1 x ‘ | l l WW1”|\|\H\|1|H|W ‘ k ‘ W — __’— — ‘I l l l “W -_-'w_-‘ no M CID-pd W 0? BEHAVIORAL WRAWAL é)? ‘E'HE BMW~A~PER$GN TEST Thesizferthemoffid.A. MECMGAN STATE UM’ERBNY C. MCI-L439 ELLCH 3966 TH ESIS LIBRARY MW" 5"" Rm”? gag cm TH E515 TH E515 ABSTRACT INDICES OF BEHAVIORAL WITHDRAWAL ON THE DRAWeA-PERSON TEST by C. Richard Ellch The purpose of the present study was to determine the relationship between certain Draw-AsPerson indices and behavior- al withdrawal. The indices of overall size, placement, motion, detail and time were used in the analysis. The method for the present study consisted of having 2% withdrawn and 21+ non-withdrawn Veteran's Administration patients draw three figures-~an automobile, a male and a female. The male and female drawings of each subject were scored on each of the five indices. The automobile drawing was used as a warm-up task and, therefore, was not scored. The findings indicated that withdrawn subjects tended to draw figures that were significantly larger, closer to the center of the page, and which contained significantly more detail than the figures drawn by the non-withdrawn patients. Furthermore, the withdrawn patients took significantly longer to complete the human figure drawings than the patients in the non-withdrawn group. There was no difference between the groups, however, on the motion index. It was suggested that the pattern of the results could be interpreted in terms of major defensive maneuvers or libido TH E515 C. R. Ellch theory. The withdrawn patient was characterized as utilizing an obsessive-compulsive defensive pattern and as having withdrawn his libido cathexes from the external world. The non-withdrawn patient, rather than withdrawing into a fantasy world, was found to handle conflict by responding to his environment in an impulsive manner and fleeing from the anxiety-arousing situation. It was concluded that an approach stressing the inter-related patterns of the indices could be of great value_in assessing and interpreting the personality variables symbolically repre- sented in human figure drawings. Furthermore, it was con- cluded that an approach which maximizes the difference between the eXperimental groups on the critical variable necessarily increases the predictive abilities of the DAP. 122% 27/ Ma [0/ [2/ THESIS INDICES OF BEHAVIORAL WITHDRAWAL ON THE DRAW A PERSON TEST By C. Richard Ellch A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1966 TH E515 ACKN OWLEDGEI-lEN T S The writer would like to express his sincere appre- ciation to his chairman, Dr. Joseph Reyher whose guidance and cooperation made this study possible. Also, the writer would like to acknowledge the help of his committee members Drs. Bertram Karon and John Hurley. A Special debt is owed to the following for their help in the preparation of this manuscript: Dr. Larry Kipperman, Heward Porter, the staff and personnel of the Battle Creek Veteran's Administration Hespital and, especially Theresa G. Ellch. ii THESIS ?. r—J J TS ‘4 TABLE OF CON Page ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . ii Chapter I. INTRODUCTION . . . . . . . . . . . . . . . . II. METHOD. . . . . . . . . . . . . . . . . . . . Subjects. Apparatus O O O O O O O O O O O O O O O O 0 Procedure 0 O O O O O O O O O O O O O O 0 O Scoring . . . . . . . . . . . . . . . . . . \J \n4r¥1u t» F4 III. RESULTS . . . . . . . . . . . . . . . . . . . IV. DISCUSSION. . . . . . . . . . . . . . . . . . 11 V. SUMMARY . . . . . . . . . . . . . . . . . . . l7 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 27 iii THESIS LIST OF TABLES Table Page 1. Average age and length of admission and percentage of patients diagnosed schizophrenic for the h3l§s by be- havioral group and test order sub- group O O O O O O O O O O O I O O O O O O O O O 19 2. Summary of reliability coefficients for modified HOyt-Baron scoring syStem O O O O O O O O O O O O O O O O O O O O 20 3. Mean values of the five indices by behavioral group and test order sub- group O O O O O O O O O O O O O O O O O O O O O 21 E. Summary of analysis of variance for size index between.withdrawn and non- withdrawn groups. . . . . . . . . . . . . . . . 22 5. Summary of analysis of variance for placement index between withdrawn and non-withdrawn groups. . . . . . . . . . . . . . 