Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The relationship between cohesiveness and productivity in small, leaderless, discussion groups Stokes, Rosemary 1954

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1954_A8 S8 R4.pdf [ 8.16MB ]
JSON: 831-1.0107036.json
JSON-LD: 831-1.0107036-ld.json
RDF/XML (Pretty): 831-1.0107036-rdf.xml
RDF/JSON: 831-1.0107036-rdf.json
Turtle: 831-1.0107036-turtle.txt
N-Triples: 831-1.0107036-rdf-ntriples.txt
Original Record: 831-1.0107036-source.json
Full Text

Full Text

THE RELATIONSHIP BETWEEN COHESIVENESS AND PRODUCTIVITY  IN SMALL, LEADERLESS, DISCUSSION GROUPS by ROSEMARY STOKES A THESIS SUBMITTED JM PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n the Department of PHILOSOPHY AND PSYCHOLOGY We accept this thesis as conforming to the standard required from candidates for the degree of MASTER OF ARTS Members of tire Department of Philosophy and Psychology. THE UNIVERSITY OF BRITISH COLUMBIA April, 1954 THE RELATIONSHIP BETWEEN COHESIVENESS AND PRODUCTIVITY IN SMALL, LEADERLESS, DISCUSSION GROUPS Abstract The present study was designed to investigate the relation-ship between cohesiveness and productivity i n small groups. A review of the pertinent literature revealed various approaches to the problem and conflicting findings concerning i t . This study was confined to investigating such a relationship i n a leaderless, discussion-type setting. Groups representative of three degrees of cohesiveness were investigated. The three types of groups were as follows: "Highly structured",-a group i n which each member i s chosen by every other member, and a l l choices are confined within the group. ( A l l choices were made on a sociometric basis.) "Semi-structured",-a group i n which the majority of members choose other members, but the majority of choices are not mutual, and a l l choices are not confined within the group. "Unstructured",-a group i n which no choices occur within the group. On the basis of findings from other studies conducted i n similar settings two hypotheses were advanced: 1. That a group which i s either p a r t i a l l y or highly structured w i l l be more productive i n carrying out a working project than a group which i s relatively unstructured. 2. That a semi-structured group w i l l be more productive i n carrying out a working project than a highly structured group. The study was carried out at a summer camp consisting of seventy-one g i r l s aged twelve to fourteen. From the camp groups of four g i r l s each were chosen to represent the three degrees of cohesive-ness l i s t e d above. These groups were equated as far as possible for age, motivation, and social acceptance. Each group was asked to prepare and present an original skit. Group interaction was observed i n order to gain f u l l e r and more comprehensive data concerning the relationship under investigation. Group productivity was assessed by means of judges1 ratings of the performances. These ratings indicated a direct relationship between cohesiveness and productivity i n this experiment, such that the more cohesive a group the greater i t s productivity. Analysis of the interaction data suggested that productivity was also affected by time and by the type of interaction present, both of which may or may not be related to cohesiveness. Further investi-gation i s indicated. ACKNOWLEDGMENTS The writer wishes to express her gratitude to Dr. Reva Potashin for her many helpful sug-gestions and her constant support and encourage-ment. Thanks are also due to Dr. E. I. Signori for advice concerning design of the experiment, to a l l the faculty members of the Department for their support and suggestions, and to the camp leaders and members, whose friendly cooperation made this study possible. C O N T E N T S Chapter Page I Introduction and statement of the problem 1 I I Theoretical background, review of related 4 research, and discussion of major concepts used Theoretical background 4 Review of related research 6 Discussion of major concepts used 14 Cohesiveness 14 Productivity 15 I I I Experimental subjects, materials, and pro- 16 cedure Subjects 16 Materials 17 Procedure 21 IV Results 26 Data re subgroups 26 Interaction results 26 Judges' ratings 32 (a) Performance 32 (b) Audience reaction 35 (c) Motivation of performers 35 Post-experimental data 36 V Discussion of the results 40 Discussion of performance differences 40 Discussion of interaction assessment 42 Discussion of changes i n motivation and 50 cohesiveness Discussion of the experimental design 52 VI Summary and conclusions 56 VII Implications for further research t 59 References 61 Appendices A Pre-Experimental sociometric questionnaire B Principles for choosing sociometric c r i t e r i a C Pre-experimental motivation scale D Interaction scale E Post-experimental motivation scale F Post-experimental sociometric questionnaire 6 Sociograms of the camp and of the experimental groups H Judges' comments on the performances I Diagrammatic representation of the changes which occurred i n cohesive-ness and motivation of the groups during the experiment T A B L E S Table Page I Age distribution i n camp 16 I I Distribution of the experimental groups with 24 respect to age III Distribution of the experimental groups with 24 respect to motivation IV Frequency and kind of contribution made during 28 practice sessions V-A Percentage and kind of contribution made during 29 practice sessions V-B Approximate percentage and kind of contribution 29 made during practice sessions VI Percentage of contributions of individual members 31 for separate and combined sessions VII Time taken for performance 33 VTII Judges' ratings of performances based on five 33 c r i t e r i a IX Total awarded each performance by each judge 34 X Placement awarded each performance by each judge 34 XI Audience reaction 35 XII Judges' ratings of motivation of groups during 35 performance XIII Sociometric choices based on the criterion of the 37 working project: a comparison of the "pre-" and "post-" scales XIV Reasons given for choices made on "post-" socio- 38 metric scale Motivation indices for members of four experi-mental groups: a comparison of the "pre-" and "post-" scales Difference between "pre-" and "post-" motiva-tion indices A comparison of choices received and amount said by each individual CHAPTER I INTRODUCTION AND STATEMENT OF THE PROBLEM Our modern society is organized into many different kinds of groups of individuals. These groups vary from one another in several respects such as size, purpose, structure, etc. The dif-ferences among groups in such characteristics acquire special signi-ficance as an area for research when i t is recognized that the acti-vity and development of society i s dependent on and limited by the efficiency of the groups within i t . The study of the products and functional processes of groups related to differences in their characteristics i s included in the recently established field of Group Dynamics. This field contains many interrelated problems. In the present study one of these problems is subjected to exploratory investigation. There have developed so far two major research approaches to the problems in the group dynamics field. One emphasizes the rela-tionship between the structure of a group and the quality of its functioning; the other emphasizes the relationship between the nature of the process or interaction within a group and group func-tioning. These two approaches are of course interrelated. For example, i f the structure of a group is emphasized and differences 2 between type of structure and functioning are to be observed, one of the ways of evaluating the la t t e r i s through observation of what differences i n interaction occur. Leadership has and s t i l l does receive much emphasis as one of the important variables for study i n this f i e l d . However this emphasis, important as i t i s , leaves many questions unanswered. There are other variables or factors which also influence the group and require investigation. One of these i s the factor of cohesiveness of the group. There has not as yet been a great deal of s t r i c t l y experi-mental investigation of the relationship between a group's cohesive-ness and i t s functioning although there has been a good deal of speculation and observationally-derived hypotheses about i t . Studies i n this area have been preoccupied with searching out hypotheses and developing appropriate methodological techniques. Consequently most research has had to be exploratory. Such i s also the case with the ; present study, which i s attempting to observe i n a specific type of situation some of the kinds of differences which appear i n group functioning. In the present study three types of groups w i l l be studied, each type representing different qualities of cohesiveness. Through what differences i n group products and processes appear, some aspects of the relationship between cohesiveness and productivity may be indicated. The study was conducted at a summer camp situated a few • miles south of Vancouver. The camp was made up of seventy-one g i r l s , 3 aged twelve to fourteen inclusive, from "whom three types of groups were selected for study during discussion and in performing an assigned task. The groups were selected on the basis of sociometric criteria for defining cohesiveness. Results are based on observation of the groups during the discussion periods and on measures obtained of their performances. CHAPTER II THEORETICAL BACKGROUND, REVIEW OF RELATED RESEARCH, AND DISCUSSION OF MAJOR CONCEPTS USED Theoretical Background The recency of research in group dynamics is logically responsible for the absence so far of any considerable body of induc-tive theory. At the same time, "armchair" or a priori theorizing has been so thoroughly discouraged'within psychological thinking that research workers in group dynamics are l e f t with very few theoretical guideposts. Here is sufficient reason for the variety of approaches in and the exploratory nature of studies in the field. But to this reason must be added at least two more considerations. First, the complexity of the subject-matter of social psychology makes the analysis and assessment of a l l of the important variables in any one social situation a very difficult task. Second, methodological techniques in the field are in a state of flux. New techniques are being devised (e.g., Bales (l), Northway ( 2 4 ) , Carter et al (7)) and the inadequacies of the old ones have led to some revision and refinement (e.g., Guetzkow (14), Schultz ( 2 9 ) ) . 5 Thus the necessity of being exploratory is underlined. Behind the continuing search for appropriate techniques lies the familiar problem of choosing the dimension of the units to be measured. That i s , i s there an "ideal" point of reference situated somewhere between the small and large units represented by "individual" and "group" which w i l l enable the observer to assess both of these extremes without giving an invalid weighting to either? The three most active centers of research i n group dynamics (Harvard, Ann Arbor, Mich., and Tavistock, Eng.), although differing somewhat in their chosen "points of reference", agree on the final importance of conducting research on both the smaller and the larger units. 