17 4 F = Clerk, different, vs. Professor, different G = Clerk, different, vs. Professor, same 90 * T stands for the first experimenter, Z for the second experimenter. 91 Table 9 Raw Scores of the Third Class on the: Sarason Anxiety Test Questionnaire, SSST, and Perceptual-Speed Test From Which Pairs Were Matched Professor (T)*, Same Sarason SSST P. ~Beed E H B I G 2 43 7 18 29 73 103 14 4 11 47 101 11 3 16 60 92 15 6 30 98 10 3 13 20 45 85 5 23 75 118 2 1 20 2 21 56 104 8 2 15 17 66 74 29 14 2 13 5 26 32 70 4 16 47 83 23 11 7 49 86 4 9 16 23 40 88 18 9 6 8 10 13 47 95 13 4 10 9 18 29 71 30 16 21 16 3 9 52 81 25 12 11 11 29 53 84 22 10 14 17 54 91 17 6 8 6 8 26 84 107 6 16 5 30 24 54 33 17 1 1 27 36 82 26 14 12 62 93 5 9 20 34 83 24 10 21 45 93 14 19 25 33 61 35 18 12 3 63 79 27 13 12 27 37 65 34 17 13 25 82 143 25 72 93 12 7 17 25 28 110 17 1 17 29 90 18 8 7 7 8 72 101 12 2 14 66 74 32 15 3 15 6 Groups matched for: E == Over-all, same (T), vs. Over-all, same ( z) B == Clerk, same, vs. Professor, same G == Clerk, different, vs. Professor, same H == Z vs. T as Professor, same I == Professor, same, vs. Professor, different * T stands for the first experimenter Table 10 Raw Scores of the Fourth Class on the: Sarason Anxiety Test Questionnaire, SSST, and Perceptual-Speed Test From Which Pairs Were Matched Professor (T)*, Different (Z)* Sarason SSST P. Speed 20 60 74 14 24 77 7 45 100 7 53 88 25 49 68 29 48 71 21 37 103 21 35 83 18 46 76 20 41 79 34 59 80 26 66 129 22 50 46 20 71 102 21 62 106 17 46 76 13 30 94 12 70 113 20 71 112 6 78 86 16 53 91 Groups matched for: F 15 14 5 13 10 7 11 8 1 12 9 3 2 6 4 I 13 12 2 8 17 16 4 10 14 11 3 5 15 7 1 9 6 A 12 13 8 14 11 7 15 2 1 5 6 10 3 4 14 9 F = Clerk, different, vs. Professor, different I= Professor, same, vs. Professor, different A= Clerk, same, vs. Professor, different * T stands for the first experimenter and Z for the second experimenter. 92 Table 11 Raw Scores of the Fifth Class on the: Sarason Anxiety Test Questionnaire, SSST, and Perceptual-Speed Test, From Which Pairs Were Matched Control* Sarason 4 4 29 22 8 12 25 27 7 24 0 6 11 SSST 39 63 51 49 60 44 45 23 41 39 53 78 74 P. Speed 104 90 76 70 123 107 74 78 67 84 97 111 106 * Group eliminated from final cal culations. 93 94 Table 12 Raw Scores of the Sixth Class on the: Sarason Anxiety Test Questionnaire, SSST, and Perceptual-Speed Test From Which Pairs Were Matched Clerk (Z)*, Same Sarason SSST P. s:12eed E D w z V C 29 49 70 30 11 9 10 12 14 26 67 108 6 4 8 29 11 58 16 26 20 95 13 6 2 4 3 4 9 53 88 20 9 4 13 6 7 13 75 111 4 2 1 18 39 76 32 12 13 14 11 13 14 37 56 3 13 18 53 91 15 8 14 6 5 6 21 53 105 7 13 10 2 3 25 57 81 25 7 9 10 19 60 109 5 3 15 36 59 35 18 49 59 2 72 114 3 1 1 20 73 103 9 5 5 1 2 2 3 63 80 8 9 11 13 25 64 34 1 9 14 15 28 48 64 6 15 19 77 86 21 15 3 11 7 9 30 45 78 14 12 7 10 12 17 68 94 14 7 11 5 4 5 21 42 84 22 10 12 8 8 Groups matched for: E = Over-all, same, vs. Over-all, same D = Clerk, same, vs. Clerk, same w - Clerk, same, vs. Clerk, different z = Clerk, same, vs. Professor, different V = Clerk, same, vs. Prof'essor, same C = Clerk, same, vs. Control * z stands for the second experimenter the Table 13 Raw Scores of the Seventh Class on the: Sarason Anxiety Test Questionnaire SSST, and Perceptual-Speed Test From Which Pairs Were Matched Clerk (Z)*, Different (T)* Sarason SSST P. Speed w X u 24 56 66 1 10 8 29 75 100 3 3 9 72 86 4 7 6 5 81 101 5 4 15 73 101 10 2 30 38 69 9 9 9 19 68 97 11 6 12 21 57 86 3 8 5 30 46 67 6 22 58 101 1 2 73 107 8 3 36 81 7 13 7 10 35 78 12 11 10 22 76 96 14 5 2 34 43 72 13 12 11 26 81 101 1 24 78 97 2 14 4 Groups matched for: w = Clerk, same, vs. Clerk, different a 12 3 7 2 11 4 6 1 8 9 5 10 X = Clerk, diff'erent, vs. Professor, different u = Professor, same, vs. Clerk, different a = Clerk, different, vs. Control * z stands for the second experimenter, T stands for first experimenter. 95 96 Table 14 Raw Scores of the Eighth Class on the: Sarason Anxiety Test Questionnaire SSST, and Perceptual-Speed Test From Which Pairs Were Matched Professor (Z)*, Same Sarason SSST P. S2eed E H V y u d 18 36 34 16 22 43 79 27 13 12 10 12 23 72 95 16 4 3 7 4 17 21 53 36 16 13 76 117 2 1 18 21 85 128 1 1 31 74 104 8 2 2 2 1 2 24 44 73 31 16 12 18 9 14 6 55 74 29 14 15 11 25 46 88 18 9 6 9 5 32 29 63 18 14 17 8 15 29 54 75 28 15 11 14 13 8 64 98 10 3 4 3 3 3 66 81 26 12 9 7 9 6 43 91 17 6 5 3 6 11 52 54 33 17 13 17 22 69 89 19 8 8 7 25 31 84 23 10 7 10 6 10 23 52 96 11 7 4 5 2 4 16 59 77 10 13 11 25 62 83 24 11 8 11 8 21 68 93 12 5 6 12 5 9 64 115 1 1 Groups matched for: E = Over-all, same, vs. Over-all, same H = Exp. A, vs. Exp. B as Pr•of essor, same V = Clerk, same, vs. Professor, same y = Professor, same, vs. Professor, different u = Professor, same, vs. Clerk, different d = Professor, same, vs. Control * z stands for the second experimenter Table 15 Raw Scores of the Ninth Class on the: Sarason Anxiety Test Questionnaire SSST, and Perceptual-Speed Test From Which Pairs Were Matched Professor (Z)*, Different (T)* Sarason SSST 19 72 30 56 6 75 2 45 32 45 23 52 18 44 6 68 12 55 10 28 13 70 3 87 16 54 17 81 15 72 12 56 24 49 6 44 7 78 7 38 11 60 o 46 1 52 P. Speed 125 96 83 123 50 76 91 95 52 69 93 102 89 99 103 75 62 99 126 78 86 79 91 Groups matched for: z 5 12 14 4 3 15 2 13 1 10 9 7 11 8 6 y 5 11 16 15 9 7 18 6 8 2 14 17 4 1 13 10 12 3 X 5 13 11 6 9 14 2 8 3 1 12 10 4 7 Z = Clerk, same, vs. Professor, different b 3 6 13 10 4 8 2 7 1 12 9 14 11 5 Y = Professor, same, vs. Professor, different X = Clerk, different, vs. Professor, different b = Professor, different, vs. Control * Z stands for the second experimenter, T stands for the first experimenter. 97 98 Table 16 Raw Scores of the Tenth Class on the: Sarason Anxiety Test Questionnaire SSST, and Perceptual-Speed Test From Which Pairs Were Matched Control Group Sarason SSST P. Speed a b C d 18 65 83 7 6 9 8 0 58 69 11 8 25 Bo 110 1 1 1 22 48 58 12 16 17 6 47 90 7 7 7 13 22 83 8 10 19 88 103 2 1 2 2 20 26 48 13 18 17 31 66 12 15 15 27 59 80 8 11 11 12 22 23 31 16 20 84 76 10 13 13 29 33 138 25 76 90 6 5 6 6 3 72 82 14 19 38 78 9 9 12 11 29 40 70 10 14 14 1 80 92 5 4 5 5 10 64 100 3 2 3 3 30 61 81 10 9 23 87 97 4 3 4 4 Groups matched for: a = Clerk, different, vs. Control b = Professor, different, vs. Control C = Clerk, same, vs. Control d = Professor, same, vs. Control Table 17 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk, Sarne, vs. Professor, Different Distribution A Clerkz Sarne Professor 2 Different Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 40 50 50 46 59 129 66 129 84 113 70 113 53 112 71 112 53 102 71 102 51 106 62 106 50 103 37 103 54 100 45 100 46 91 53 91 64 95 30 94 45 71 48 71 26 73 60 74 53 75 24 77 53 86 53 88 54 86 35 83 Mean: 52.33 92.80 51.66 92.60 99 Table 18 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk, Clerk, Same, vs. Professor, Same Distribution B Same Professorz Same 100 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 40 50 24 54 53 75 66 74 26 73 66 