23 6. Summary of analysis of variance for motion index between withdrawn and non-Withdrawn grouPSQ o o o o o o o o o o o o o 2"} 7. Summary of analysis of variance for detail index between withdrawn and non-w1thdram groups. a o o o o o o o o o o o o 25 8. Summary of analysis of variance for time index between withdrawn and non withdrawn groups. . . . . . . . . . . . . . 26 iv TH ESIS CHAPTER I INTRODUCTION Despite conflicting evidence concerning the validity of the technique, the DragegrPerson ngt is being extensively used by the clinician to draw inferences concerning the nature of an individual's drives, conflicts, motives and behavioral tendencies. Traditionally, clinicians have based these inferences on the projective hypotheses of Buck (l9h3), Machover (l9h9) and their followers. Recent research, how- ever, has left serious doubt as to the validity of a number of the original hypotheses of these workers (Swenson, 1957; Handler 8c Reyher, 1965). Much of the ambiguity concerning the interpretation of the drawings seems to be the result of the attempt to re- late certain conflicts to specific symbolic aspects of the drawings in a one-to-one fashion. A new trend in the process of the interpretation of these drawings is an attempt to move away from the minute examination of separate details to a consideration of all those drawing indices which taken together might symbolize this conflict (Reyher, 1959). Furthermore, Handler and Reyher (1965) suggest that a possible reason for a number of the non-significant findings might be the choice of criteria by which the DAP's validity THESIS is determined. Many of the previous studies have used diagnostic categories to set up experimental groups. Taking into consideration the impreciseness of diagnostic methods and the large variance in behavior within any given category, the reliability of such selection procedures is left in ser- ious doubt. Other studies have used paper and pencil tests of anxiety to set up groups. The relationship between such anxiety scores and conflicts is also somewhat tenuous. Accordingly, a selection method which would maximize the difference on the critical variables between the two ex- perimental groups would be an improvement in design. One such critical variable which can be objectively assessed in order to maximize differences between groups is behavioral withdrawal. The purpose of this study, therefore, was to determine the relationship between certain DAP indices and the variable of withdrawal. The following DAP indices were used in the study: (1) overall size of the figure; (2) placement of the figure on the page; (3) the amount of detail in the drawing; (H) the motion of the figure in the drawing; (5) the amount of time it takes to complete the drawing. These five indices were selected because they seemed to be sensitive to differences in the tendency to withdraw from the environment. THESIS CHAPTER II METHOD ..... The.§s consisted of #8 male patients at the Battle Creek Veteran's Administration Hospital. The‘Ss‘were selected by the working staffs of four wards in the hospital who were in constant contact with the patient population. Three members of each ward staff were asked to select a group of 12 patients composed of the six men whom they felt were the most behavior- ally withdrawn and the six men whom they felt were the least behaviorally withdrawn. Avoidance of staff members and other patients, noneinvolvement in ward projects and the tendency toward restricted behavior were considered to be indicative of behavioral withdrawal. The criteria involved in the selec- tion of non-withdrawn §s, on the other hand, included the tendency to seek out staff members and other patients, constant involvement in ward projects and the inability to restrict their need for interpersonal contacts. Three other criteria were also considered in the selec- tion process. These criteria were: (1) residency on an Open rather than a closed ward; (2) residency in the hospital of six or more months; (3) consideration of only those patients who were in fairly good contact, i.e., he does not openly talk to himself, lose control of himself from time to time and 3 THESIS is not openly hallucinating or delusional. These criteria were used to insure that any differences in the results could not be accounted for by differences between the groups in terms of environmental setting or treatment methods, institu- tionalization factors or degree of contact with reality. Apparatus Each drawing was made on a sheet of 3% x 11" unlined white paper and each.