1' 2 This again tends to increase the difficulty in finding well-developed experimental research in the whole field, including the specific field of group productivity. Apparently organized or traditional theoretical frames of reference have proved inadequate for Group Dynamics, as evidenced by the lack of studies using them. Only two of the studies reviewed used a field theory approach (Thibaut (31) and Schachter et al (28))j while one article exhibited a psychoanalytic orientation (Cohen (10)). The rest seem to be combining various frames of reference and/or 1 For a discussion of this problem by a member of the Ann Arbor group the reader is referred to the introductory section of Lewin's "Frontiers in Group Dynamics11, Part I (19). 2 It i s refreshing to find an emphasis on cooperation rather than competition between these centers. In 1947, for example, a new journal entitled "Human Relations" began publication under the joint editorship of The Tavistock Institute of Human Relations and The Research Center for Group Dynamics. branching out on their own. The.state of confusion in the field has been summed up in the following statement concerning the University of Michigan Conference 3 Research Project: Concerned with evaluating results rather than describing process, i t is faced with the problem of devising appropriate criteria of effectiveness. The search for relationships between the criteria and potentially influential factors has tended to be exploratory, since few theoretical guide-lines were available (30, p.182). Review of Related Research As with most of the earlier studies in social psychology, a large proportion of recent studies concerned with group functioning stress the relationship between leadership and productivity. Many of these studies s t i l l showthe influence of Lewin's early investigations of autocratic and democratic leadership (20, 21), (23, p»49l)« In general they tend to corroborate his findings that "participatory" leadership is more effective than "supervisory" leadership (e.g., Hare (15), Preston and Heintz (26)). Weschler, Kahahe and Tannenbaum, however, found that group productivity was higher under a "restrictive" leader than under a "permissive11 leader (32). It is hard to estimate how contradictory these results are to the above, since the authors do not define what 3 The U. of Mich. Conf. Res. Proj. is one of the most comprehensive pieces of research yet undertaken in the field. Al-though i t is not yet completed and i t s findings have not been published, i t has been reported in the "Annual Reviews" (18, 30). 7 they mean by "restrictive" and "permissive". Studies which have emphasized relationships between '.other variables than leadership and productivity include the following: Coch and French (9) found that increased group participation in goal-setting tends to increase group productivity by overcoming competing influences in the situation. Their findings were corro-borated by a series of studies edited by Bradford and French (5). Rosenthal and Cofer (27) have indicated how group cooperation and motivation can be decreased by the non-participation of a "planted" member of the group. Finally, Berkowitz (3) found that group cohesiveness and satisfaction decrease when leadership is shared by other members than the nominal leader. Four studies concerned primarily with the relationship between cohesiveness and productivity will be more fully described here. To the best of the author's knowledge, two of these (French (12), Darley et al (28, p.230)) have not yet been published. Con-sequently the information about them is incomplete. The four studies in this area have been carried out in a variety of settings using groups of differing sizes. These differen-ces, coupled with differences in experimental procedure have resulted in correspondingly inconsistent conclusions. Goodacre (13) finds a 4 Smith (30, p.183) has emphasized the danger of any attempt to generalize from these findings on the effects of leadership to larger groups, that is to groups i n which the interaction is no longer "face-to-face". The caution which he advocates may very well be applied to a l l experimental data arising from small group studies. 8 direct linear relationship between degree of cohesiveness and group productivity. Schachter et a l (28) found "no necessary relationship" between the two variables. French (12) and Barley et a l (28, p.230) do not find the same kind of relationship suggested by Goodacre. I t i s not clearly stated, however, what kind of relationship, i f any, these studies did find. A brief description of the experimental design used i n each of these four studies w i l l demonstrate the differences i n procedures and results. Goodacre (13) worked i n a military setting. He used twelve already established combat teams each of which was composed of four members plus two formal leaders. The degree of cohesiveness for each of the teams was determined sociometrically. The experimental situation chosen by Goodacre consisted of a f i e l d examination of each of the selected groups i n combat maneuvers. Goodacre's groups were equated for sex and previous military training. Ratings made by military officers and by members of the Institute sponsoring the study were the c r i t e r i a of effectiveness. Using these measures he obtained a correlation of +.77 between group cohesiveness and group performance. French's study (12) was also carried out i n a military setting. Sociometric techniques were used to assess the cohesiveness of sixteen companies of about sixty naval cadets each. French intro-duced a variety of working projects': company d r i l l i n g , inspections, academic performance, athletics, and community participation. Again the groups were equated on the basis of sex and of previous military 9 training, and no assessment of interaction has been reported. French had hypothesized a direct relationship between com-pany cohesiveness and company ratings on the various projects; a l l that could be found concerning the results he obtained was the state-ment that the anticipated relationship was not upheld (28, p.230). Darley, Gross, and Martin (28, p.230) defined cohesiveness sociometrically but the size of their groups i s not known. Neither i s i t known for what variables (including leadership) they equated the groups, or i f they assessed interaction. Their working project was a group-written essay, an act i v i t y of a more sedentary nature i n which group discussion was probably foremost. Darley et a l had been working from a similar hypothesis as French's. Again this hypothesis was not upheld. The study by Schachter, Ellertson, McBride, and Gregory (28) differs i n several respects from the previous three.^ Schachter et a l were interested i n an indirect relationship between cohesiveness and productivity; they studied the relationship between group cohesiveness and reception of group "induction" to either speed up or slow down production.^ Using groups composed of three strangers each, they super-imposed either high or low cohesiveness by suggestion from false "test" 5 This study i s interesting both for i t s results and be-cause i t presents both the consistencies and the limitations arising from the use of one particular theoretical framework (in this case a f i e l d theory approach). 6 Group induction i s used here apparently to mean the i n -ducement or influencing of the group members by each other through group interaction to modify certain of their attitudes towards the working project. 10 results. The groups were equated on the basis of sex and of educational background. The working project used was a simple motor task: cutting out pieces of cardboard to a specified size. Group interaction was not "face-to-face" but took the form of written notes, and by this means either positive or negative i n -duction was introduced. Schachter et a l had hypothesized that for both positive and negative induction the high cohesive groups would be more successful i n influencing their members than the low cohesive groups. Results upheld this hypothesis only under conditions of negative induction. A l l of these studies have certain things i n common. They are using formal groups and emphasizing motor tasks7 while minimizing emphasis upon observation of group interaction. There are some studies which suggest that given a different type of organization and a different type of task there i s not neces-s a r i l y a one to one relationship between cohesiveness and productivity. In 1937 Bos published a study (4), based on earlier work by Joubert (17), which was concerned with the relative productiveness of individuals and of groups of two. Bos suggests from her results that a f a i r l y intensive i n -vestigation of the interaction within each group i s essential to an intelligent appraisal of the relationship between cohesiveness and 7 With the exception of the study by Darley et a l . 11 productivity. During her observations of the interaction of children i n a leaderless, discussion-type setting Bos noted: ...that a too l i v e l y personal interest of the children i n each other was a stumbling block i n the way of f r u i t f u l work being done to-gether, the fact being that the contact which stimulates mental activity and intensity must be rooted i n the task"in hand and not i n the person. I f personal interest dominates, the attitude necessary for the work i s interfered with (4, p.415). Support of this particular finding by Bos i s contained i n an unpublished paper by Carter and Couch (16, p.280). In this study factor analysis was applied to ratings of behavioural t r a i t s of group members. Three factors were identified and labelled Group Goal Facilitation I, Striving for Individual Prominence I I , and Group Sociability I I I . With respect to this last factor the authors found that: Sociable behavior—Factor III—tends to reduce group motivation and competition, but to increase group talkativeness, friendliness, and interest i n social interaction (16, p.283). suggesting that the more cohesive the group the more interaction i s l i k e l y to occur. The third study which presents pertinent findings was centered 8 Several of the observations made during group interaction by Bos are in close agreement with those made i n the present study. See "Discussion of Interaction", Chapter V. 12 around an investigation of children's friendships by Potashin (25)• This study differentiates between two types of the mutual friendship relationship: (1) A"closed" or "clique" type of relationship i n which the sociometric choices of two mutual friends are largely identical. (2) An "open" type of relationship i n which the sociometric choices of two mutual friends d i f f e r . The author found that mutual friends of the f i r s t or "closed" type of relationship did not contribute as much in a discussion setting as mutual friends of the second or "open" type of relationship, whose interests are presumably more diverse. The findings from these three studies, a l l involving the use of discussion groups and some interaction analysis, are not especially congruent with the studies described previously. As has already been mentioned, the latter did not make use of interaction analysis. Again, Schachter et a l , whose groups were composed of strangers, studied only the two extremes of high and low cohesiveness. Goodacre and French both used highly organized, m i l i -tary settings. There i s at the same time some incertitude arising from the failure of French to corroborate Goodacre's findings.