74 54 86 49 86 53 86 45 85 45 87 40 88 64 90 29 90 46 91 54 91 56 93 62 93 64 95 47 95 56 100 47 101 50 101 72 101 54 100 30 98 53 102 73 103 50 103 56 104 51 106 84 107 54 109 28 110 67 96 43 98 57 96 45 93 65 120 75 118 45 71 29 71 Mean: 52.50 91.90 50.7 92.1 Clerk, Table 19 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk, Different, vs. Clerk, Same Distribution C Different Clerk, Same 101 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 72 87 45 87 50 71 45 71 34 84 54 86 44 72 26 73 80 117 84 113 56 89 64 90 58 91 46 91 60 124 65· 120 56 74 53 75 65 89 56 93 63 95 64 95 57 90 53 86 65 46 40 50 70 95 67 96 Mean: 59.3 87.4 54.4 87.6 Table 20 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk, Same, vs. Clerk, Same Clerkz Same Clerkz Same 102 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 72 114 84 113 75 111 53 112 60 109 54 109 67 108 51 106 73 103 50 103 20 95 67 96 68 94 64 95 53 91 46 91 53 88 64 90 42 84 53 86 49 70 45 71 39 76 26 73 53 105 53 102 45 78 53 75 77 86 54 86 Mean: 56.40 94.13 54.46 93.86 Table 21 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Over-all Z* (Clerk, Same and Professor, Same) vs. Over-all T* (Clerk, Same and Professor, Same) Distribution E 103 Over-all Z Perceptual- Symbol Symbol Speed Over-all T Perceptual- Symbol Symbol Speed 85 76 72 75 60 67 53 74 73 64 52 68 20 68 53 72 43 46 69 53 77 42 31 62 57 66 43 54 55 49 44 39 52 25 36 21 128 117 114 111 109 108 105 104 103 98 96 93 95 94 91 95 91 88 89 88 86 84 84 83 81 81 79 75 74 70 73 76 54 64 59 53 59 75 84 53 54 84 51 56 50 30 67 72 47 45 46 64 54 40 29 45 54 53 47 34 52 36 63 53 66 29 26 66 24 37 33 40 129 118 113 112 109 107 106 104 103 98 96 93 95 93 91 95 91 88 90 87 86 84 83 83 81 82 79 75 74 71 t~ 54 65 61 50 Symbol Symbol Mean: 55.44 Table 21 (continued) Perceptual Speed 88.69 Symbol Symbol 104 Perceptual Speed 88.69 * T stands for the first experimenter, Z stands for the second experimenter. Table 22 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk, Different, vs. Professor, Different Distribution F Clerk 2 Different Professor 2 Different 105 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 60 124 66 129 80 117 70 113 63 95 30 94 58 91 53 91 57 90 53 88 72 87 78 86 34 84 35 83 57 80 59 80 56 74 46 76 50 71 48 71 44 72 46 76 65 46 50 46 50 69 49 68 70 95 45 100 56 71 60 74 Mean: 58.1 84.4 52.5 85.0 Table 23 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Professor, Same, vs. Clerk, Different Distribution G Professor 2 Same Clerk 2 Different 106 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 24 54 49 57 75 118 80 117 29 71 50 71 32 70 56 71 66 74 56 74 66 74 44 72 29 90 57 90 54 91 58 91 47 95 63 95 40 88 65 89 52 81 57 80 33 61 24 60 37 65 50 62 53 84 34 84 60 92 56 89 49 86 72 87 72 93 70 95 Mean: 48.1 81.59 55.4 81. 41 Table 24 Perceptual-Speed and Symbol Symbol Substitution Scores t'or Pairs Matched t'or Perceptual-Speed Scores: Prot'essor, Same, vs. Prot'essor, Same Distribution H Prot'essor :1. Same Prot'essor 1 Same 107 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 76 117 75 118 74 104 56 104 64 98 30 98 72 95 47 95 68 93 62 93 43 91 54 91 52 96 43 98 69 89 29 90 46 88 40 88 31 84 53 84 62 83 47 83 66 81 52 81 43 79 63 79 55 74 66 74 54 75 66 74 44 73 29 71 52 54 24 54 29 63 33 61 Mean: 55.55 85.4 48.27 85.3 Table 25 Perceptual-Speed and Symbol Symbol Substitution Scores ror Pairs Matched ror Perceptual-Speed Scores: Proressor, Same, vs. Proressor, Dirrerent Distribution I Proressorz Same Proressor 2 Dirrerent 108 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 28 110 71 112 72 101 45 100 47 101 71 102 73 103 37 103 84 107 62 106 54 91 53 91 72 93 30 94 40 88 53 88 49 86 78 86 34 83 35 83 52 81 59 80 63 79 24 77 66 74 60 74 36 82 41 79 66 74 46 76 29 71 48 71 37 65 49 68 Mean: 53.01 87.6 50.7 87.6 Table 26 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Professor, Same, vs. Clerk, Different Distribution U Professorz Same Clerk 2 Different 109 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 74 104 81 101 52 96 76 96 64 98 75 100 72 95 78 97 46 88 57 86 31 84 72 86 66 81 36 81 29 63 56 66 44 73 38 69 43 79 35 78 55 74 43 72 68 93 68 97 Mean: 53.67 85.67 59.58 85.75 Table 27 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk 2 Clerk, Same, vs. Professor, Same Distribution V Sarne 110 Professor 2 Same Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 72 114 64 115 73 103 74 104 20 95 72 95 68 94 52 96 53 91 43 91 53 88 46 88 77 86 31 84 42 84 62 83 63 So 66 81 45 78 59 77 39 76 54 75 49 70 44 73 37 56 52 54 25 64 29 63 Mean: 51.14 84.21 53.42 84.21 Symbol 25 20 77 53 73 48 57 67 49 53 68 45 39 53 Mean: Table 28 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerkz Clerk, Same, vs. Clerk, Different Distribution W Same Clerkl Different 111 Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 64 56 66 95 78 97 86 57 86 88 72 86 103 81 101 64 46 67 81 36 81 108 73 107 70 38 69 105 73 101 94 68 97 78 35 78 76 43 72 91 76 96 51.93 85.93 59.43 86.oo Table 29 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk, Different, vs. Professor, Different Distribution X Clerk 1 Different Professorz Different 112 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 58 101 72 103 73 101 87 102 75 100 81 99 81 101 44 99 76 96 56 96 68 97 68 95 72 86 60 86 57 86 54 89 38 69 28 69 56 66 49 62 35 78 52 76 43 72 56 75 36 81 75 83 78 97 70 93 Mean: 60.43 87.93 60.86 87.64 Table 30 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Professor, Same, vs. Professor, Different Distribution Y Professorz Same Professor 1 Different 113 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 85 128 78 126 74 104 72 103 43 91 52 91 64 98 44 99 52 96 56 96 68 93 70 93 72 95 68 95 69 89 54 89 46 88 44 91 31 84 60 86 62 83 75 83 43 79 46 79 59 77 38 78 54 75 56 75 55 74 52 76 21 53 45 50 29 63 49 62 44 73 28 69 Mean: 53.94 85.72 54.83 85.61 Table 31 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Clerk, Same, vs. Professor, Different Distribution Z Clerkz Same Professori Different 114 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 73 103 72 103 53 105 87 102 37 56 55 52 20 95 68 95 68 94 56 96 53 91 52 91 45 78 38 78 63 80 46 79 25 64 49 62 49 70 56 75 77 86 60 86 42 84 75 83 53 88 54 89 39 76 52 76 48 64 28 69 Mean: 49 .67 82.26 56.53 82.40 Table 32 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Control, vs. Clerk, Same* Control Clerk.z Same 115 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 80 110 75 111 88 103 73 103 64 100 53 105 87 97 20 95 80 92 68 94 76 90 53 91 47 90 53 88 22 83 42 84 65 83 77 86 61 81 57 81 59 80 63 80 38 78 45 78 84 76 39 76 40 70 49 70 31 66 25 64 48 58 11 58 Mean: 60.62 84.81 50.18 85.25 * Both groups from the second semester only. Table 33 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Control, vs. Clerk, Different* 116 Control Symbol Symbol Perceptual Speed Clerk, Different Symbol Symbol Perceptual Speed Mean: 80 88 64 87 80 76 65 59 38 40 58 31 110 103 100 97 92 90 83 80 78 70 69 66 86.50 