§ was given a well-sharpened pencil for each drawing. A stopwatch and time sheets were used to record the length of time (in seconds) each'fi needed to complete each drawing. Procedure The experiment was set up in such a way that the MSISS were divided into two groups-- Group I consisting of those. ‘§s judged behaviorally withdrawn and Group II consisting of those‘fis judged non-withdrawn. Each of the two groups was then divided randomly into two sub-groups-- sub-groups IA and IIA were asked to draw the human figures in a male-female order and sub-groups IB and IIB were asked to draw the human figures in a female-male order. The design, thus, consisted of four sub-groups composed of l2'§s each. ' Each‘S'was asked to draw three figures-- an automobile, a male and a female. As a result of recent investigations (Handler, 196%; Handler & Reyher, 196%; Sanders, 1965) which have shown that the act of drawing an automobile is much less anxiety-arousing than drawing human figures, the §s were first THESIS 5 asked to draw this relatively neutral figure so that they would feel less threatened by the testing situation. The remaining order of presentation was dependent upon the sub-group member- ship of each §. Each S was seated at a desk which allowed him ample freedom for movement while drawing the pictures. The‘SS'were instructed as follows: "Now, draw a picture of an automobile (male, female) and tell me when you have finished.” If‘§ asked for additional information he was told, "Just draw it (him, her) anyway you like, but be sure to draw the whole figure." Occasionally, if the,§ seemed quite anxious about the task, he was told, "This is not a test of drawing ability; just do the best you can." If the §,failed to draw a whole figure, he was given a new sheet of paper and told, "This time draw the whole figure." Scoring Each human figure was scored for size of the figure, diSplacement of the figure from the center of the page, motion, detail and elaboration of the drawing and length of time needed to complete the drawing. The automobile drawings were not scored. Displacement from the center and size of the figure were determined by means of a ruler. The displacement and size scores were reported in units, where one unit was equal to i". Time scores were recorded in terms of length of time in seconds each § needed to complete each drawing. The indices of kinesthetic motion and detail and elaboration were scored for degree of presence or absence of THESIS each index using a modified Heyt-Baron scoring system (Hoyt & Baron, 1959; Handler & Reyher, 196%). Each drawing was scored on the two indices by the experimenter and a graduate student in clinical psychology who was unfamiliar with the research project. Previous studies have shown that such indices can be scored with a high degree of reliability (Hoyt & Baron, 1959; Mogar, 1962; Handler 8c Reyher, 1961+). Score values ranging from one to four were assigned, where a rating of one represented a relative absence and a rating of four represented a high degree of presence. THESIS CHAPTER III RESULTS The average age and length of admission of the patients and the percentage of patients diagnosed schizophrenic were tabulated for each of the groups and sub-groups to determine the degree to which inter-group differences might effect the DAP results. The data are summarized in Table l. InEeItITEbIeIEanEch‘érE The differences between the groups and orders on the age and average length of admission variables were investi- gated by means of a two-tailed t-test for differences be- tween means assuming equal variances. The value of t obtained on the age variable for between withdrawn and nondwithdrawn groups was .0% and for between MF and FM orders was 1.32. The value of t obtained on the length of admission variable for between withdrawn and nondwithdrawn groups was .68 and for between MF and FM orders was 1.69. Since none of these values are significant at the .05 level with #6 degrees of freedom, it can be said that inter-group differences on the age and length of admission variables do not significantly effect the DAP results. The reliability of the modified Hbyt-Baron scoring procedure for rating the indices of motion and detail-elab- 7 THESIS oration was determined by computing the Pearson product-moment correlation coefficient between the scores assigned by the experimenter and those independently assigned by the second psychologist. The reliability of these ratings is summarized in Table 2. ln§e1t_T able_2_a_bo ut_h_er e The results indicate that such indices can be rated ‘with a high degree of reliability. The differences between the groups on the five indices were investigated as follows. 