^ The present study, while combining elements from both types of study to some extent, v a i l generally follow more closely the ap-proach used by Bos, Carter and Couch, and Potashin. This study i s interested i n observing group interaction as an aid to exploring the relationship between cohesiveness and productivity. I t would also l i k e to avoid the complicating factor of formal leadership, which i s 9 This may be an unfair comparison because of the d i f f e r -ence i n size of group used i n the two studies. 13 present i n both French's and Goodacre's studies, and to work with more than just the extremely high and extremely low degrees of cohesiveness, as did Schachter. The kind of group used by Bos, Carter and Couch, and Potashin (leaderless, discussion groups) w i l l be used i n the present study because the verbal, sedentary nature of the interaction i n this type of group i s adaptable to observational analysis, particularly since Bales 1 technique of interaction analysis—dealing with this type of group—has been published ( l ) . Using a leaderless, discussion-type setting the present study w i l l attempt to compare the relative effectiveness of three types of groups: (1) Highly structured - a higly structured group w i l l be 10 defined as a group i n which each member i s chosen by every other member, and a l l of the choices are confined within the group (i.e., a "clique"). (2) Semi-structured - a semi-structured group w i l l be de-fined as a group i n which the majority of the members choose other members, but the majority of choices are not mutual, and a l l choices are not confined within the group. (3) Unstructured - an unstructured group w i l l be defined as a group i n which no choices occur within the group, Using these three types of groups two hypotheses w i l l be tested: 10 A l l choices w i l l be made on a sociometric basis. 14 (1) That a group which i s either p a r t i a l l y (A) or highly structured (B) w i l l be more productive i n carrying out a working project than a group which i s instructured (C). (2) That a semi-structured group (A) w i l l be more productive i n carrying out a working project than a highly structured group (B). The f i r s t hypothesis was chosen i n order to test what seems to be an unwritten assumption i n a l l of these studies that groups possessing l i t t l e or no cohesiveness w i l l be less productive than any other group(s).^" The second hypothesis i s based on suggestions from the results of Bos' and Potashin's studies, as set forth earlier. Combining these two hypotheses, i t i s hypothesized that i n terms of productiveness A>B>C. Discussion of Major Concepts Used Cohesiveness Judging from studies i n the f i e l d there are different ways of defining cohesiveness. Some have used a sociometric basis. In this case cohesiveness necessarily becomes the product of the degree of attraction or "tele" between members of the group. Others prefer units which emphasize the group as a whole. For example, Schachter et a l (28) define cohesiveness as group attractiveness for i t s members; Thibaut defines i t as "the t o t a l f i e l d of forces which acts on members 11 There i s some evidence i n the study by Potashin that non-friends are not as productive i n a discussion setting as friends (25, P-49) 15 to remain i n the group" (31, p.25l)« The two approaches may not be as different as they appear. However the use of sociometric techniques allows a certain amount of objectivity and does allow measurement, which i s s t i l l very d i f f i c u l t to attain using other approaches. Productivity The seeming looseness of usage associated with this term probably arises from a continuing i n a b i l i t y to dissociate i t from the particular setting involved. To i l l u s t r a t e : productivity as found i n this chapter could be defined as the end-result of comparative judgements, based on previously set-up c r i t e r i a , of selected conse-quences of interaction. This definition brings out two other related points. F i r s t , i n actual practice a concept becomes the product of the assessment technique(s) used. Second, the major technique so far developed re productivity has been judges' ratings. The above definition, with i t s implied problems, i s appli-cable to the present study. CHAPTER I I I EXPERIMENTAL SUBJECTS, MATERIALS, AND PROCEDURE Subjects The setting i n which the experiment was carried out was a ten-day summer camp situated a few miles south of Vancouver. The camp consisted of seventy-one1 g i r l s aged twelve to fourteen years inclusive. The age distribution i n the camp i s shown i n Table I. TABLE I AGE DISTRIBUTION IN CAMP Age Frequency Percentage 12 31 43.66 (44) 13 18 25.35 (25) 14 22 30.98 (31) Sixteen of the campers were chosen as subjects for the experiment, making up four groups of four g i r l s each. 1 The original number was seventy-two, but one of the gi r l s had returned home prior to commencement of the study. 17 Materials For purposes of selecting possible experimental groups a sociometric questionnaire was designed (see Appendix A). This questionnaire consists of four c r i t e r i a with three choices allowed on each criterion, making a t o t a l of twelve choices for each subject to make. The c r i t e r i a used i n this questionnaire were chosen i n ac-cordance with principles l a i d down by Brofenbrenner (6) and Northway (24)• In Appendix B these principles and their application i n this study have been outlined. Because the primary purpose of the questionnaire was to yield data concerning the friendships established i n the camp i t was f e l t that relatively general c r i t e r i a would be more appropriate than c r i t e r i a which emphasize any particular s k i l l ( s ) . On this basis the f i r s t three c r i t e r i a may be differentiated from the fourth one. The f i r s t three have been made as general as possible; the fourth i s directed specifically to a particular a c t i v i t y which involves several of the " s k i l l s " (e.g., acting a b i l i t y , imagination, and perhaps more) that are related to the experimental task. The results from only the f i r s t three c r i t e r i a were used i n selecting the experimental groups. Data based on the fourth criterion were used later i n order to assess changes in group structure which took place during the experiment (see Chapter V). With respect to the number of choices allowed for each criterion, the decision was an arbitrary one.2 Three choices per 2 Northway points out the lack of research needed to help determine the most appropriate number of both c r i t e r i a and choices to be used i n any given situation (24, p-3)« 18 criterion was eventually decided upon (making a to t a l of nine possible choices on the general criteria) because this number i s neither so large as to be clumsy nor so small as to distort the results by pre-venting the recording of a l l of the subject's major choices. Further, previous studies have suggested that results are not changed s i g n i f i -cantly by freedom to choose beyond this level. The i n i t i a l motivation of each camper with respect to the experimental project was assessed by means of an attitude-type scale. More specifically, i t i s an adaption of the Likert type of attitude scale (11, p.372). A copy of the scale may be found i n Appendix C. I t w i l l be seen that i n this scale the experimental project has been broken down into the three major sub-activities involved i n i t . Values ranging from +2 to -2 were assigned to the five possible choices allowed for each act i v i t y , and each subject's index of moti-vation was derived from the algebraic sum of the choices made. In this way the experimenter sought to gain a more detailed and graded appraisal of i n i t i a l motivation than would have been gained by leaving the scale i n a simpler form and the working project whole. Results of the motivation scale w i l l be i'used as one index of equation of the experimental groups, and also as a criterion to assess changes occur-ring during the experimental process. In order to assess the interaction of the experimental groups a "sieve" type recording scale was devised i n preference to a more elaborate, verbatim system. That i s , only certain specific types of interaction were to be recorded, rather than everything which occurred during group discussion. 19 The scale used i n this study i s based on the type of scale developed by Bales ( l ) . I n i t i a l l y the following five c r i t e r i a were selected: (1) Introduces inside idea^ (2) Introduces outside idea^ (3) Wanting to do a good 30b (4) Evidences of competition (5) Rejects idea ot plan During the course of the f i r s t observation of a discussion period a sixth criterion was added: (6) Irrelevant remarks A copy of the scale may be found i n Appendix D. Three judges (S, T, and U) were instructed to rate the per-formances of the experimental groups on five c r i t e r i a : (1) originality (2) best ideas (3) best presentation (smoothness, etc., of performance, not individual acting s k i l l ) (4) probable degree of thought and planning (5) overall degree of entertainment 3 Included here would be a l l ideas which originated from within the group. 4 Included here would be a l l ideas which originated from outside the group, usually from discussion between a member and a non-member outside the formal discussion period. 