73 81 75 68 76 57 72 36 35 43 38 56 59.16 * Both groups from the second semester only. 107 101 100 97 96 86 86 81 78 72 69 66 86.58 Table 34 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Control, vs. Professor, Same* 117 Control Symbol Symbol Perceptual Speed Professor, Same Symbol Symbol Perceptual Speed Mean: 80 88 64 87 80 76 47 65 61 22 38 59 84 40 31 23 48 26 56.6 110 103 100 97 92 90 90 83 81 83 78 80 76 70 66 31 58 48 79.78 64 74 64 52 68 43 69 62 66 31 59 43 54 44 29 36 52 21 * Both groups from the second semester only. 115 104 98 96 93 91 89 83 81 84 77 79 75 73 63 34 54 53 80.11 Table 35 Perceptual-Speed and Symbol Symbol Substitution Scores for Pairs Matched for Perceptual-Speed Scores: Control, vs. Professor, Different* Control Professorz Different 118 Symbol Symbol Perceptual- Symbol Symbol Perceptual- Speed Speed 88 103 72 103 64 100 87 102 87 97 56 96 80 92 44 91 76 90 52 91 65 83 75 83 47 90 54 89 58 69 28 69 38 78 38 78 84 76 52 76 59 80 46 79 48 58 49 62 26 48 45 50 72 82 60 86 Mean: 63.71 81.86 54.14 82.50 * Both groups from the second semester only. Table 36 Raw Scores Where the Median SSST Fell Question·>E- 4 5 6 7 8 9 10 11 12 13 14 15 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Men 7 6-7 4 5 5 5 6-7 6 3 2 2 6 5 6 4 7 6 6 6 6 6 3 3 9 6-7 5 6 5 5 5 6 5 3 3 2 Women 6 6 6 5 5 5 6 6 2 2 3 5 6 5 5 6 6 5-6 5 5 5 3 3 8 7 7 5 6 4 4-5 5-6 5 2 2 4 119 * The assigned number is that which the question actu ally has in the questionnaire. Questions 1, 2, 3, and 16 were 11 f'iller 11 i terns and were not scored. 90- 86- 82- 78- 74- 70- .-I 66- ; 0 .D 62- E >- 58- ' C/) ' .-I 54- 0 50- '..0 E 46- ' >- : Cl) 42- 38- 34- 30- 26- 22- 18- Correlation Between Anxiety Test Scores and SSST • • • . • • . • . : . . . . . . . . • .. • .. . . . . • • . . . • . • . . .. . . ! .. ; •· • • . • . 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Anxiety Graph No. 1 I-' I\) 0 90- 86- 82- 78- 74- 70- 66- 62- 58- 54- 50- 46- 42- 38- 34- 30- 26- 22- 18- Correlation Between SSST and Perceptual-Speed Scores Control Group • • • • • • t • • • • • Control Group P Correlation .43 • ----------------------------------------------- 30 34 38 42 46 50 54 58 62 66 70 74 78 82 86 90 94 98 102 106 110 114 118 122 126 130 134 138 142 Perceptual-Speed Graph No. 2 I-' (\) I-' APPENDIX B TEST MATERIALS APPENDIX B Test Materials Instructions for the Visual Speed and Accuracy Test (the perceptual-speed te'st"; and the Anxiety Test Questionnaire We are conducting a survey on how students feel toward tests and examinations. This is a general survey and not especially related to how you feel toward any course in par ticular. I have two tests for you, one is rather short, as part of the survey. Please write your names or your ini tials, and sex. Although we are not especially interested in each person individually, we still would like your true feelings. The first test is timed. (Distribute tests.) Please read the instructions and when everyone is ttrough reading, we will begin. Your co-operation is greatly appreciated. 123 TIME LIMIT: 5 minutes NATURE OF THE TEST EMPLOYEE APTITUDE SURVEY TEST 4-Visual Speed and Accuracy Developed by: G. Grimsley, F. L. Ruch & N. 0. Warran MAXIMUM SCORE: 150 The Employee Aptitude Survey Visual Speed and Accuracy Te st is a measure of the ability to see small details quickly and accurately, as in visual inspection and clerical work. The more detailed the "paper work" required on a job, the more import.ant it is that the worker be high in this ability. Performance on this test is especially important in the evaluation of personnel for positions as bookkeepers, account ants, general office clerks, typists, stenographers, and operators of most kinds of office machines. Most sales, supervisory, and executive positions require above-average performance on this test. ADMINISTRATION Distribute the test forms with the directions side up, instructing the subjects not to turn the sheet ova:.· until the signal is given. When all examinees have entered their names and other pertinent information in the blanks on the face of the test sheet, say: I am going to read the test directions with you. Follow them carefully. Look at the pairs of numbers below. The first pair of numbers, 792 and 792, are exactly alike. Therefore the space to the right under S (same) has been filled in. The second pair of numbers, 6122 and 6123, are not exactly the same. Therefore the space to the right under D (different) has been filled in. The next pair, $898 and $898, are marked ta shaw that they are the same. The fourth pair, 72.10 and 72, 10, are marked as different because one has a period in it while the other has a comma. Now mark the next four items for practice. Allow ti,11e for the examinees to fill in the answers to the sa1nple problems, Then say: Yau should have marked them D, S, D, and S. Have you any questions? Clarify any questions which may arise. Then go on: When the signal is given, turn this sheet over and check as many pairs of numbers as you can in five minutes. Put a heavy black mark between the little dotted lines under the S if the two numbers are exactly the same. If they are different in ~ way, put a heavy black mark between the little dotted lines under the D. If you want to change an answer, be sure to erase completely. Do not waste time making pretty marks. A heavy black mark between the two little dotted lines is all that is needed. Work as fast and accurately as possible. Ready •.. Go! At the end of exactly five minutes, say "Stop!" and collect the papers. PSYCHOLOGICAL SERVICES, INC.' 909 Wost Jefferson Blvd. LOS ANGELES 7, CALIFORNIA SCORING The test may be scored either by hand or by IBM Scoring Machine. For either type of scoring, the special scoring stencils may be used. All forms of the test may be scored by the same keys, The scoring formula for the test is the number of right items minus the number of wrong items. Items which have been omitted or II skipped" by the examinee are not counted -either right or wrong. To score the test by hand, follow these steps: 1. Lay the RIGHTS KEY over the answer spaces. Count the total number of answer marks showing through the holes in the stencil. This total is the Rights score. 