2.0. 40:20 ._ . _ V 7 L A__ ,__ ,7, _ W fl-fi _ —- AM ' :31 J" l“MICHIGAN STATE UNIVER I Y Immullw ll"'lllllllllfilflll mm 3 12553 00971 3254 LIQRARY Michigan State University This is to certify that the thesis entitled SMALL GROUP INTERACTION: SOME NOTES ON THE EMERGENCE OF PROCEDURAL RULES AND THEIR RELATIONSHIP TO THE DEVELOPMENT OF A STATUS-PRESTIGE ORDER IN CERTAIN SMALL GROUPS presented by Noel Dean Young has been accepted towards fulfillment of the requirements for Masters degree in Sociol 091 QUE/tum 7L COMM/Ck Major professor 'Ihanas Conner 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES “ RETURNING MATERIAL§; Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. SMALL GROUP INTERACTION: SOME NOTES ON THE EMERGENCE OF PROCEDURAL RULES AND THEIR RELATIONSHIP TO THE DEVELOPMENT OF A STATUS-PRESTIGE ORDER IN CERTAIN SMALL GROUPS BY NOEL DEAN YOUNG A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology 1985 Copyright by NOEL DEAN YOUNG 1985 ABSTRACT SMALL GROUP INTERACTION: SOME NOTES ON THE EMERGENCE OF PROCEDURAL RULES AND THEIR RELATIONSHIP TO THE DEVELOPMENT OF A STATUS-PRESTIGE ORDER IN CERTAIN SMALL GROUPS BY Noel Dean Young This study establishes the empirical existence of certain categories of speech acts and the development of a hierarchical structure of group interaction in three- person, task-focused, collectively oriented groups. A secondary analysis of the transcripts from four experi- mental situations in which participants "talked about” while ”doing" distributive justice is the basis for the evaluation of this problem. The data suggest that such a relationship between the development of a status-prestige order and the emergence of procedural rules does exist. Several additional interesting features are also implied. The exact nature of these associations cannot be established empirically with these data. ACKNOWLEDGEMENTS I would like to thank my advisor, Tom Conner, and the other members of my committee, Hans Lee and S.F. Camilleri for their support in this project. In addi- tion, Bo Anderson provided insight into the original experimental design. Lastly, Rick Hurst has provided encouragement throughout. iii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O O O 0 LIST OF FIGURES . . . . . . . . . . . . . CHAPTER I. INTRODUCTION . . . . . . . . . . . The Phenomenon . . . . . . . . Investigating Task Procedures an Hierarchies . . . . . . . . . The Problem . . . . . . . . . . Scope and Initial Conditions . Kinds of Behavior . . . . . . . The Analysis . . . . . . . . . II. METHOD AND DESIGN . . . . . . . . . Recruitment . . . . . . . . . . . Characteristcs of the Participants Setting . . . . . . . . . . The Task . . . . . . . . . Experimental Procedures . . Debriefing the Participants III 0 RESULTS 0 O C O O O O O O I O O O 0 Operational Referents of th Interaction Categories . . . . . Differentiation . . . The Data . . . . . . Results . . . . . . . IV. DISCUSSION AND CONCLUSIONS . . . . . Ranking . . . . . . . . . . . . . Phases . . . . . . . . . . . . . Differential Distribution of the Types of Acts . . . . . . . . . Structural Change . . . . . . . . Issues . . . . . . . . . . . . . V. SUMMARY 0 O O O O O O O O O O O O 0 NOTES 0 O O O O O O O O O O O O O O O O 0 LIST OF REFERENCES . . . . . . . . . . . . iv HH \IO‘U'lubw 10 11 12 12 15 17 17 18 19 21 46 46 47 51 54 55 58 61 63 TABLE 1. LIST OF TABLES Final proportions of acts initiated by category . . . . . . . . . . . 21 thh‘ UJNFH NNN NNN HHH UJNH O‘U'Iih O‘U‘lnb www WNH LIST OF FIGURES FIGURE Entire Sample: Proportion of participation cumulated over time each basic initiation rank. . . . . . . . . . Performance outputs . . . . . . . . . . . . . PrOblem-SOIVIDQ attempts o o o o o o o o o o o Task-organizing procedures .'. . . . . . . . . Entire Sample: by 23 23 23 24 Proportion of participation by time period by each basic initiation rank. . . . . . . . . . . . Performance outputs . . . . . . . . . . . . . . Problem-solving attempts . . . . . . . . . . . Task-organizing procedures . . Differentiated Subset: Proportion of participation cumulated over time each basic initiation rank. . . . . . . . . . Performance outputs . . . . . . . . . . . . . . Problem-solving attempts . . . . . . . . . . . Task-organizing procedures . . . . . . . . . . Differentiated Subset: 26 26 26 27 by 29 29 29 30 Proportion of participation by time period by each basic initiation rank. . . . . . . . . . . . . Performance outputs . . . . . . . . . . . . . . PrOblem-SOIVIHQ attempts o o o o o o o o o o o Task-organizing procedures . . . . . . . . . . Equal Status Characteristic Subset: Proportion of participation cumulated over time each basic initiation rank. . . . . . . . . . Performance outputs . . . . . . . . . . . . . . Problem-solving attempts . . . . . . . . . . . Task-organizing procedures . . . . . . . . . . vi 32 32 32 33 by 35 WWW GUI-b bub-b mm» Equal Status Characteristic Subset: Proportion of participation by time period by basic initiation rank. . . . . . . . Performance outputs . . . . . . . . . Problem-solving attempts . . . . . . Task-organizing procedures . . . . . Unequal Status Characteristic Subset: each . 