20 Three more judges (V, W, and X) rated motivation of the per-formers on three c r i t e r i a : (1) enthusiasm of participants (2) desire to do well (3) concentration on the performance Two judges (Y, and Z) rated audience reaction to the perfor-mances on the basis of applause. I t was originally intended that individual interviews would be used to gather a l l of the data concerning any changes i n motivation and cohesiveness following the performances. However a crowded camp schedule and a lack of time forced the substitution of the two ques-tionnaires plus appended questions described below for the less "ob-jective" but possibly more f r u i t f u l interviewing technique. Any major changes i n the motivation of the experimental groups towards the working project were determined from the results of a condensed version of the pre-experimental motivation scale (see Ap-pendix E). The question which was appended to this scale was designed to allow the subject to elaborate on her attitude toward the working project. Additional space was l e f t for any other comments and the subjects were encouraged to make use of i t . A f i n a l , sociometric-type questionnaire was administered to the members of the experimental groups. This questionnaire (see Appendix F) made use of one criterion and three possible choices. The criterion used was the same as the fourth criterion on the pre-experimental scale, i.e., the working project. Again a question 21 was appended which was designed to encourage the subject to l i s t the reasons for the choices given. In a l l cases the author i s convinced that most, i f not a l l of the campers were sincerely cooperative throughout. Procedure In order to select groups from the camp to use i n the experiment proper the sociometric questionnaire described i n the pre-ceding section was given to the seventy-one camp members. I t was given four days after the camp session began. A longer time lapse than four days would have been desirable, but the t o t a l length of the camp session made further postponement impossible. The results from this sociometric questionnaire were then analyzed i n order to discover the existing relationships within the camp and so select groups to represent the three types of cohesiveness to be studied. The f i n a l selection out of a l l the possible structured and semi-structured groups rested on how closely these groups were equated on the basis of number of members, motivation of the members towards the experimental task, the age, and the social acceptance levels of the group. The size of group decided upon was four members, since that was the size of the possible cliques. Groups which were concentrated either close to the centre of the sociogram (most socially accepted), which had been constructed 22. on the basis of the sociometric questionnaire, or close to the outside of the sociogram (least socially accepted)were eliminated.-* The motivation of a l l possible groups towards the experimental task had previously been determined on the basis of the attitude scale, described i n the preceding section, which was given immediately after the sociometric questionnaire.^ The equation of possible examples of each type of group on the basis of motivation and age was more d i f f i c u l t than i t had been for 7 size and social acceptance-Of the two most l i k e l y cliques, one was markedly older, and more interested than the most l i k e l y semi-structured and unstructured groups; while the other was markedly younger and less interested (see Tables I I and I I I ) . Such differences would increase the d i f f i -culty of evaluating the productiveness of the various types of groups. The inclusion of both cliques, i n order to avoid such ambiguity, necessitated the use of four rather than three experimental groups: two examples of a structured group and one example each of a semi-structured and unstructured group. In order to simplify this and the following sections, each of 5 The term "social acceptance" i s used here to denote the to t a l number of choices any one g i r l received from the remaining seventy g i r l s , i.e, her "popularity rating". 6 The two scales were administered i n this order so that the choices made on the sociometric questionnaire would not be influenced by speculations which might be aroused by the motivation scale. 7 Because of the setting, chosen a l l group members were pre-equated on the basis of sex. Although any results w i l l be correspon-dingly limited, the confusion of factors involved i n the use of a coeducational setting has been avoided. J 23 the experimental groups w i l l be represented by a letter which i s i n agreement with the letters used i n the hypotheses i n Chapter I I , thus: Group A The semi-structured group Group B]_ The "weaker" clique Group B2 The "stronger" clique Group C The unstructured group The working project which was assigned to each of the experi-mental groups involved the preparation and presentation of an original s k i t , i n which each group member must take part, and which must be entitled "A Pleasure Cruise". Such a project necessitates a consider-able amount of discussion; i t s interest value i s high, and although skit presentation i s a familiar project i n a camp setting, the fact that these skits must be original made the discussion part of the task a novel one. (In the leaders' experience, camp skits at this age level are seldom original.) The frequency of skit presentation i n camp l i f e ensured that a l l members of the groups were experienced i n the non-discussion part of the project. Each of the four groups was allowed two periods of fifteen minutes each on two successive days i n which to prepare their enter-tainment. During each of the practice sessions the interaction of the groups was observed and recorded by means of the interaction scales described earlier. In addition to the quantitative data so gathered certain qualitative observations were made and these w i l l also be considered i n the discussion of the results. 24 TABLE I I DISTRIBUTION OF THE EXPERIMENTAL GROUPS WITH RESPECT TO AGE Group Age Frequency Average Age 12 1 A . 13 2 13-0 14. 1 12 3 B l 13 0 12.5 14 1 12 0 *2 13 1 13-75 04 3 12 1 C 13 2 13.0 14 1 TABLE I I I DISTRIBUTION OF THE EXPERIMENTAL GROUPS WITH RESPECT TO MOTIVATION Group Member+ Motivation + + Average Motivation 36 4 A 12 59 4 4 4 26 4 14 2 % 13 11 4 2 2.5 39 2 1 5 B 2 20 56 5 5 5 18 5 46 5 C 34 29 4 4 4.25 57 4 + Each camper was assigned a number at the outset of the study and w i l l be referred to by that number throughout. ++Indices of motivation range from 0 to 5; 0 = lowest possible motivation, 5 = highest possible motivation. 25 At the conclusion of the practice sessions each skit was presented before the assembled camp. A t o t a l of eight judges was used: three to rate independently the quality of the performance, three to rate independently the motivation of the performers, and two to rate independently audience reaction. The following day each member of the experimental groups was given two f i n a l questionnaires designed to discover any changes i n motivation and cohesiveness respectively which had occurred during the experiment. The complete procedure lasted six days i n a l l • CHAPTER IV RESULTS Data re Subgroups Results from the three general c r i t e r i a of the pre-experi-mental sociometric questionnaire have been summarized i n the sociogram i n Appendix G, p.S3. The age, social acceptance, and motivation to-wards the working project of each camp member, as well as the inter-relationships among a l l of the g i r l s at the outset of the study are depicted i n this sociogram. In addition.,.;, a separate sociogram was drawn up for each of the experimental groups (Appendix G, pp.79-82). It would be interesting to see i f there i s any relationship between 11 enthusiasm" and "popularity". Accordingly a correlation .coefficient was computed between each g i r l ' s degree of motivation towards the experimental task and her social acceptance, as measured by the three general sociometric c r i t e r i a . The resulting c o e f f i c i e n t — computed according to the method described by Lindquist (22, p.l66-167)—is very'close to zero: +0.1253, or +0.13 • Interaction Results Data gathered during observation of the interaction within the experimental groups were analysed i n terms of reaction time, 27 characteristics of group interaction, and individual participation. When the working project was presented to each of the four groups at the beginning of their f i r s t sessions their "reaction times" varied somewhat. Unfortunately the actual number of minutes taken by each group to begin speaking was not recorded; however the order was B 2 < A < B]_ < C. When analysis was begun of the different types of remarks made during interaction, the si m i l a r i t y of two of the c r i t e r i a used i n the interaction scale ("Wanting to do a good job" and "Evidences of competition") prompted the combining of the two sets of data under one new heading (Number of "Boosters"). Since the sources of the ideas introduced i n the discussions were seldom mentioned, data gathered under-"Introduces inside idea" and under "Introduces outside idea" had also to be regretfully combined under the one heading (Number of Ideas). Therefore f i n a l analysis of the data from the interaction scale w i l l be based on the four headings: Boosters, Ideas, Rejections, and Irrelevancies. Tables IV and V-A present the total number of contributions made in each group for each of the practice sessions and indicate what proportions of the various types of contributions make up these totals. Table V-B i s essentially a repetition of Table V-A, but the percentages have been rounded i n order to f a c i l i t a t e an examination of the various trends which are present. I t w i l l be noted from these Tables that there are no out-standing differences between the groups i n the distribution of the TABLE IV FREQUENCY AND KIND OF CONTRIBUTION MADE DURING PRACTICE SESSIONS Group Session Boosters Ideas Rejections Irrelevancies Total 1 0 ) 37 ) 5 ) 7 ) 49 ) A ) o ) 79 ) 14 ) 9 ) 102 2 o ) 42 ) 9 ) 2 ) 53 ) 1 2 ) 50 ) 5 ) 2 ) 59 ) h ) 2 ) 55 ) 6 ) 2 ) 65 2 0 ) 5 ) 1 ) 0 ) 6 ) 1 1 ) 66 ) 4 ) 2 ) 73 ) B 2 ) 1 )109 ) 9 ) 2 ) 121 2 0 ) 43 ) 5 ) 0 ) 48 ) 1 0 ) 7 ) 1 ) 0 ) 8 ) C ) o ) 17 ) 1 ) o ) 18 2 0 ) 10 ) 0 ) 0 ) 10 ) 29 TABLE V-A PERCENTAGE AND KIND OF CONTRIBUTION MADE DURING PRACTICE SESSIONS Group Session Boosters Ideas Rejections Irrelevancies 1 0.0 ) 75.51) 10.2 ) 14.28 ) A )0.0 )77.45 )13.72 )8.82 2 0.0 ) 79.24) 16.98) 3.77 ) 1 3.38) 84.74) 8.47) 3-38 ) h )3-07 )84.6l ) 9.23 )3.07 2 0.0 ) 83.33) 16.66) 0.0 ) 1 1.36) 90.41) 5.47) 2.73 ) B2 )0.82 )90.08 )7-43 )1.65 2 0.0 ) 89.58) IO.41) 0.0 ) C 1 0.0 ) 87-5 ) 12.5 ) 0.0 ) )0.0 )94-44 )5.55 )0.0 2 0.0 ) 100.0 ) 0.0 ) 0.0 ) TABLE V-B APPROXIMATE PERCENTAGE AND KIND OF CONTRIBUTION MADE DURING PRACTICE SESSIONS Group Session Boosters Ideas Rejections Irrelevancies 1 0 ) 76 ) 10 ) 14 ) A ) 0 ) 77 ) 14 ) 9 2 0 ) 79 ) 17 ) 4 ) 1 3 ) 85 ) 8 ) 3 ) B l ) 3 ) 85 ) 9 ) 3 2 0 ) 83 ) 17 ) 0 ) B 2 1 1 ) 90 ) 5 ) 3 ) ) 1 ) 90 ) 7 ) 2 2 0 ) 90 ) 10 ) 0 ) 1 0 ) 88 ) 13 ) 0 ) C ) 0 ) 94 ) 6 ) 0 2 0 ) 100 ) 0 ) 0 ) 30 various categories of interaction. However differences do appear re intensity" of interaction as indicated by this data. Very l i t t l e interaction took place i n Group C, compared to the other three groups. I t w i l l also be noted in Table IV that while the t o t a l number of comments decreases from the f i r s t to the second session for the two cliques, i t increases for groups A and C.**-In order to examine the relative contribution made by each member within a group, the percentage of contributions for each member for each session and for the two sessions combined has been computed and w i l l be found i n Table VI. Since the present study specifies that groups to be studied shall be leaderless, i t i s of importance to note whether this condition i was violated by the appearance of any "unofficial 1 1 leaders during group interaction. I f leadership i s defined on the basis of the amount said during group discussion, 2 and a r b i t r a r i l y set at > 50%, then no such unofficial leaders did appear, with the possible exception of individual 46 i n group C. 1 Group Bj_ informed the author of four rehearsals which they had held between the two formal sessions, and their rehearsal during the second session was somewhat "finished" or stereotyped. Group B2 rehearsed also between sessions, but not as frequently (only twice). Their rehearsal during the second session did not start u n t i l half-way through and was not so stereotyped. Groups A and C got together twice and once respectively between sessions but did not reach the rehearsal stage before Session I I . For a discussion of some of the implications involved see Chapter V. 2 Other definitions are of course possible. 