2. Lay the WRONGS KEY over the answer sheet. Count the total number of marks showing through the holes in the stencil. This total is the Wrongs score, 3. Subtract the Wrongs score from the Rights score. INTERPRETATION In the practical employment situation, it is important to interpret a person's Visual Speed and Accuracy score in relation to the requirements of the particular job for which he is being evaluated, Because different job categories vary widely in their ability requirements, an individual's score is be st interpreted in terms of norms for an appropriate occupational or educational group. Norms for a variety of groups are provided on the back page of this Manual. These norms have been compiled over a period of several years in cooperation with large business organizations, manufacturing concerns, high schools, and colleges. GENERAL POPULATION NORMS In addition to the norms for special occupational and educational groups, norms are provided for the general productive population. For the purpose of constructing these norms, the general productive population was defined as that part of the population above 16 years of age with !Q's above 75. In terms of California Mental Maturity IQ' s, the normative group formed a normal distribution, truncated at the lower limit of 7 5, with a modal IQ of 100 and a standard deviation of 16. The IQ' s in the sample of 891 cases ranged from 76 to 148, For practical purposes, this range includes essentially all of the employ able population. RELIABILITY AND VALIDITY In a study conducted with a wide-range sample of 90 high school students, test-retest scores on Form A of the Visual Speed and Accuracy Test yielded a correlation of . 84, which is equivalent to an index of reliability of . 92. For another sample of 70 high school commerce students, alternate-forms scores yielded a correlation of . 86, which corresponds to an index of reliability of. 93. For a sample of 335 junior college students, alternate-forms scores yielded a correlation of . 87, with a corresponding index of reliability of. 93. Substantial correlations have been obtained between the Visual Speed and Accuracy Test and other standardized tests of clerical ability. For a heterogeneous sample of 89 job applicants for a variety of positions, a correlation of. 82 was obtained with the Minnesota Clerical Test 1. From a sample of 100 successive inmates passing through a prison reception center, a correlation of. 65 was obtained with the DAT Clerical Speed and Accuracy. In a study conducted in a California high school, the Visual Speed and Accuracy Test was administered to 70 graduating commerce students. During the final six weeks prior to graduation, these students were temporarily "placed" as clerks in the offices of local concerns. At the end of this trial period, 33 of the students were permanently hired by the firms where they had been "apprenticed." For those who were hired, the mean Visual Speed and Accuracy score was 94. 45, compared with a mean score of 82. 30 for those who were not hired. The correlation between Visual Speed and Accuracy scores and the criterion of hirability was . 48. In a study conducted at the AiResearch Manufacturing Corn.pany, the Visual Speed and Accuracy Test was administered to 37 inexperienced engineering drawing trainees. All subjects had completed high school courses in algebra, trigonoinetry, plane and solid geometry, and two semesters of drafting, At the end of four months of training, each trainee received a proficiency rating based on quality and quantity of work, adaptability, and technical knowledge. Visual Speed and Accuracy score correlated . 45 with pro ficiency rating. -2- In a study conducted by Lockheed Aircraft Corporation, California Division, the Visual Speed and Accu racy Testwas administered to 47 draftsmen enrolled in a junior engineer training program. Visual Speed and Accuracy scores correlated ,40 with grades in strength of materials, and. 36 with grades in descrip tive geometry. In another study conducted in the same company, a significant relation was found between Visual Speed and Accuracy scores and efficiency ratings of 142 production supervisors. The difference between the mean scores of the high-rated supervisors and the low-rated supervisors yielded at-ratio of 2. 99, which is beyond the 1% level of significance. Another indication of the validity of the Visual Speed and Accuracy Test may be obtained from a com parison of the median scores of persons employed in various occupational categories requiring different levels of this ability. Such a comparison is provided in the table below. It may be seen that persons in job categories requiring heavy, detailed paper work obtained relatively high Visual Speed and Accuracy scores, Group No. of Cases Median Score Mathematical analysts 42 115 Accountants 131 109 Industrial relations clerks 47 108 Graduate de sign engineers 117 96 Stenographers 100 95 Typists 100 90 General population 891 83 Industrial leadmen 131 71 Power truckers 44 -6s Note - Data on math analysts, industrial relations clerks, de sign engineers, and power truckers obtained from research conducted by Lockheed Aircraft Corporation, California Division. Other data were obtained from re search conducted by the authors and publishers. UNIQUENESS OF THE TEST To obtain 1naximum validity per minute of testing time from a battery of employment tests, each test should be as nearly independent as possible with respect to the other tests in the battery. To avoid wasteful overlap from one test to another, each test should be a unique measure of some characteristic not measured by any of the other tests. In the ideal situation, each test in the battery should be 100% independent of the others. Table III on page 4 indicates coefficients of independence obtained for the Visual Speed and Accuracy Test with respect to the other tests in the Employee Aptitude Survey battery. AL TERNA TE FORMS The Visual Speed and Accuracy Test is currently available in two alternate forms -- Form A and Form B. Repeated studies with various industrial and educational samples indicate that these two forms are statistically equivalent and may be used interchangeably. Studies in which both alternate for1ns have been administered during the sarne testing session indicate a definite practice effect on this test. The extent of this practice effect .fluctuates at different levels along the range of scores. Because of this non-uniform practice effect, special retest norrns are provided for interpreting ~he scores of individuals who are retested on an alternate form of the test. -3- G~· o- E .9;: N -"' ..!l• T 5..~ I 0 L 11. E ~ . " c 90 104 75 92 60 87 55 84 50 83 15 BO 40 79 25 69 10 59 ~-- C e- o- E ·-"' - D N ~ n T 5.~ I 0 11. L E ] " Q ----- ... '}0 118 75 10•1 60 9'! 55 % 50 95 45 ~2 -10 'JI 25 7'} IO 63 . I I, ;:-1 - - 0 - c:: ~:;; ! m " ~ • N ""' ""' - a " " " "" ~~ " "'• z ·;;; uiZ g~ -~ ~ t;- ··- 0. ;;" " 0 • 'El, 0 . - .5 • C " " " . "' ~-M "1 j " C 0 " "' 0 . 8 ~ "1 ;:"1 . p. c l) Ji 0 u f-' .,. ---· .. 123 121 119 125 132 107 108 I 07 Ill 122 97 IOI 99 I DI 114 95 99 97 97 112 93 96 95 93 109 91 95 93 91 106 88 93 91 8'} 103 83 87 85 81 93 74 76 75 73 82 TAllLE I NOllMS SCORE " 0 -~ r: "- N E~ •o "' -"' . ~ C n 'O" . " .. ~ 7. -;; '-l'lz ~~ ...i- l/J -- ~ ~ ~ ] :~i ~ . ~ n. . 0 u.t, -;; 'O <X l/J ..'i =-:=-'--- -- ·---- 121 101 125 I 07 87 113 97 77 l 05 95 75 101 93 71 99 91 69 97 89 67 95 83 59 87 75 53 77 TABLE II RETEST NORMS " 0 •o • •D "' O..c, 0 u ,-. .,., -~" ,O " I ~ II C " ~ Ill Z ~~ t/l7, E~ uc- -- 0 . ui 0 0 . ...':i -~ _,: , '-'JO. . l/J 0 "' ~ C 0. . p,-t; fa ,.. d :i: . f-' " ,..___,. ___ 119 129 109 89 107 117 95 82 99 I 07 89 78 95 105 87 71 93 101 85 76 91 99 83 7-1 89 97 81 73 83 89 75 65 75 83 65 5-1 _To be used for interpreting retest scores following initial test on allcrnatl! form. SCORE . 