38 Proportion of participation cumulated over time by each basic initiation rank. . . . . . Performance outputs . . . . . . . . . Problem-solving attempts . . . . . . Task-organizing procedures . . . . . Unequal Status Characteristic Subset: Proportion of participation by time period by basic initiation rank. . . . . . . . Performance outputs . . . . . . . . . Problem-solving attempts . . . . . . Task-organizing procedures . . . . . vii each CHAPTER I INTRODUCTION The Phenomenon Bales (1950) noted that small group members, to solve a collective task, do not interact randomly. To further investigate this phenomenon, he developed a general cate- gory scheme for coding behavior in the small group set- ting to serve both as a basis for an accumulation of research and to provide a method of observation and analysis for a wide range of problems. Among his find- ings was that the rates of various categories of partici- pation are distributed unequally throughout the course of interaction and among persons (1950:126). Some members will be more active than others in presenting possible solutions, soliciting opinions from others, or facili- tating the discussion. These differences in rates of participation indicate a status-prestige order among members as Berger, Cohen, and Zelditch (1966) have pointed out. The emergence of such a status-prestige order in informal task-focused groups has been attempted to be explained through theoretical and empirical inves- tigations stemming from this original research (Berger, 1974z5). In particular, Berger et a1. (1966:29) presented a theory explaining how “prior status factors determine the emergent power-prestige order in a group.” The early work on this original set of problems has developed into the formulation of Expectation States Theory (Berger et al., 1974). Berger and Conner (1969:1974), in relation to Expectation States Theory, developed a typology of kinds of behavior and explained the emergence of role and status systems as consequences of the effect of estab- lished status systems upon interaction. Their main pur- pose was to show that ”patterns of behavioral inequali- ties in task-focused small groups can be explained by positing an underlying structure of ranked performance expectation states” (1974:106). Since the interest in these studies is directed at task-focused groups and not at groups for social and emotional support, there is expected to be some emphasis on efficient completion of the task (Berger and Conner, 1974:101). Accordingly, McMahon et a1. (1979:5) propose that the members of any task group must solve two classes of problems. ”They must, of course, solve the problems defined by the explicit task assigned to the group. To do that, they must solve a second set of problems con- cerning the procedures the group will follow in solving the tasks.” McMahon et a1. call this second set of prob- lems Task Procedure and distinguish them from Task Con- tent problems. Although Expectation States Theory explains ”the effects of status characteristics in organizing social interaction“ (Berger, 1974:4), it has not specifically explored the relationship between the development of any procedural rules and the emergence of a status-prestige order. Investigating Task Procedures and Status Hierarchies A study designed to examine the relationship between task-procedural acts and emerging status hierarchies would first involve determining the status order and then distinguishing between acts related to Task Content problems and those related to Task Procedure problems. In establishing the status order for the group, two questions need to be examined. First, what is the order in which the status positions are established? That is, perhaps the number one position is differentiated first and the other positions are defined later. Second, how much stratification is established, and when is it established? Next, questions relating to task—organizing pro- cedures need to be examined. First, does the group develop some organizational principles or procedural rules? Second, if the group does develop some rules, when in clock-time do they become apparent or manifest? Third, where is this point in experimental time? These questions then lead us into another set of questions regarding the relationship between the establishment of the status order and the emergence of procedural rules. That is, are the procedural rules established first or is the status order decided first, or is one inextricably bound up in the other? Stated differently, how much of the stratification has to be established for the emer- gence of procedural rules? Examination of the interaction process itself will help to shed light on these questions and