31 TABLE VI PERCENTAGE OF CONTRIBUTIONS OF INDIVIDUAL MEMBERS . FOR SEPARATE AND COMBINED SESSIONS Group Member Session 1 Session 2 Sessions 1 and 2 B l B 2 36 30.95 (3D 27-45 (27.5) 29.03 (29) 12 45.23 (45) 37.25 (37) 40.86 (41) 59 4.76 (5) 13.72 (14) 9.67 (10) 26 19.04 (19) 21.56 (22) 20.43 (20) 14 22.807(23) 16.66 (17) 22.22 (22) 13 45.61 (46) 33-33 (33) ' 44-44 (44) 11 7.01 (7) 0.0 (0) 6.34 (6) 39 24-56 (25) 50.0 (50) 26.98 (27) 1 19-71 (20) 27.08 (27) 22.68 (23) 20 29-57 (30) 14.58 (15) 23.52 (23.5) 56 15.49 (15.5) 12.50 (12.5) 14.28 (14) 18 35.21 (35) 45.83 (46) 39.49 (39-5) 46 62.50 (62.5) 40.0 (40) 50.0 (50) 34 0.0 (0) 0.0 (0) 0.0 (0) 29 12.50 (12.5) 40.0 (40) 27.77 (28) 57 25.0 (25) 20.0 (20) 22.22 (22) 1 32 Judges' Ratings (a) Performance Only three, instead of four, skits were presented before the camp at the conclusion of the practice sessions. Group C, the unstructured group, did not succeed i n preparing a s k i t , and at the conclusion of their second practice session they asked to be allowed to withdraw. Therefore the results pertaining to judgements of the performances w i l l deal only with three groups: A, BQ_, and Table VII gives the time taken for each performance. As mentioned i n Chapter I I I , the performances were rated independently by three judges on the basis of five c r i t e r i a which stressed the quality of the skit material rather than acting a b i l i t y . The judges' ratings of the performances are set forth i n Table VTII and summarized i n Tables IX and X. According to Tables IX and X, there i s no disagreement between the judges on placing group A third, or la s t . The decisions with respect to groups and B 2 are consistent for Judges S and U. The difference exhibited by Judge T's ratings should be considered in the light of her statement that she could not see a large part of group B2's performance. A l l three judges were asked to comment on the performances, and the interested reader may find these comments i n Appendix H. I t i s noteworthy that both Judges S and U mentioned that group B2 (the "strong" clique1) displayed more confidence than the other two groups.-3 The members of this group were superior to the other groups members i n terms of age (Table I ) . 33 TABLE VII TIME TAKEN FOR PERFORMANCE Group Time A Bo 6 minutes 10 » 6 » TABLE VIII JUDGES' RATINGS OF PERFORMANCES BASED ON FIVE CRITERIA Criterion Judge S: 1st 2nd 3rd Judge T: 1st 2nd 3rd Judge U: 1st 2nd 3rd 1. Originality B2 B l A B2. A % B 2 - A BJ 2. Best Ideas B2 B l A B l B2 A B2 \ A 3- Best Presentation B l B 2 A B l B 2 A B 2 B3_ A 4. Most Thought and Planning B l B2 A B l B 2 A B! B 2 A 5. Overall Degree of Entertainment B2 % A B l B 2 A B 2 A B± Overall Rating B 2 B l . A Bl B 2 A not given II very close" TABLE IX TOTAL AWARDED EACH PERFORMANCE BY EACH JUDGE Group Judge s , Judge T Judge U Total A 5 6 7 18 B l 12 13 9 34 B 2 13 11 14 38 TABLE X PLACEMENT AWARDED EACH PERFORMANCE BY EACH JUDGE Group Judge S Judge T Judge U A -v Third Third Third B l Second First Second B2 First Second Firs t 35 (b) Audience Reaction Audience reaction, as indicated by applause, was judged independently by two judges. Their findings, * which are i n agreement with the above results, may be found i n Table XI. TABLE XI AUDIENCE REACTION Applause First Judge Y B 2> B^y A Second JudgeZ B 2> Bj_ > A (c) Motivation of Performers Judges V, ¥, and X, who had been asked to rate the motivation of the performers a l l expressed concern over the d i f f i c u l t y they had experienced i n trying to evaluate this quality. This d i f f i c u l t y seemed to arise partly because a l l of the groups were well motivated and partly because of the lack of a clear-cut understanding of the thing to be judged. TABLE XII - JUDGES1 RATINGS OF MOTIVATION OF GROUPS DURING PERFORMANCE ~~~~ Judge Judge Judge Group v w x A Second- First & BT Third Second • r-i B 2 First Third ^ 36 Post-Experimental Data The results from the two questionnaires administered to members of the experimental groups at the conclusion of the study-w i l l have more meaning i f they are compared with similar measures taken at the beginning of thi s study; therefore they are presented in that form i n Tables XIII, XV, and XVI. The same data have been compiled diagrammatical1y and may be found i n Appendix I. Several interesting changes can be seen i n this data. Among the more obvious i s the increase i n cohesiveness which occurred i n a l l of the groups except C, and the decrease i n motivation for both the semi-structured and unstructured groups. As mentioned i n Chapter I I I , a question was appended to the "post-" sociometric questionnaire: "Why did you choose these three g i r l s ? " The answers to this question have been arranged according to general type of and frequency of reason given i n Table XIV. 37 TABLE XIII SOCIOMETRIC CHOICES BASED ON THE CRITERION OF THE WORKING PROJECT A COMPARISON OF THE "PRE-" AND "POST-" SCALES Group Individual Chosen by number of the group prior to experiment Chosen by number of the group after experiment A 36 12 59 26 1 1 0 2 2 2 3 0 B l 14 • 13 11 39 3 3 1 3 2 3 3 3 . B2 1 20 56 18 3 0 2 3 3 3 3 3 C 46 34 29 57 0 0 0 0 0 0 0 0 38 TABLE XIV REASONS GIVEN FOR CHOICES MADE ON "POST-" SOCIOMETRIC SCALE Group f z t Reason A 1 Bl 1 6 imagination (or "good ideas") B2 11 C 11 A 0 BL 11 5 friendship ("get along well together", B2 11 etc.) C 1 A 0 Bl 11 5 acting ability-B 2 0 C 111 A 1 Bl 1 3 generalized a b i l i t y to do well i n B 2 0 putting on a skit C 1 A 1 Bl 0 2 cooperation B 2 1 C 0 A 11 Bl 0 2 dependability B2 0 C 0 A 0 B l 0 2 success of previous mutual efforts B2 11 D' 0 A 0 Bl 0 1 enjoy putting on a skit B 2 0 D 1 N = 27 N(people) = 1 6 39 TABLE XV MOTIVATION INDICES FOR MEMBERS OF FOUR EXPERIMENTAL GROUPS A COMPARISON OF THE "PRE-" AND "POST-" SCALES "Pre-" "Post-" Group Individual Motivation" A v e r a S e Motivation" A v e r a g e 36 4 12 4 59 4 26 4 34 2 13 4 11 2 39 2 1 5 20 5 56 5 18 5 46 5 34 4 29 4 57 4 3 3 2 3 4-25 | 2.5 2 2.75 5 %  2 , 5 4 3 , 5 3 5 5 B 2   5 5 5 5 2 ++ 5 = maximum motivation; 1 = minimum motivation TABLE XVI DIFFERENCE BETWEEN "PRE-" AND "POST-" MOTIVATION INDICES Sum of "Pre-" Sum of "Post-" Difference ^ o u p Motivation Motivatibn ( "post-" - "pre-") A 16 11 -5 &L 10 14 +4 B 2 25 25 0 C 17 10 -7 CHAPTER V DISCUSSION OF THE RESULTS Discussion of Performance Differences That there i s a relationship between cohesiveness and produc-t i v i t y i s suggested on a gross level by the fact that the unstructured group i n this study was non-productive. Of the groups who did pro-duce skits , judges' ratings place the two cliques ahead of the semi-structured group; this suggests that the more cohesive the group, the greater i t s productivity. This suggestion i s reinforced by the high degree of agree-ment between the three judges. As indicated i n Table X a l l three judges' ratings of the performances agree, with the exception of the placements awarded groups and B 2 by Judge T. This discrepancy becomes less significant when i t i s recalled that this judge was unable to see a good deal of group B2's performance. The same trend i s shown by the ratings of audience reaction, which are i n agreement with the results of the performance ratings.^ 1 The audience was not aware that the emphasis i n judging the performances was on quality of the skit material rather than on the dramatic performance of same. Therefore this agreement might be par t i a l l y due to the fact that those groups having the better sk i t material also gave the better dramatic performance, or i t might indicate that dramatic performance did influence the judges. u The trend shown so far (i.e . , that the most cohesive groups are the best producers) may be accounted for by differences i n cohesive-ness, but i t might also be the result of other differences between the experimental groups (e.g., age, etc.) for which perfect equation was not possible. Since the groups were equated for sex and number of members, differences i n these variables could not account for the above trend. The groups were not perfectly equated on the.basis of social acceptance. However on the basis of inspection of the sociogram they were generally equated so that with one exception*' none of the members of the groups l i e s within either the extreme outer or the extreme inner c i r c l e on the sociogram (see Appendix G). Therefore since there i s no great variation with respect to this variable, differences i n performance cannot be accounted for here either. I f differences i n productivity were due to differences in age, then the quality of the performances should have been i n the order B2 > C = A > BQ_ (assuming that age of performers and quality of per-formance i s positively correlated i n this age group). Since the performances were not rated i n that order then i t i s concluded that i n this study cohesiveness i s a stronger factor than age. Similarly, i f differences i n productivity could be accounted for by differences i n pre-experimental motivation then the performances would have been i n the order B2 > C > A > B]_. Except for group B]_, differences between the groups for pre-experimental motivation are 2 Individual 46 i n group C l i e s i n the extreme outer c i r c l e . 1 42 small. Therefore, i f this were the determining variable then group Bi would have done very poorly. Since this did not happen then cohesiveness i s a stronger factor than pre-experimental motivation. However given groups of relatively equal cohesiveness (e.g., groups B]_ and B2), their performances do di f f e r . Therefore under these circumstances age and motivation do make a difference. Thus the clique which was weaker for cohesiveness but stronger for both age and motivation (Bo) did better than the clique which was stronger for cohesiveness but weaker for age and motivation (Bi). Therefore, taking a l l of the factors so far mentioned into account, there appears to be a general trend such that increasing productivity i s accompanied by increasing cohesiveness, i n this type of situation.^ Discussion of Interaction Assessment The following discussion may be roughly divided into two sections on the basis of the kind of material used in each. The f i r s t of these two sections i s based on the data gathered by means of the c r i t e r i a for interaction assessment. The second of the two sections i s based on some qualitative observations made by the experimenter while gathering the above quantitative data. The two sections are complementary. Table IV indicates certain differences i n the t o t a l amount 3 The length of each performance does not seem to be re-lated to i t s quality, and was only included as a matter of interest. 