0 - -~r:- ~:; § §' ·- 0 ,-. - ~:;; ;;; -;;; "' "o 0 ,n C ,-. •o ~ 7i " •"' _g,o ;; "' - • -0 ",,.. - _,_, u ,- _,., c- "' 0 " .~ II "" CJ II " ~ n . " .... II C II ~ II . " ~:~ '1l Ul~ ~z ez • :es. ~?;. Zz ~-z ~ Ul z •z U) z oz . ·- .- i'o- ...i- -- u~- "'- g- ~- l) ~ Cl • c • C 0 0 ~ . n. 1 .. ui ~ ~ o C C . .:: ~i ...:I-~ -5 0 ~ ·c:o "1 • § co'n. . -~ A "' "' a C ,g .: . s . " n. " a. e gw .w 0 0. . 0 ~< ·o.,<:( 0 fa . <) 0 " 0 ~ Ji -0 :,.. :i: " l/J f-' .s l/J f-' Ll ·----~--,- ------ , .. .. ----- .. -- .. 135 l J-1 l 3Z I 38 141 !J.l 115 138 132 141 123 IOI 12 I 122 I I 2 I I 24 ll1 121 99 126 121 130 109 94 85 115 Ill 115 127 Ill 89 119 113 121 I 01 90 I 09 113 ! Ill Ill 125 109 87 115 109 119 99 8') 107 110 l 09 107 123 107 83 113 107 l 15 97 88 103 lW3 101 103 IZO 103 H Ill 103 l 13 I ')$ BL 100 107 103 J 01 l 17 I DI 76 109 101 Ill 93 85 95 CJ9 97 ~3 I I 07 95 63 99 95 IOI 87 73 86 88 87 85 94 87 50 89 87 95 73 50 TAllLE Ill Coefficients of Independence Between the Visual Speed and Accuracy Test and Other Tests in the Employee Aptitude Survey Dattcry Employee Freshman Junior Aircraft High Aptitude Engineering College Production School Survey Student a Students Leadmcn Seniors Test (N•335) (N-139) (N •178) (N•90) I. Verbal Comprehension JOOo/a 90% 86% 86% z. Numerical Ability 82 75 69 65 3, Visual Pursuit 94 94 85 86 5. Space Visualization 86 93 71 92 6, Numerical Reasoning 99 88 77 82 7. Verbal Reasoning 100 96 · 66 83 8, Word Fluency 99 83 80 77 9, Manunl Speed 98 Bl 93 JO. Symbolic Reasoning 94 94 n 88 Note - The coefficient of independence ia computed as 100(1 - r2). -4- -~- '" 0 " . ,g_ 0 -- .. 0 "o c,,O . 0 o- " - C ,1 U •f) ,1 ~z i::,...: ... ,, : ,;,i 1-. ~. ,;;- lJ. - ~ ~u :- "' 0 0 n. 0. E E w "1 - ....,. _____ 121 12.1 107 105 99 97 97 '} 5 95 92. 93 89 91 87 83 78 I 75 66 . " u . 5 _,: n. ·- g "o ~ "'o o- - " " 0 C • ti~ <:.J.-.: z 0 "- . 'O .,,- . ,u ~ ~ p. p. E E ul "1 .. I 3,1 134 121 119 113 Ill Ill 109 109 l 04 l 07 l Ol l 03 '!9 95 90 87 74 w .J VJ-;- LLJ b°-. a:: 0 u VJ 3: <( 0: ...J t- j:: Ill 0 .... t <( Cl w ::i.: <( z a:: ~ "' < ~ j w 1/l j + U) C - "' I- U) "' ...J - .. :,;: Ill 0 l!) - u < "' sf C - < 0: M - Cl I- "' < N ...J - ... .... 0 w w 0: > "' 0 OJ 0 :E _, ::, a.. z ::i.: 0\ "' w ...J CXl u 0: r-- u z "' U) ;:: z z 0 <( j:: u < :J u a.. ::> a.. C c:( L,J ~ I I I >- z < t- a.. z ::i.: <( 0 .J u a.. EMPLOYEE APTITUDE SURVEY TEST 4-Visual Speed and Accuracy FORM A Developed by: G. Grimsley, F. L. Ruch & N. D. Warren Look at the pairs of numbers below. The first pair of numbers, 792 and 792, are exactly alike. Therefore the space lo the right under S (same) has been filled in. The second pair of numbers, 6122 and 6123, are not exactly the same. Therefore the space to the right under D (different) has been filled in. The next pair, $898 and $898, are marked to show that they are the same. The fourth pair, 72.10 and 72,10, are marked as different because one has a period in it while the other has a comma. Now mark the next four items for practice. s D 792 792 I .. s D 6123 6122 .. I s D $898 $898 I s D 72,10 72.10 I s D 33333 33323 s D 117! 117! .. .. s D 42 24 .. s D 6696 6696 .. .. You should have marked them D, S, D, and S. Have you any questions? When the signal is given, turn this sheet over and check as many pairs of numbers as you can in five minutes. Put a heavy black mark between the little dotted lines under the S if the two numbers are exactly the same. If they are different in~ way, put a heavy black mark between the little dotted lines under the D. If you want to change an answer, be sure to erase completely. Do not waste time making pretty marks. A heavy black mark between the two little dolled lines is all that is needed. Work as fast and accurately as possible. COPYRIGHT. 1952 PSYCHOLOGICAL SERVICES, INC. 909 Wost Jefferson Blvd. LOS ANGELES 7, CALIFORNIA PRINTED IN U.S.A. IBM FORM I.T.S. 1100 /1. 3107 ]()/) 31.o7 38 :19156 100 31.67 83 // 63291 Sh3291 $63291 .222227 .222227 89(100.0 8900.0 596 $674 .5')() 000.001 00.0001 634 56.03' $63.89 1,793 634 56.03';i $6,1.89 5786 C S786 C 83-t.01 '., 834.01 ',, -167.n $(d9 ·1677.3.1 $6.48 697.S3 -l~.\T6 45ATC> 56.-132 111117 11117 $55.38 78/'IJJ 78')/l 3 S D D s l) lJ S D [) S D S D S D S D s u s lJ S D S D S D S D S D s [J s I) S D s I) S D S 11 S D D D s [) S D 875.566 87.:'i66 $5879.3 'r 5879.3 ,\0()()()() 40000 4789P 11679 4789Q 11678 21 $,() .2234).1:i .2233½ .5890 58&21 2IJ45 .5890 5fr821 21345 121,21 131,:n'' 81.12 42.6 l 0 813.2 4.2610 56? 13 ,, 21 ,;. 13 j; 21 31,435 2/.f56 31,435 2/456 U 12345 U12345 :'i46,21 l 54,62 l l 489,211 -189,211 .C).j()() $79..1-5 9400 $79A6 98X-500 98-XSOO 45,Bl 02[(; 45,131 021C $385.00 $385.01) A6996W J\69%V 90.002'/ ')0.0021/c s l) • • 5()0(i(1() ,'i(){)I)()() [J s I) •. $90.{)J $90.02 S D · -15678 s I) .00005 .OllOOOS S D .. 900001 900001 S 0 689 698 s n 6934721 693421 S D $(166.99 $666.99 S D S D 58W30 SSW30 S D . 3222~!2 3222222 S D . 640/37 3 640/372 s [1 • $00.33 $00.33 S D 2223444 2233444 S D ' "6903 '! 1 6903 S D •.·. *79145 $79145 S D •• 68. 99'}il 68. 99';, S D •• 5566885 5566885 S D S D 5 D S D S D S D 8435 791L 8435 7911, 3156(31 3156(31 .5.6.70 . .'i,11.70 $589.76 U589.76 . 32567 3256711 s [) :: 21111121 :; 2lml21 S D •• 690.111 69<H 1 s u • 112,321 112,321 S Ll ' 115'/cl ll 115')D 11 S D 115,211 11,521\ D S D l' I) S D " [1 s [J 0 $ D S D S D S D S 0 s fJ S D S D s () S D S D S D S D S D s n s l) s u S 0 S D s n S 11 s l) .21,356 2 U56 3$'ir671 3$'/o671 365BA 35hBA ZJ.410 21..!IO 612.B 6LU2 D23456 D23456 $28,312 $28.312 31,21,31 31,21,31 53998N 53998N 28,710 28.71() 1/569 1/5.69 8934!<\ 39341,; 345679 345678 58.4';'c: 58.4'}{. 4938621 493821 29 5/7 29 5/7 ''304 '30-1 -533.22 '533.22 MU IOU 681.1011 943/571 94/:',;571 $.25 $.25 : 856,1).1 :! 856,0-1 777-11 77741 ''5.-122 ': 5.422 54E566 54F56(i '}5S.4')o 955.-J\1c .11112 .lllB 793.22 -'793.'2 l) 33.222 33,222 S D 9.-1/68 9·1/69 841 841 S D $111.1 $11.11 S D 57 75 S D -803.4 -803.5 S D 56G666 56G666 s [) • 90.02~{, 90.02~~ S D • If 81.79 f! 81.79 S 11 .. 