'The index scores were subjected to an analysis of variance for a three factor experiment with repeated measures on only one of the three factors (Miner, 1962). The main effects and all interactions of the three factors - behavioral group, test and test order - were deter- mined using this method. The mean values for the two groups and four sub-groups were obtained by scoring the drawings on each of the five indices. The summary of the mean scores appears in Table 3. Ina9£fi_Iahle_3_abont_h2ra Size S . The F-ratio for the between groups of 12.02 obtained from the analysis of variance was significant at the .01 level. This indicates that, irreSpective of test or test order, the withdrawn'Ss drew significantly larger human figures than the non-withdrawn‘fis. Also significant at the .01 level was the F-ratio of 8.76 for the test by test order interaction. This finding points to the fact that THESIS those'§s who were asked to draw the male first drew larger females, while those §s who were asked to draw the female first drew larger males. The analysis of variance is sum- marized in Table H. InEeEtITEbIeZMIanfitIhErE Placement ‘ N O The F-ratio; of H.29 for the between withdrawn and non-withdrawn groups obtained from the analysis of variance was significant at the .05 level. This result indicates that, irrespective of test or test order, non-withdrawn‘Ss dis- placed their human figures further from the center of the page than withdrawn §s. The analysis of variance is sum- marized in Table 5. ' InEeEtZTEbIeZSZanEtIhErE None of the F-ratios for the motion index were signifi- cant at the .01 or .05 levels. This indicates that there were no differences between the two main groups or the four sub-groups in terms of the amount of motion displayed in their human figure drawings. The summary of the analysis of variance appears in Table 6. InEeitZTEbIeZ6Ian'fitIhErE Detail ‘ O ' The obtained F-ratio of 13.25 for between experimental groups was significant at the .01 level. This finding indicates that, irrespective of test or test order, the drawings of the withdrawn‘Ss contained significantly more THESIS 10 detail than the drawings of the non-withdrawn‘Ss. The analysis of variance for the detail index is summarized in Table 7. inEeEtITEbIeI7IanEchEr§ Time * ' ' ' ' The-F-ratio for between experimental groups of 19.93 obtained from the analysis of variance was significant at the .01 level. This result indicates that, irrespective of test or test order, withdrawn‘Ss needed significantly more time to complete the human figure drawings than DOD! withdrawn.§s. In addition, the F-ratio of 5.9% for the test main factor was significant at the .05 level. This finding points to the fact that, irrespective of experimental group or test order, it took all.§s significantly longer to complete the female drawing than the male drawing. |nzauo CHAPTER IV DISCUSSION Our findings which indicate that withdrawn patients tend to draw significantly larger figures than non-with- drawn patients contradict the traditional clinical hypothesis that small size indicates a definite tendency to withdraw from the environment. According to Buck (l9#8), many withdrawn schizophrenics draw tiny figures in order to symbolize their limited interaction with the environment. Similarly, Machover (l9h9) states that a high energy level and self- esteem are suggested by large figures, while small figures indicate low energy level and self-esteem. Our results, however, tend to support the work of Helzberg and wexler (1950) who found that normal women drew figures that were smaller significantly more often than schiz0phrenic women. In addition, it was found that the patients tended to draw the second human figure significantly larger than the first human figure. This supports the impression of Handler and Reyher (1965) that size increase and decrease is a re- liable index of the anxiety-arousing task of drawing human figures. The interpretation of diSplacement from the center of the page differs from one authority to another. Our results, 11 THESIS 12 which indicate that non-withdrawn patients tend to displace their human figure