43 of interaction between the four groups. Before any implications are drawn from a' comparison of these differences i t would be wise to examine them i n terms of the two formal sessions and events between the two formal sessions. Differences between the t o t a l amount of interaction for each of the groups during the f i r s t formal session are interesting, since they indicate the degree of intensity of interaction i n each group during their i n i t i a l contact on the working project. I t w i l l be noted that the two more intense interaction than the other two groups, the order being B2 > B]_> A > C. Furthermore, an interesting trend can be observed by com-paring, total group interaction for each group i n the two formal sessions: the subtotals i n Table IV. I t w i l l be noted that for both cliques the totals f a l l for the second session; while for the semi-structured and the unstructured groups they r i s e . At least three explanations are possible. The drop-off i n interaction for the two cliques and the ris e i n interaction for the semi-structured group lends support to the second hypothesis i n this study (and so also to Bos', Haythorn^s, and Potashin's findings) by suggesting that i f these two trends continued then group A would eventually overtake and surpass the two cliques. However other indications to be discussed tend to counteract this f i r s t explanation. Two other, interrelated explanations seem more relevant to the present data. The f i r s t of these two concerns particularly the r i s e i n interaction recorded for groups A and C. The said rise and the 44 difference i n degree of r i s e for the two groups suggests that as cohesive-ness of a group decreases, the length of time taken for the group to reach a "peak" of interaction (i.e., the%arming up" period) increases.'4' The "warming up" period for the cliques was negligible. The second explanation concerns particularly the observed f a l l i n interaction for the two cliques. While observing the inter-action sessions the experimenter noted a general "pattern of develop-ment", an orderly sequence which the interaction for each group f o l -lowed. First the groups "drafted out" a plot; during this period suggestions (recorded as "ideas" on the interaction recording sheet) are p r o l i f i c and welcomed. After the plot has been outlined the group begins to rehearse. At this point there i s a sharp drop i n the number of new ideas or suggestions which are offered. This drop continues, and as the rehearsals progress fewer and fewer ideas appear. The concern of the group i s now to "polish" their performance.-* Thus the severe drop i n recorded interaction for group B]_ i s the product both of the relatively large number of rehearsals they held between the two sessions and of an artifact of the recording procedure. Similarly group B2*s recorded interaction dropped between sessions. The drop here i s not so severe, since they did not rehearse as 4 Studies edited by Bradford and French (5) report that the productivity of discussion increases as members come to accept the group goal and their "group roles" become stronger than other competing roles. 5 General trends i n conference group procedure have been hypothesized by at least one experiment i n the literature—Bales and Strodbeck (2). 45 frequently as did group B]_. Group A met between the two formal sessions, but they did not reach the rehearsal "stage"; i n fact they had not reached i t by the end of the second formal session.^ Group C did not even reach the f i r s t "stage"; they were s t i l l trying—not very successfully—to "warm up". From this discussion of the postulated "stages" of group interaction arises the suspicion of an important and hitherto neglected variable—that of time. - The experimental results indicate that the time taken to pass through these "stages" i s inversely propor-tional to group cohesiveness, thus: cohesiveness C < A < B 2 < B^ time taken to reach rehearsal "stage" c ? A > % > TL A study of differences between the groups on the individual interaction c r i t e r i a yields less clear-cut results and trends, but i t draws attention to certain differences among the groups and i t also suggests certain refinements i n procedures used for assessing inter-action. Very few comments were made .in any group which could be placed under the heading of "Boosters"; the implications involved were omnipresent during the group sessions, but they were seldom overtly expressed. Therefore, this criterion did not differentiate very clearly between the groups. (However such a criterion might 6 Group A may well have handicapped themselves by choosing to write out their script as they thought i t up, rather than just talking about i t , as the other groups did. 46 prove more useful for groups meeting over a longer period of time than the groups i n this study.) The criterion of "Rejections" included comments made with varying degrees of apparently personal "feeling" involvement. I t therefore becomes very d i f f i c u l t to interpret the overall differences between the groups on this criterion. I t may be pointed out, however, that the group for which the highest number of "Rejections" was re-corded (group A) i s the same group i n which the greatest amount of f r i c t i o n between members was observed. A regret that extra-sessional rehearsals took place follows from the realization that an incomplete record of the number of "Ideas" for each group makes the relationship between t o t a l number of ideas and success of the skit impossible to assess. I f the number of "Irrelevancies" for each group i s regarded as an indication of the lack of concentration of that group upon the formal goal, then the semi-structured group (A) was not as concerned with achieving the goal as the other three groups. Cn the other hand, the fact that the unstructured group (C) made no irrelevant remarks suggests that some cohesiveness may be necessary i n a group before any social interaction, or "ordinary conversation" w i l l occur. While observing the interaction sessions the experimenter noted some further, qualitatively distinguishable differences between the groups.7 each of which includes implications for further research 7 The concept of syntality (syntality i s to a group as personality i s to an individual), which was introduced by Cattell (8), has received considerable attention i n the literature. Recently Haythom (16) has attempted to explore some of the relationships between the syntality of the group and the behavior of the individual group members. 47 and for refinement of present recording techniques. (l) The members of group A often interrupted one another and/or talked over each other. I f a contribution does not gain a complete hearing i t s f u l l import cannot reach each member of the group. At least some of i t s value (i.e., the possib i l i t y of i t s being used either i n part or as a whole by the group) i s lost to the group. How much group A's skit suffered as a result of thi s behaviour i s of course impossible to estimate on the basis of the present data. v (2) One 'of the members of group A tended to make angry remarks far more often than the other members of that or any other group. An individual with a "quick temper" can sidetrack or com-pletely halt a forward-going discussion. During this period several either potential or already expressed ideas may be lost (i.e., for-gotten) and the continuity of the discussion has been broken. (3) In general, the more "permissive" (i.e., relaxed and friendly) the group atmosphere the more suggestions w i l l be contributed. Fear of negative group reception of an idea probably prevents many suggestions from becoming overt, at least i n the age range studied here. Here i s a consideration which has been neglected so far.® I t is demonstrated i n a statement written by one of the subjects used i n this experiment when she was asked the reason for her choices for a future (hypothetical) working project: "...these kids are i n my cabin and I know them well, so this helps, because then you relax 8 Although friendship and "getting along well together" 48 and are not shy" (subject l ) . The study by Bos (4) , mentioned i n Chapter I I , was conducted before the concept of syntality was known as such. However, during this study Bos noticed certain qualitative differences i n the manner in which groups of two discussed a common task. Her categorization of these differences i s of particular interest here because i t copes adequately, although with a sl i g h t l y different emphasis, with the differences discussed i n sections (2) and (3) above. Bos discriminated between three major types of group inter-action (4, p« 394): (1) "Actively-collaberative form",-In this type there i s marked inter-member stimulation, and interaction i s carried on i n the group with v i t a l i t y and intensity. The behaviour of groups B^ and B 2 i s well described i n these terms. Particularly i n group B 2 the suggestions were rapid and provoked much comment and elaboration or counter-suggestions. (2) "Resistance-determined form",-Here the degree of positive mutual stimulation i s low and interaction results from pressure of resistive or negativistic contributions from a member of the group and of the necessity of achieving "results". Similarly the interaction observed i n group A of the present study f i t s this category. In this case individual 59 provided the was one of the most frequent reasons given for choice of future work-,ing mates (see Table XIV) i t i s interesting to note that only one of the four members of group G (the unstructured group) gave this as a reason. Yet this is the group which presumably suffered most from a lack of some such element i n the situation. 4? resistance which, for the majority of the time remained passive but which occasionally became active i n the form of such statements as "I think i t ' s cornyi , 9(see Tables V-B and VI). (3) "Collective Individual work",-Here iriter-member stimu-lation i s minimal and no effective collaboration i s achieved.• Group C's behaviour i s well described here. In this group there was no "pick-up" on the suggestions made; they (the suggestions) f e l l into silences which were amazingly painful and prolonged; while each member tried to develop a plot independently. It i s also interesting to note that Bos found the rate of achievement of these groups to be greatest for the f i r s t , or "actively-collaberative" form, next for the second, or "resistance-determined" form, and least for the third, or "collective individual" form. Present findings are i n agreement, i f the experimental groups are equated with Bos's categories i n the above manner. It should be recalled, however, that Bos also found that-groups i n which the members were interested i n each other were not as productive as groups i n which the members concentrated more upon the working project than upon each other (see Chapter I I ) . A consideration of these findings leads to the following suggestions. Although degree of s k i l l w i l l help or hinder achieve-ment,- i t remains dependent upon the type of interaction present. Thus i n the present study whatever s k i l l (imagination, etc.) was 9 Individual 59's index of motivation was equal to the indices of the other members of her group prior to the experiment. By the end of the project however i t had fallen off more than the other members' (see diagram i n Appendix I ) . 