579843 579873 S D S D s [) S 11 S D S D 56 831 561I33 65/312 55/312 36578 36578 689 2/4 689 3/4 -21.31 -21.31 • P3789 P3689 S D S D s lJ S D 21456 21456 $35.69 ~1d5.69 .28999 .82999 s !j l) S i.l S D S D S D S D S D 5 D S D S D s l} S D S D S D S D S D s !) s [) $J5LJ,60 $359,60. s [) s lJ 84%.38 84 ii .38 ::; (1 s I) -90.24 90.24 s [) S D :?345b7 243567 s I) s I) $ll<i.75 $116.75 s [) s [J 1921 1921 s ]) s [J Vl'l 90 \'El 90 s ,, s lJ '291 $291 s u s (_l 921Y 92.lY s f) fl 42~Sl 42$18 •; D :;; fl 121.11 112.l l l:JM ror~M 1.1 s. 11rr I\ J1ou.2 126 QUESTIONNAIRE ON ATTITUDES TOWARD THREE KINDS OF TESTING SITUATIONS NAME (Please Print): This questionnaire is designed to give you an opportun ity to indicate how and what you feel in regard to three types of testing situations: al the group intelligence test or aptitude test. b the course examination. c the individual (face-to-face) type of intelligence test. One of the main reasons for constructing this question naire is the fact that very little is known about people's feelings toward the taking of various kinds of tests. We can asswne that people differ in the degree to which they are affected by the fact that they are going to take a test or by the fact that they have taken a test. What we are particularly interested in here is how widely people differ in their opinions and reactions to the various kinds of testing situations. The value of this questionnaire will depend in large part on how frank you are in stating your opinions, feelings, and attitudes. We are requesting you to give your name only because it may be necessary for research purposes. Each of you has taken a course examination and a group intelligence or aptitude test, but not all of you have taken an individual intelligence test. Those of you who have not taken such a test are requested to answer in terms of how you think you would react to them. We want to know what you think your attitudes and feelings toward taking such a test would be and not what you think it ought to be. Those of you who have taken an individual intelligence test will, of course, answer the questions in terms of what you actually experienced. For each question there is a line or scale on the ends of which are statements of opposing feelings or attitudes. In the middle of the line you will find either the word "Midpoint'' or a phrase, both of which are intended to re flect a feeling or attitude which is in-between the state ments of opposing feelin~s described above. You are re quested to put a mark (X) on that point of the line which you think best indicates the strength of your feeling or at titude about the particular question. The midpoint is only for your guidance. Do not hesitate to put a mark on any point on the line as long as that mark reflects the strength of your feeling or attitude. 127 THERE ARE NO "CATCH" QUESTIONS ON THIS QUESTIONNAIRE. PLEASE READ EACH QUESTION AND EACH SCALE VERY CAREFULLY. THERE IS NO TIME LIMIT. SECTION I The following questions relate to your attitude toward and experience with group intelligence or aptitude tests. By group intelligence tests we refer to tests which are ad ministered to several individuals at a time. These tests contain different items and are usually paper and pencil tests, with answers requiring either fill-ins or choices of several possible answers. Scores on these tests are given with reference to the standing of the individual within the group tested or within specific age and educational norms. Please try to remember how you usually reacted to these tests and how you felt while taking them. THE MIDPOINT IS ONLY FOR YOUR GUIDANCE. DO NOT HESITATE TO PUT A MARK (X) ON ANY POINT ON THE LINE AS LONG AS THAT MARK REFLECTS THE STRENGTH OF YOUR FEELING OR ATTITUDE. 1. How valuable do you think group intelligence tests are in determining a person's ability? Very valuable Valuable in some respects and valueless in others Valueless 2. Do you think that group intelligence tests should be used more widely than at present to classify students? Should be used less widely Should be used as at present Should be used more widely 3. Would you be willing to stake your continuance in col lege on the outcome of a group intelligence test which has previously predicted success in a highly reliable fashion? Very willing Uncertain Not willing 4. If you know that you are going to take a group intelli gence test, how do you feel beforehand? Feel very unconfident Midpoint Feel very confident 128 5, After you have taken a group intelligence test, how confident do you feel that you have done your best? Feel very unconfident Midpoint Feel very confident 6. When you are taking a group intelligence test, to what extent do your emotional feelings interfere with or lower your performance? Do not interfere at all Midpoint Interfere a great deal 7. Before taking a group intelligence test, to what extent are you aware of an uneasy feeling? Am very much aware of it Midpoint Am not aware of it at all 8. While taking a group intelligence test, to what extent do you experience an accelerated heartbeat? Heartbeat does not accelerate at all J\lidpoint Heartbeat notice ably accelerated 9. Before taking a group intelligence test, to what extent do you experience an accelerated heartbeat? Heartbeat does not accelerate at all Midpoint Heartbeat notice ably accelerated 10. While taking a group intelligence test, to what extent do you worry? Worry a lot Midpoint Worry not at all 11. Before taking a group intelligence test, to what extent do you worry? Worry a lot Midpoint Worry not at all 129 12. While taking a group intelligence test, to what extent do you perspire? Perspire not at all Midpoint Perspire a lot 13. Before taking a group intelligence test, to what extent do you perspire? Perspire not at all Midpoint Perspire a lot 14. In comparison with other students, how often do you think of ways of avoiding a group intelligence test? Less often than other students Midpoint More often than other students 15. To what extent do you feel that your performance on the entrance test, if taken, was affected by your emotional feelings at the time? Affected a great deal Midpoint SECTION II Not affected at all The following paragraph relates to your attitude toward individual intelligence tests and your experience with them. By individual intelligence tests we refer to tests which are administered to one individual at a time by an examiner. These tests contain different types of items and thus pre sent a variety of tasks. Those tasks can be both verbal and manipulative, i.e., verbal or written answers to questions or manipulation of objects such as is involved in puzzles, form boards, etc. Examples of tests of this type would be the Stanford-Binet test, and the Wechsler-Bellevue test. Please try to remember how you have usually reacted toward these tests or how you would expect to react to them. 16. Have you ever taken any individual intelligence tests? Yes No (Circle appropriate answer) If your answer to the above question is YES, indicate in the following questions how you did or do react to indi vidual intelligence tests. 130 If your answer to the above question is NO, indicate in the following questions how you think you would react to or feel about individual intelligence tests. 