drawings further from the center of the page than withdrawn patients, support the notion of Buck (19h8) who states that there is a tendency toward exact centering of the figures in withdrawn, rigid individuals. The results, on the other hand, seem to contradict the hypo- thesis of Machover (19%9) who suggests that placement to the left side of the page is characteristic of self-oriented rather than environment-oriented individuals. Our finding that there was no significant difference between withdrawn and non-withdrawn patients in terms of motion displayed in their drawings also contradicts one of Machover's original hypotheses. According to Machover (19%9), schizophrenics who are entirely withdrawn and fantasy bound produce drawings which display a clear impulse to movement which is counteracted by introversive features. Our finding not only fails to support this theory but presents evidence that withdrawn §s actually produce drawings with somewhat more kinesthetic motion than nonawithdrawn‘Ss. This result supports the findings of Helzberg and Wexler (1950) and Royal (19H9) who found no significant differences on this index between normals and schizophrenics and normals and neurotics. The greater elaboration of detail in the withdrawn group supports the earlier work of Buck (l9k8), Machover (19%9) and Witkin (1962). Each has suggested that, although detailed drawings may represent a high level of interest and THESIS l3 respect for the body, they may also indicate a high degree of narcissistic self-absorption and withdrawal from reality. Furthermore, the results obtained from comparing the groups on the time index lend support to the work of Barman, Klein and Lippman (1951) who found that introverted, withdrawn‘Ss took much longer to produce figure drawings than outgoing. or impulsive §s. The findings of the study which indicate that, irre- spective of group or order, patients took significantly longer to complete the drawing of the female than the male seem to corroborate the findings of Handler and Reyher (196%) who found that the female drawing, drawn by a male, is much more anxiety-arousing than the male drawing. Furthermore, it would seem that the limited contact with the opposite sex due to the hospitalization of these patients would also increase the anxiety associated with the female figure. The interpretation of these results would be less meaningful if each finding was looked at as being independent from the others. This would be comparable to a cook-book, analytic approach. Rather, the stress must be laid primarily on the inter-related patterns indicated by the results. It was found that the withdrawn subject tended to draw large figures that were placed near the center of the page. Furthermore, the drawings showed a high degree of elaboration and detail and took a relatively large amount of time to be drawn. 0n the other hand, the drawings of the non-withdrawn subject tended to diSplay small figures which were randomly THESIS 1% displaced from the center of the page. Mereover, the drawings contained little detail and were completed in a relatively short period of time. These patterns of results can be accounted for by inves- tigating the differing modes of defensive maneuvers used by the two groups in the stressful clinical situation. One as- pect of their differing defensive structures is reflected by the obsessive-compulsive - impulsive dichotomy. The obses- sive-compulsive defensive pattern is extensively utilized by the withdrawn patient who attempts to combat intrapsychic and external stress by excessive preoccupation with details and the rigid structuring of his environment. He tends to cope with stress. 0n the other hand, the impulsive indivi- dual combats anxiety by acting out. This mode of defense is utilized by the non-withdrawn patient who must avoid details and structure and act upon the task as quickly as possible to avoid anxiety. The striking differences between the groups can also be accounted for by libido theory. The introverted patient has withdrawn from the frightening external world to the inner world of fantasy. He has invested the libido he has withdrawn from external cathexes on his own self and body. This high fantasy self-esteem and narcissistic self-absorption is reflected in the large size of the figures, the wealth of detail, the tendency toward exact symmetry and the reluctance to leave the task until fully