50 present i n groups and B 2 was relatively free to operate; while the s k i l l potential i n group C never had a chance to show i t s e l f . I t would seem doubtful (although we do not know as yet) that measures of cohesiveness could be used exclusively to predict what type of interaction w i l l occur i n a group. Certain personality t r a i t s of the members and the degree of motivation towards the group project are probably equally important. Much of which indicates that cohesiveness may predict the length of the period of "warming up" which w i l l be needed—and may or may not set a ceiling on how "warm" the atmosphere w i l l get—but i t may not predict the type of interaction which w i l l occur after the group has "warmed up"; which i n turn directly determines the productivity of that group. In the present study group A and particularly group C were handicapped both by the time factor and by the type of interaction involved. Discussion of Changes i n Motivation and Cohesiveness: a comparison of the "pre-" and "post-" questionnaires A comparison of these findings suggests some of the effects of working together and of success or failure on the working project upon attitudes toward the group and motivation toward a similar future project. These effects could be regarded as another kind of criterion of group productivity. Table XIII illustrates certain changes i n cohesiveness which occurred during the experiment. In a l l groups except the unstructured 51 one cohesiveness increased, and this increase parallels the success of the group. That i s , as degree of group success increases, adient relations within the group increase. 1^ Thus,', group B 2—the most successful group—had become a completely closed clique by the end of the experiment. These changes, as well as the changes i n motivation, have been illustrated diagramatically (see Appendix I ) . Tables XV and XVI i l l u s t r a t e certain changes i n motivation 11 towards the working project which took place during the experiment. Again, as success of the group increases, the motivation of the group rises. (The motivational indices for group B 2 could not ris e , since they were at a maximum prior to commencement of the experiment.) Therefore, results of the present study suggest a relation-ship between success of a group and i t s cohesiveness and motivation such that as success on an a c t i v i t y increases, cohesiveness (based on that activity) increases and motivation towards the a c t i v i t y rises. A qualification should be made concerning the above postulated relationships. Complete agreement between success of the groups as rated by the judges and as perceived by the group members themselves cannot be unhesitatingly assumed. The group members were 10 • I t seems unnecessary to point out the well known f a l -lacy of concluding that because two things correlate, there i s a causal relationship between them. Degree of enjoyment derived from working together i s probably an interrelated factor here also. 11 The relative motivation of the groups during their per-formances has not been included i n this section. The concern expres-sed by the judges over the d i f f i c u l t y of judging motivation " i n action" leaves the v a l i d i t y of their judgements open to serious question. 52 informed that group B2 had "won", according to the audience reaction ratings. The members of group C were aware that they were the only ones who had not been able to complete the working project. I t has been assumed that they would regard this as failure. The positions of groups A and B]_ were not announced. Although the difference i n audience reaction towards these two (B]_> A) seemed marked to the judges, the members of these two groups may have listened to the applause with somewhat biased ears. One might predict that the individual receiving the highest number of choices from within the group prior to the experiment would play the dominant r o l e 1 2 during group discussion. One might also predict that the individual who played the dominant role during the experimental task would receive the largest number of choices from the group on the post-experimental sociometric questionnaire. A comparison of the data from Tables VI and XIII does not support either of these predictions. Discussion of the Experimental Design The adequacy of the experimental design should be assessed before proceeding further, i n order to determine the degree of con-fidence with which the results can be generalized beyond the present experiment. As the study progressed some deficiencies i n i t s design 12 "Dominant" i s here defined as that member who contributes the greatest number of remarks during group discussion. Other defini-tions are of course possible. TABLE XVII A COMPARISON OF CHOICES RECEIVED AND AMOUNT SAID BY EACH INDIVIDUAL (TABLES VI AND XIII) Q Individual with Highest Individual with Highest Individual with Highest p Number of Choices Proportion of Remarks Number of Choices "~ ("pre-") ("post-") A 26 > 36 - 12 > 59 12 > 36 > 26 > 59 59 > 12 = 36 > 26 &L 14 = 13 = 39 > 11 13 > 39 > 14 > H 39 = 11 - 13 > 14 B 2 1 » 18 > 56 > 20 18 > 20 > 1 > 56 1 = 20 = 56 = 18 C — 46 > 29 > 57 > 34 54 became apparent. The following modifications would lend greater support to the findings. The lack of any control groups i s a serious omission. I f the two "pre-experimental11 questionnaires had been readministered to the camp at the conclusion of the study then: (1) I f the sociometric relationships within the camp had remained f a i r l y stable over the period of the experiment proper, then the changes which occurred within the experimental groups could be attributed with some confidence to factors within the experiment proper. In the present study this point was assumed without any such substantiating data. (A related point: ideally, a l l of the camp members should have been strangers at the beginning of camp in order to perfectly equate the length of acquaintanceship between the subjects) (2) The same criticism may be made with respect to the indices of motivation. I f at least two examples of each of the three types of experimental groups had been used i n the study (as did happen i n the case of the clique) then not only could the combined results be generalized with greater confidence but also an interesting study of individual differences within the same type of group struc-ture could have been made. There i s an implicit assumption i n this study and i n a good many other studies i n the f i e l d of Group Dynamics of a general consistency of behaviour for any sample drawn from the same population of group structure (see "Discussion of Interaction Assessment" i n this chapter). A number of working projects, a l l of which could remain i n 55 the same category of "leaderless group discussion", would have given the experimental results added weight. The f a l l i b i l i t y of the methodological techniques used has already been discussed (see Chapter I I ) . CHAPTER VI SUMMARY AND CONCLUSIONS The present study was designed to investigate the relative productiveness of groups of differing degrees of cohesiveness operat-ing i n an "exchange of ideas" situation. A review of the pertinent literature suggested two hypotheses: (1) That a group which i s either p a r t i a l l y or highly structured w i l l be more productive i n carrying out a working project than a group which i s relatively unstructured. (2) That a "semi-structured" group w i l l be more productive i n carrying out a working project than a highly structured group. Working in a camp setting, groups were chosen on a socio-metric basis to represent the various degrees of cohesiveness set forth i n the hypotheses. These groups were equated as far as possible for age and motivation. Each of the experimental groups was allowed two periods of fifteen minutes each i n which to prepare a ski t for presentation before the camp. Observations of group interaction during these sessions was made, and any changes i n motivation and cohesiveness of the group members was assessed after their performances. Three independent judgements of the performances yielded the 57 following ratings: n First - the highly structured group which was superior to the other groups i n both age and motivation. Second - the highly structured group which was inferior to the other groups i n both age and motivation. Third - the semi-structured group. The unstructured group could not complete the working project and failed to give a performance. On the basis of these findings the f i r s t hypothesis must be accepted and the second rejected. Within the limits of the design of the present study certain conclusions can be made. (1) There i s a relationship between the cohesiveness and productivity of a group such that as group cohesiveness increases i t s productivity also increases. Investigation of group interaction, however, suggests that productivity i s also dependent on other variables i n the situation (time, type of interaction) and these variables i n turn may or may not directly depend on cohesiveness. (2) Group productivity i s related to the type of interaction in the group. Although the exact nature of this relationship has not been f u l l y explored i t would appear.that the more intense and more collaborative the interaction, the greater the productivity of the group. (3) Group interaction i s related to time and cohesiveness such that the more cohesive the group the less time i s needed to reach a high intensity of interaction. 58 (4) Degree of group cohesiveness changes i n relation to degree of group success so that as success of the group increases ' i t s cohesiveness rises. (5) Similarly the motivation of the group members towards a common task rises as success of the group on this task, increases. CHAPTER VII IMPLICATIONS FOR FURTHER RESEARCH In a f i e l d of research which i s as unexplored and which contains as many undiscovered variables as this one the necessity of studying the.groups " i n action" i s paramount. The name of the whole field—"Group Dynamics"—indicates what should be the primary area of investigation. To ignore process i s to impose a gratifying but a r t i f i c i a l and unrealistic neatness upon the particular study and i t s findings.^ A second major implication for further research which arises out of this and of a l l other related studies l i e s i n the need to continue to develop and to refine methodological techniques to f a c i l i t a t e the accurate study of a l l aspects of group behaviour. As , was mentioned earlier, when a concept or a quality i s assessed or measured, i t loses i t s original identity and becomes a product of whatever technique of measurement i s used. Therefore to the extent that that technique i s inappropriate or unrefined, the concept or quality measured becomes biased and distorted. Again, the present study i s centered around a working project which depended largely upon group interaction of a verbal, sedentary nature. How appropriate i t s hypotheses and i t s results 1 I t i s hoped that the U. of Mich. Conf. Res. Proj. (18, 30) w i l l not hold to i t s avowed purpose of "evaluating results rather than describing process" (see Chapter I I , p. 6). 