17. When you are taking an individual intelligence test, to what extent do (or would) your emotional feelings in terfere with your performance? Would not interfere with at all Midpoint Would interfere a great deal 18. If you lmow that you are going to take an individual intelligence test, how do you feel (or expect that you would feel) beforehand? Would feel very unconfident Midpoint Would feel very confident 19. While taking an individual intelligence test, how con fident do you feel (or expect you would feel) that you have done your best? Would feel very confident Midpoint Would feel very unconfident 20. After you have taken an individual intelligence test, how confident do you feel (or expect you would feel) that you have done your best? Would feel very unconfident Midpoint Would feel very confident 21. Before taking an individual intelligence test, to what extent are you (or would you be) aware of an "uneasy feeling?" Am not aware of it at all Midpoint Very much aware of it 22. While taking an individual intelligence test, to what extent do you (would you) experience an accelerated heartbeat? Heartbeat does not accelerate at all Midpoint Heartbeat notice ably accelerated 131 23, Before taking an individual intelligence test, to what extent do you (would you) experience an accelerated heartbeat? Heartbeat does not accelerate at all Midpoint Heartbeat notice ably accelerated 24. While taking an individual intelligence test, to what extent do you (would you) worry? Worry a lot Midpoint Worry not at all 25. Before taking an individual intelligence test, to what extent do you (would you) worry? Worry a lot Midpoint Worry not at all 26. While taking an individual intelligence test, to what extent do you (would you) perspire? Would never perspire Midpoint Would perspire a lot 27. Before taking an individual intelligence test, to what extent do you (would you) perspire? Would never perspire Midpoint Would perspire a lot 28. In comparison with other students, how often do you (would you) think of ways of avoiding taking an indi vidual intelligence test? More often than other students Midpoint SECTION III Less often than other students The following questions relate to your attitude toward an experience with course examinations. We refer to major examinations, such as mid-terms and finals, in all courses, not specifically in any one course. Try to represent your usual feelings and attitudes toward these examinations in general and not toward any specific examination you have taken. We realize that the comparative ease or difficulty 132 of the course may influence your attitude toward the examina tion; however, we would like you to try to express your feel ings toward course examinations generally. THE MIDPOINT IS ONLY FOR YOUR GUIDANCE. DO NOT HESITATE TO PUT A MARK (X) ON ANY POINT ON THE LINE AS LONG AS THAT POINT REFLECTS THE STRENGTH OF YOUR FEELING OR ATTITUDE. 29, Before taking a course examination, to what extent are you aware of an "uneasy feeling?" Am not aware of it at all Midpoint Am very much aware of it 30. When you are taking a course examination, to what ex tent do you feel your emotional reactions interfere with or lower your performance? Do not interfere with it at all Midpoint Interfere a great deal 31. If you know that you are going to take a course examina tion, how do you feel beforehand? Feel very unconfident Midpoint Feel very confident 32. After you have taken a course examination, how confi dent do you feel that you have done your best? Feel very confident Midpoint Feel very unconfident 33. While taking a course examination, to what extent do you experience an accelerated heartbeat? Heartbeat does not accelerate at all Midpoint Heartbeat notice ably accelerated 34. Before taking a course examination, to what extent do you experience an accelerated heartbeat? Heartbeat does not accelerate at all Midpoint Heartbeat notice ably accelerated 133 35. While taking a course examination, to what extent to you worry? Worry a lot Midpoint Worry not at all 36. Before taking a course examination, to what extent do you worry? Worry a lot Midpoint Worry not at all 37. While taking a course examination, to what extent do you perspire? Never perspire Midpoint Perspire a lot 38. Before taking a course examination, to what extent do you perspire? Never perspire Midpoint Perspire a lot 39. When, in your opinion, you feel well prepared for a course examination, how do you usually feel just before the examination? Confident Midpoint Anxious INTRODUCTION FOR CLERK Our school is participating in a project comparing abilities of students who attend local colleges. Miss 134 (name of experimenter) is a statistical clerk employed by this project and has been asked to administer a test to you. INTRODUCTION FOR PROFESSOR Our school is participating in a project comparing abilities of students who attend local colleges. Dr. (name of experimenter) is a research associate on the staff of the Psychology Department, University of California at Berkeley. She is conducting this particular project and is going to administer a test to you. Dr. perimenter). THE THRENI'ENING INSTRUCTIONS (name of ex- TO THE SYMBOL SYMBOL SUBSTITUTION TEST (SSST) The test you are going to take is a revision of a sub test that appears on the Wechsler Adult Intelligence Test. The reason we have made this revision is that we found that this test is a good indicator of I.Q. This revision is ex tremely difficult. However, it correlates very highly with total I.Q. This test has been given to university students at both UCLA and USC so we know what to expect from univer sity students. Former research has indicated that there are differences between university students and junior college 135 and city college students, and that city college students do not do as well as university people. We are checking that fact now by giving you this revision of tl1is intelligence test. Even though the test has not been perfected yet, it is regarded as a good predictor of success in college. There fore, a copy of your score will be kept on file here at school. If the question of your continuation in college ever comes up, this score will be one of the determining factors. After the tests are scored, the results will be returned to Dr. (name of instructor) and he will post the scores so you can all see how well or poorly you did. INSTRUCTIONS FOR THE SYMBOL SYMBOL TEST Please write your name and y0ur sex. Now look at the divided boxes or squares at the top of the page. Notice that each has a mark on the upper part and another symbol on the lower part. Every mark has a differ- ent symbol. And now look where the boxes have marks on the upper part only. I want you to put in each of these squares, the first seven squares up to the double line, the marks that should go there. Stop when you come to the double line. When I give the signal begin after the double line and fill in as many squares as you can without skipping any. Keep working until I tell you to stop. GO AHEAD. (3 min utes.) (Ordinarily additional verbal instructions were nec ~ssary. THE SYMBOL-SYMBOL SU.i3S 1 rITlJTiON TBS·r (Revised from the Wechsler Adult Intelligence Test) ' 2 6 0 @ � � \] G VI 'v v1 8 � 2 0 � 6 @ 1 ~ 0 ~ .6 /\ 0 0 6. 2, � \/ � I •~ � 0 ·~ '0' V Q l i ' ~VI 2 G 0 V 6 <) Vl c) 0 V � 6 8 N v • G � 2 111 <> ~ 0 [I v ~ � c" c) � c) 0 i.. 2 � 0 \I c) 0 2 G VI � � 0 G v @ 2 -1/1 ~ D 2 o. c) 0 2 8 G \/ 2 0 � G VT ' \/ D \I @ VI � 2 1/1 c) � ~u \I � 2 �· 0 0 6. 8 6 G \/ 2 D f/1 � Vl \j 0 2 6 D '2 V 2 G Vf V � 0 G V 1/l Q 6 ·� 2 6 ~�. - G c) 2 � 8 G 6 2 � 2 0 2 V d 6