satisfied. The outgoing, extroverted subject, on the other hand, THESIS 15 has not withdrawn his libido cathexes from the external world. As a result of the fact that he is not as narcissistically involved in the task as the withdrawn subject, he is unable to cOpe with the situation and, instead, attempts to control his anxiety by avoiding the task. This flight reaction, as reported by Handler and Reyher (1965), is reflected in the varied placement of the figure on the page, the minimal amount of detail and the tendency to leave the situation as quickly as possible despite the poor quality of the finished drawing. Witkin (1962) and his associates suggest that the type of drawings produced by our non-withdrawn group may also indicate a lack of interest in the body itself. This would seem to explain the relative small size of the human figure drawings. However, perhaps the smallness is not simply the manifestation of too little interest, but rather, an indication of the great interest in the body coupled with a great deal of anxiety associated with such thoughts. On the other hand, the introverted subject is able to express his great interest and cOpe with the situation because he has withdrawn himself into a fantasy world in which his own body is of the highest esteem. The overwhelming significance of many of the results in- dicates that the DAP can be successfully applied in the area of research, especially when the experimental design is set up in such a way that the differences between the experimental groups are maximized. Furthermore, the findings suggest that an interpretive approach which utilizes the inter-related THESIS l6 patterns of the figure drawing indices can lead to the discovery of striking differences between eXperimental groups who are known to differ in some significant way. The results of this study suggest that further research along these lines would be beneficial. Due to the findings concerning size and placement, in particular, the study should be replicated. The results concerning these two indices contradict the traditional clinical hypothesis that withdrawn subjects draw small figures which are displaced from the center of the page. Secondly, research utilizing variables akin to behavioral withdrawal might disclose the superiority of such methods as compared to those utilizing diagnostic categories and paper and pencil tests of anxiety. THESIS CHAPTER V SUMMARY The purpose of the present study was to determine the relationship between certain Draw-A-Person indices and behavioral withdrawal. The indices of overall size, place- ment, motion, detail and time were used in the analysis. The findings indicated that withdrawn subjects tended to draw figures that were significantly larger, closer to the center of the page, and which contained significantly more detail than the figures drawn by the nonewithdrawn patients. Furthermore, the withdrawn patients took significantly longer 'to complete the human figure drawings than the patients in the non-withdrawn group. There was no difference between the groups, however, on the motion index. It was suggested that the pattern of the results could be interpreted in terms of major defensive maneuvers or libido theory. The withdrawn patient was characterized as utilizing an obsessive-compulsive defensive pattern and as having withdrawn his libido cathexes from the external world. The non-withdrawn patient, rather than withdrawing into a fantasy world, was found to handle conflict by responding to his en- vironment in an impulsive manner and fleeing from the anxiety- arousing situation. 17 THESIS 13 It was concluded that an approach stressing the inter- related patterns of the indices could be of great value in assessing and interpreting the personality variables symboli- cally represented in human figure drawings. Furthermore, it was concluded that an approach which maximizes the difference between the experimental groups on the critical variable ne- 'cessarily increases the predictive abilities of the DAP. 19 Table 1 Average Age and Length of Admission and Percentage of Patients Diagnosed Schizophrenic for the %8‘§s by Behavioral Group and Test Order Sub-Group Group Mean Age Average Length % Schizo- (in years) of Admission phrenic* (in months) Withdrawn 40.33 92.25 96% MF Order 39.03 78.17 92% FM Order %1.58 106.33 100% Non-Withdrawn no.29 80.16 83% MF Order no.33 57.75 83% FM Order h0.25 103.17 