60 are to groups engaged i n a different type of project (for example, the physically active, non-verbal type used by Goodacre (13)) remains debatable. Until this question i s answered, i t would seem wisest for studies i n this area to keep their projects a l l of the same type, i n order to avoid possible distortion of their results (which may have occurred i n French's study (12), for example). The more, specific implications for further research arising out of the present study are numerous. Some of the more important ones, however, arise out of the conclusions stated i n the last chapter. Each of these conclusions i s i t s e l f i n need of further investigation before i t can safely be generalized beyond the limits of the present study. Suggested hypotheses for future research might include the following: (1) That type of group interaction i s directly dependent upon group cohesiveness. ( 2) That the more cohesive the group the less time i t takes to pass through observed interaction "stages". (3) That as success of the group rises i t s cohesiveness increases. (4) That as success of the group rises the members' moti-vation increases. REFERENCES Bales, R.F., & Gerbrands, H. The "Interaction Recorder". Hum. Relat., 1947-1948, I, 456-463-Bales, R.F., & Strodtbeck, F.L. Phases i n group problem-solving. J. abnorm- soc. Psychol.. 1951, 46, 485-495< Berkowitz, L. Sharing leadership i n small, decision-making groups. J. abnorm. soc. Psychol., 1953, 48, 231-238. Bos, M.C. Experimental study of productive collaboration. Acta psychol., 1937, 3, 315-426-Bradford, J.P., and French, J.R.P. (ed.)• The dynamics of the discussion group. J. soc Issues, 1948, IV, No. 2, 75 pp-Bronfenbrenner, U. A constant frame of reference for socio-metric research. Sociometry Monogr., 1944, No. 6. Carter, L.F., Haythorn, W., Meirowitz, B., & Lanzetta, J. Note on a new technique of interaction recording. J. abnorm. soc. Psychol.. 1951, 46, 258-260. Cattell, R.B., & Lauren, G.W. The dimensions of syntality i n small groups. J. soc. Psychol., 1948, 28, 57-78-Coch, L-, and French, J.R.P. Overcoming resistance to change. Hum. Relat., 1948, I, 512-532-Cohen, J. Some working hypotheses and provisional conclusions i n the study of committees and conferences. Occup. Psychol.,  Lond.. 1952, XXVT, 70-77-62 Cronbach, L.J. Essentials of psychological testing. N.Y.: Harper, 1949-French, R.L. Sociometric measures i n relation to individual adjustment and group performance among naval recruits (abstract). Amer. Psychologist, 1949, 4, 262. Goodacre, D.M. The use of a sociometric test as a predictors of combat unit effectiveness. Sociometry, 1951, XIV, 148-152. Guetzkow, H. Unitizing and categorizing problems i n coding qualitive data. J. c l i n . Psychol., 1950, VI, 47-58. Hare, A.P. Small group discussions with participatory and supervisory leadership. J. abnorm. soc. Psychol., 1953, 48, 273-275-Haythorn, W. The influence of individual members on the characteristics'of small groups. J. abnorm. soc. Psychol. 1953, 48, 276-284-Joubert, G.J. Indiwiduele en Kollektiewe Prestasie, 'n bi.jdrae tot die experimentele groepspsigologie. Swets en Zeitlinger, Amsterdam, 1932. * Katz, D. Social psychology and group processes. Ann. Rev. Psychol., 1951, 2, 137-172-Lewin, K. Frontiers i n group dynamics I and I I . Hum. Relat., 1947-1948, I, 5-41, 143-153-Lewin, K., & Lippitt, R. An experimental approach to the study of autocracy and democracy: a preliminary note. Sociometry. 1938, I, 292-300. Lewin, K., Lippitt, R., & White,'R.K. Patterns, of aggressive behaviour i n experimentally created "social climates". J. soc. Psychol., 1939, 10, 271-299-63 22. Idndquist, E.F. A f i r s t course i n stat i s t i c s (rev. ed.). N.Y.: Houghton M i f f l i n , 1942. 23- McCurdy, H.E., & Lambert, W.E. The efficiency of small human groups i n the solution of problems requiring genuine co-operation. J. Personality. 1952, XX, 478-494. 24. Northway, M. A primer of sociometry. Toronto: U. of T. Press, 1952. 25- Potashin, R. A sociometric study of children's friendships. Sociometry Monogr.. 1947, No. 11, 31-53-26. Preston, M.G., & Heintz, R.K. Effects of participatory versus supervisory leadership on group judgement. J. abnorm. soc. Psychol.. 1949, 44, 345-355-27- Rosenthal, E., & Cofer, C.N. The effect on group performance of an indifferent and neglectful attitude shown by one group member. J. exper. Psychol.. 1948, 38, 568-5.77-28. Schachter, S., Ellertson, N., McBride, D., & Gregory, D. An experimental study of cohesiveness and productivity. Hum. Relat.. 1951, IV, 229-238. 29. Schultz, W.C. Re l i a b i l i t y , ambiguity, and content analysis. Psychol. Rev., 1952, 59, 119-129. 30. Smith, M.B. Social psychology and group processes. Ann. Rev.  Psychol., 1952, 3, 175-204-31. Thibaut, J. An experimental study of the cohesiveness of under-privileged groups. Hum. Relat.. 1950, I I I , 251-278. 32. Weschler, I.R., Kahane, M. and Tannenbaum, R. Job satisfaction, productivity and morale: a case study. Occup. Psychol..  Lond., 1952, XXVI, 1-14-APPENDIX A THE PRE-EXPERIMENTAL SOCIOMETRIC QUESTIONNAIRE NAME AGE DATS OF BIRTH Suppose the cabin groups were going to be rearranged, who of a l l the. g i r l s would you most l i k e to have i n your.cabin? Mao would be your second choice? lour t h i r d choice? < TO BE IN SAME CABIN WITH YOU: CHOICE NAME F i r s t Choice Second Choice Third Choice Suppose a l l the g i r l s i n the camp were going to be divided into small groups so as to t a l k about certain topics, who would you most l i k " to have i n your group? Who would be your second choice? Your t h i r d choice? TO BS IN SAME DISCUSSION GROUP WITH YOU . CHOICE NAME ; F i r s t Choice Second Choice • Third Choice . • I f you were going on an all-day hike, who of a l l the g i r l s would you most l i k e to have go with you? Who would be your second choice? Your t h i r d choice? TO CrO ON A HIKE "WITH YOU CHOICE NAME i F i r s t Choice j Second Choice I Third Choice i I f you had been asked to help get some entertainment ready for the camp, who would you most l i k e to have help you? Who would be your second choice? Your t h i r d choice? TO HELP YOU VJITH ENTERTAINMENT CHOICE NAME ' | F i r s t Choice Second Choice Third Choice APPENDIX B PRINCIPLES FOR CHOOSING SOCIOMETRIC CRITERIA 67 THE APPLICATION IN THE PRESENT STUDY OF PRINCIPLES SET FORTH BY BRONFENBRENNER AND NORTHWAY RE CHOOSING SOCIOMETRIC CRITERIA Both Bronfenbrenner (6) and Northway (24) stress the impor-tance of using criteria which are relevant for the particular group which i s to be studied* In the present study, a l l four criteria are based on activities which are familiar ones in a camp setting. Northway suggests that sociometric criteria should embrace varying aspects of the activities relevant to the group. Accordingly, criteria were chosen which emphasize different features of camp l i f e . Bronfenbrenner suggests that the criteria should provide equal potentiality of participation for a l l group members. With respect to this condition, the third criterion used (see Appendix A) would have been seriously invalidated i f any of the camp ers was physically handicapped. Fortunately none of them was. Finally, Bronfenbrenner lists clarity of meaning as an important factor. For this reason the wording of the criteria and of the instructions was made as simple as possible. In addition, time and opportunity were given to ensure that the subjects understood a l l aspects of the questionnaire. The paucity of questions asked by the subjects indicated that this condition had been satisfied. APPENDIX C THE PRE-EXPERIMENTAL MOTIVATION SCALE NAME Directions: In each of the following, c i r c l e whichever answer states best how you f e e l about the a c t i v i t y : I would l i k e to help think of ideas for a s k i t -VERY QUITE INDIFFERENT NOT VERY NOT AT MUCH WELL MUCH ALL I would l i k e to act i n a s k i t -VERY MUCH QUITE WELL INDIFFERENT NOT VERY MUCH NOT AT ALL I would l i k e to help put on a s k i t -INDIFFERENT VERY MUCH QUITE WELL NOT VERY MUCH NOT AT ALL APPENDIX D THE INTERACTION SCALE J N T f t e o u c e s i w s i o e loe-f) • i. r inet 7 <! ? ' > Tine < o / 5" secoNO sersstoN PitlST session 1ST strc»Nt> sersstoN Tine o s~ /o <£V»OeNC£ Opt COMPETITION FI*ST se-sstoN IS" 21 z: \*IAHTS To OO ft Co o 0 3o & P i f i s T XIMC o session j-ff-CoMO sess to M 0 /J N R ME . \ -t GRou P ( ) s e c * HO S KS £ to N 10 /£-FI AST Se-ssicN S1 IS" GROUP ( ) SCCffiQ S*S%ftf ^?£ 2 1 -• -APPENDIX E THE POST-EXPERIMENTAL MOTIVATION SCALE I would l i k e to help plan and present another s k i t : VERY QUITE INDIFFERENT NOT VERY NOT AT MUCH WELL WELL ALL Why did you c i r c l e the one you did? COMMENTS: APPENDIX F THE POST-EXPERIMENTAL SOCIOMETRIC QUESTIONNAIRE NAME I f you were asked to help plan and present another skit, which g i r l from the camp would you most l i k e to have on your team? Who would be your second choice? Your third choice? CHOICE NAME First Choice  Second Choice Third Choice  Why did you choose these three girls? APPENDIX G SOCTOGRAMS OF THE CAMP AND OF THE EXPERIMENTAL GROUPS G R O ~ — - •> Afcariy /?a.»>«wiTj«7i| A V«?*y W E I L & Quire VJcu. Nor Ve«y We-tL O R O U P C — — - fl»ITTty RtXlHNf'Ttfl • |> UMeciP«o«rt7*t> & Qu 'Tf W r u HOT Uect APPENDIX H JUDGES' COMMENTS ON THE PERFORMANCES 85 COMMENTS BY JUDGES S, T, AND U ON THE EXPERIMENTAL GROUPS' PERFORMANCES GROUP A Judge S - not too much appeal. - depended too much on narrator rather than individual acting. - not a punchy enough ending: lacked originality. - sincere. Judge T - narrator a good reader. - everything went fair l y smoothly. - weak ending. Judge U - I don't think the f i r s t group put as much as they could have into i t . They didn't seem to get into the play as well as the other groups. They were too conscious of the audience. GROUP h Judge S - very definite evidence of careful preparation. - a dominant character. - ending not original. - very earnest attempt. For some unknown reason I felt rather partial to this group. Judge T - 13 very good. Carried most of skit. - very good. - weaker beginning but stronger than group A. Judge U - I think the one g i r l had too much of a part. If the others could have done more I think i t would have improved i t a l i t t l e - or maybe they were supposed to and didn't. GROUP B2 Judge S - good beginning, more of an original ending. - had more confidence. Judge T - not loud enough. - very good start. - good ending. - carried by 18. (I couldn't see most of this performance.) Judge U - These girls seemed well prepared and confident. The one g i r l (l) did slip out of character to answer wSquamishn. APPENDIX I DIAGRAMMATIC REPRESENTATION OF THE CHANGES WHICH OCCURRED IN COHESIVENESS AND MOVITATION OF THE GROUPS DURING THE EXPERIMENT C H A N G E S CO H ES } VEN BSS AND MOTlVATio^ OF FOUR B M C N T f l L G R O U P S G R O U P R B a. C P R E - ffX P E « I Mff N T f l L choice's Bfiseo O N T H R f e £ E N C R « L C R I T E R cHoices &Aseo ON ONE (FOURTH) CRlTfi-flJON P O S T - e * PERI WENT ft C H O I C E S ft^seo O N O N B ( F O U R T H ) C R I T F « | 0 M sby£ \ $ R E L A T I O N S H I P S M O T I V A T I O N < 1 > R r c t r n o c R T e o ^ PflRTty ffecirKocmva - UNKNOWN Q Q u i TIT U/trci_ Q IN D\FFe*e*rr O N o r VERY Wext, 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items