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Effects Of Prior Part-Experiences On Visual Form Perception In The Albino Rat

PDF

Dominance And Contiguity As Interactive Determinants Of Autonomic Conditioning

PDF

The Influence Of The Control Of Personal Set Upon Prediction By Factored Tests Of Temperament And Interest

PDF

Transfer Of The Partial Reinforcement Extinction Effect Across Tasks In Normal And Retarded Boys

PDF

The Effects Of Ether And Electroconvulsive Shock On One Trial Appetitive And Adversive Learning

PDF

Classical Discrimination Conditioning As A Function Of Probability Of Reinforcement

PDF

The Modification Of Partial Reinforcement Effect As A Consequence Of Electrical Stimulation Of The Caudate Nucleus In Cats

PDF

Human Gsr Classical Conditioning And Awareness Of The Cs-Ucs Relation

PDF

The Concept Of Sexual Identity In Normals And Transvestites: Its Relationship To The Body-Image, Self-Concept And Parental Identification

PDF

The Effects Of Group Experiences On The Aged

PDF

The Effect Of Motor Ability Loss On Cognition And Emotion

PDF

Habituation Of The Multiple Unit Discharge Response To White Noise Stimulation In The Unanesthetized Rabbit

PDF

An Effect Of Verbal Conditioning On Nonverbal Behavior

PDF

Autonomic and cognitive indices of semantic conditioning and generalization

PDF

Noxious Auditory Environment And Psychomotor Performance

PDF

The Effect Of Discriminability On The Partial Reinforcement Effect In Human Gsr Conditioning

PDF

Use Of College Students In A Social Therapy Program With Chronic Schizophrenics To Produce Changes In Mood And Social Responsiveness

PDF

Tension And Anxiety In Deconditioning

PDF

Cue-Dependent Amnesia

PDF

Verbal Reports Of Emotional States And Onsets And Offsets Of Conditioned Stimuli

Asset Metadata

Creator Ilanit, Tamar (author)

Core Title The role of cues in the arousal of anxiety

Degree Doctor of Philosophy

Degree Program Psychology

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

Tag OAI-PMH Harvest,psychology, experimental

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Jacobs, Alfred (committee chair), Grings, William W. (committee member), Locke, Harvey J. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-908488

Unique identifier UC11356021

Identifier 5906387.pdf (filename),usctheses-c18-908488 (legacy record id)

Legacy Identifier 5906387.pdf

Dmrecord 908488

Document Type Dissertation

Rights Ilanit, Tamar

Type texts

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

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

Repository Name University of Southern California Digital Library

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