83% *Index Scores of Non-Schizophrenics not significantly dif- ferent from Schizophrenics. THESIS Index Detail Detail Motion Motion Table 2 Summary of Reliability Coefficients for Modified Heyt-Baron Scoring System Test Male Female Male Female Reliability .99 .90 .93 .35 20 THESIS 21 «.55 5.56 :.m5 m.m5H 5.05, 6.:6e m.:5 m.5m m.oo muonounsm Howno pmoa cam macho Hmnowbmsom an mmoHGGH m>Hm may go 5H.N 5o.a mm.H 09m Rae. mad m mr.m mm.H oo.m 5H.m N0.N mo.m 2 Hannah mm.a mm.H mo.m oo.m mm.a m5.a m5.a 5H.m mm.m m:.m so.m om.m a an scape: om.m 5H.m 5a.: mm.: mm.m 5m.m om.m om.m oo.m 5F.m mm.m :m.m a a pcmEoomHm mmoaeeH m manna II mr.ea mm.ma 5m.ma 5e.mm mm.mm om.mm a 60.5H mm.ma No.5H m5.:m mm.om mm.mm ea mNfim amass am geese an samaeseazusoa 90690 EM 90690 bx n3wh©£pH3 macho mosam> new: 'HESIS Table % Summary of Analysis of Variance for Size Index Between Withdrawn and Non-Withdrawn Groups Source Between Groups: A Group B Test Order A x B (error b) Within Groups: 0 Test A x C B x C A x B x C (error w) *Significant at .01 level df f: t4 t4 +4 I; I4 an P4 l4 MS 1327.60 119.26 3.01 llO.%3 12.76 5.51 78.85 2.3% 8.99 *12.02 1.08 1.%2 *3.76 22 Summary of Analysis of Variance for Placement Index Table 5 Between Withdrawn and Non-Withdrawn Groups Source Between Groups: A Group B Test Order A x B (error b) Within Groups: C Test A x C B x C A x B x C (error w) *Significant at .05 level df f: F’ F4 l4 f: F4 I4 a: t4 MS 58.59 2.3% 12.53 13.66 .09 .02 1.27 .63 1.62 *%.29 23 2% Table 6 Summary of Analysis of Variance for Motion Index Between Withdrawn and Non-Withdrawn Groups Source df MS F Between Groups: A Group 1 2.0% 1.21 B Test Order 1 3.38 2.00 A x B l .0% (error b) %% 1.69 Within Groups: C Test 1 .0% A x c 1 1.05 1.92 B x C l .0% A x B x C l .37 (error w) %% .60 Table 7 Summary of Analysis of Variance for Detail Index Between Withdrawn and Non-Withdrawn Groups Source Between Groups: A Group B Test Order A x B (error b) Within Groups: C Test A x C B x C A x B x 0 (error w) *Significant at .01 level df f: F4 t4 l4 I: +4 l4 F’ t4 23.99 5.03 1.05 1.81 .36 .05 .13 .16 .12 *13.25 2078 3.00 1.50 1.33 25 THESIS Table 8 Summary of Analysis of Variance for Time Index Between Withdrawn and Non-Withdrawn Groups Source Between Groups: A Group B Test Order A x B (error b) Within Groups: C Test A x C B x C A x B x C (error w) *Significant at .01 level #Significant at .05 level df fHHI—J f: F4 14 F3 +4 MS 178192.67 87k0.17 823.37 3919.63 3k30.0k 340.17 n.19 268.35 585.%6 *19.93 #5.9% l.%% 26 ~IESIS 2. BIBLIOGRAPHY Barman, A., Klein, A. & Lippman, A. Human Figure drawings as a projective technique. g, gpp, £§1§hgl., 1951, 1+5, 57-70. Buck, J. The HTP technique: a qualitative and quantita- tive scoring manual. g, 211p, Psychol., 19%3, %, 317-396. ' ' ' Handler, L. The relationship between GSR and anxiety indices in projective drawings. Unpublished doctor- al dissertation, Michigan State Univ., 196%. Handler, L. & Reyher, J. The effects of stress on the DAP. .g. consult. Psychol., 196%, 23, 259-26%. Handler, L. & Reyher, J. .Figure drawing anxiety indices: a review of the literature. g, p321, Eggh., 1965, 29. 305-313. ' ‘6'" "' Holzberg, J. & wexler, M. The validity of human form drawings as a measure of personality deviation. 9.. 21:21- :29.5. 1950, 11+. 343-361. Hoyt, T. & Baron, M. Anxiety indices in same sex drawings of psychiatric patients with high and low MAS scores. ‘g. consult. Pszchol., 1959, 23, %%3-%52. Machover, Karen Personality projection in ph§_drawing g£_ the may . Springfield, 111.3 Charles ' Themas, l9n9. 27 THESIS 9. 10. ll. 12. 13. 1%. 15. 23 Mogar, R. Anxiety indices in human figure drawings: a replication and extension (extended report). ‘1. consult. Psychol., 1962, 26, 108. Reyher, J. Use of figure drawings in differential diagno- sis. Paper read at the State Diagnosticians' Con- ference, Michigan, 1959. Royal, R. Drawing characteristics of neurotic patients using a drawing-of-man-and-woman technique. .1. glig. Psychol., 19%9, 5, 392-395. I Sanders, R. The clinical applicability of nude figure drawings. Unpublished master's thesis, Michigan State Uhiv., 1965. Swensen, C. Empirical evaluations of human figure drawings. Psychol. £311., 1957, 5%, %3l-%66. Winera B. Jo W W in W design. New Y0rk: McGraw-Hill, 1962. Witkin, H., Dyk, 3., Faterson, H., Goodenough, D. & Karp, S. W W. New York: John Wiley & sens, 1962. f w THESIS "'TITIIIIIIIILIIIIIMlflffljlflflfllflffliifllfl”