issues. While the interaction process is made up of several components, both verbal and non-verbal, the study here is concerned with rates of talking. Focusing attention on verbal statements of the subjects assumes that the speech is a presentation of reasoning relevant in accomplishing a group task (Conner, 1975:7), and is a creative process as well as a reflection of thought. The Problem The objective of this project is to describe, by establishing the frequency of certain speech acts for each actor, the development of a status-prestige order and its temporal organizational relationship to the emer- gence of procedural rules in three-person, task-focused, collectively oriented groups. The study attempts to establish the empirical existence of these categories of speech acts and the development of a hierarchical struc- ture of group interaction. The working hypothesis is that if certain speech acts (which can be thought of as expressing norms and standards of judgment for task con- tent problems) are put forth by people with high status, then certain other speech acts (which also can be thought of as expressing norms and standards of judgment but in relation to task procedure problems) are put forth by people with high status. Scope and Initial Conditions This research focuses on the initiation of certain verbal acts. The condition of collective orientation implies that the group must be working together on a common and valued task. If the group is collectively oriented, then it is both necessary and legitimate for the members to take into account the behavior of others in solving the task (Fisek, 1974:56; Berger and Conner, 1974:86). Further, following Fisek (1974) and McMahon et a1. (1979), the group must develop its own methods and procedures for regulating behavior and structuring the interaction. The imposition from an external source of such regulatory methods and procedures could structure the interaction in such a manner that would unnecessarily hide and obscure an important process of the interaction and a concern of this study. Finally, the theoretical perspective of this study calls for the absence of differentiated status characteristics at the outset of the discussion period. Differentiated status charac- teristics can serve as a basis for developing task- performance expectations (Fisek, 1974:56), thereby affecting the emergence of a status-prestige order and confounding the results. Kinds of Behavior Adopting the strategy of Berger and Conner (1969:1883 1974:88), the analysis of the interaction will be limited to certain kinds of behavior which can be classified as task-related. Thus, behavior which is purely ”social- emotional“ for Bales (1950:8-9) will be ignored since it falls outside the focus of this research. An act is the smallest unit of social behavior that can be classified in this system. An act, for present purposes, is a mini- mal utterance capable of being evaluated as a unit which is task-relevant. It is a unit of speech equivalent to Scheflen's point (1974:22-23; see also Conner, 1975:1). Berger and Conner (1969; 1974) identify four kinds of acts; however only the type they call a performance 1 output is of interest here. "A performance output is an act that is an attempt to solve or partially solve a sub- problem” (1969:188). It is a task-relevant contribution to the discussion by one of the actors. Since we have identified two types of task-related problems, it is useful to distinguish two corresponding types of performance outputs. Certain performance out- puts can be understood as problem-solving attempts (Conner, 1975:?) which are oriented to task—content prob- lems. A problem-solving attempt, as a speech act, is a minimal utterance capable of being evaluated as a unit and relevant to the task-content problem. Certain other performance outputs can be understood as task—organizing procedures and are concerned with task-procedure prob- lems. A task-organizing procedure is a speech act expressing a complete thought, the content of which attempts to orient, guide, or give direction to the co- ordination of the group's time, behavior, and energy towards solving the primary task given the group. The Analysis A secondary analysis of the transcripts from four small group interaction experiments designed to study a related problem from an ethnomethodological perspective is the basis for the evaluation of this problem. As with McMahon et a1. (1979:5), a structural inter- pretation of the interaction process itself will examine the changing emphasis of certain speech acts over time and will avoid reducing the analysis to psychological traits and attributes of individuals. CHAPTER II METHOD AND DESIGN This investigation called for the examination of an emerging process in the development of status hierar— chies. The principal operational concern of the study was to have a situation where there would be a good deal of interaction between individuals. This would enable the researcher to observe and document the process by which individuals become organized and go about solving their task. This section will relate the method and design of the experiments from which the data were collected. Recruitment Students enrolled in introductory French and socio- logy classes at Michigan State University were asked by the researcher during classes if they would like to par- ticipate in a study concerning allocation of financial awards. A brief recruitment speech was given, stating that the researcher was employed by the Michigan Educa- tional Granting Association. The students were told that this organization was revising its guidelines in relation to how financial awards ought to be allocated, and that the purpose of the study was to obtain student input 9 10 regarding the subject. They were told that if they decided to volunteer for the study, it would involve two separate sessions, each lasting approximately one to two hours, and that they would be paid $10.00 for their par- ticipation. The students were also told that they would be videotaped and that the video tapes would be viewed only by the researcher, the researcher's advisor, and the committee. The researcher strongly emphasized that the participants would not be asked to do a nonsensical task or be deceived in any way. Next, the researcher arbitrarily selected groups of three from the list of individuals who had volunteered to participate in the study. A group was then instructed to meet in the Sociology Laboratory at a specified time, at which point they were completely briefed with regard to the intention of the study. These procedures were repeated until all groups of three had been used in the study. Characteristics of the Participants Participants ranged from nineteen to twenty-two years of age. They were from a variety of majors (sociology, secondary education, advertising, philosophy, etc.). The experimental group consisted of twelve participants, five 11 females and seven males. There was no particular control over gender and race. Setting The Sociology Laboratory consists of two rooms, an observation room and a group-activities room, which are separated by a wall-length, one-way mirror. Two rectangular tables, four chairs, a portable chalkboard, video camera, monitor, special effects generator, video cassette recorder, and microphone were placed in the group-activities room. The recording equipment, except for the camera and microphone, was shielded from direct observation by the participants with a divider. A pad of paper and some pencils were placed on top of the tables. The chalkboard had the words 'Michigan Educational Granting Association' written on it. The microphone was placed on a stand directly in front of the tables. Cur- tains were drawn across the one-way mirror since the observation room was not utilized during the experiments. The two rectangular tables, placed together, were large enough to allow each person to view the other individuals simultaneously. The participants sat around three sides of the table arrangement in one corner of the room. This arrangement had the middle participant directly facing the camera. The other two participants 12 were directly facing each other. The recording equipment was located in the adjacent corner. The researcher sat on the fourth chair in front of the divider. The Task The experimental design, as presented here, was adapted from Anderson and Hurst (1977). The task for the participants in this study, as with Anderson and Hurst's design, was to make a group decision of how financial awards ought to be allocated. The primary difference between these two designs was that this design was not as formally and highly structured as that of Anderson and Hurst. This was accomplished by having two phases, each taking place on separate days. The first phase consisted of an open-ended interaction among three individuals who were assigned the task of developing a list of Ell relevant criteria to be used in evaluating applicants and awarding educational grants.2 Experimental Procedures Once the group had assembled in the laboratory, they were given a tour of the facilities, including the obser- vation room, and reminded that the video equipment would be used, but not the observation room. Next, the 13 researcher explained the operation of the video equip- ment. The group was told that it was necessary to videotape the interaction for this study because it would record precisely, and without bias, the group proceed- ings. The researcher further explained that this was important in the future when reviewing the group's deci— sions and their reasoning behind those decisions. The participants were then told that if they still had any objections to being videotaped that they did not have to participate and were free to leave; no person chose to leave. The researcher then asked the participants to take a seat at the table, and gave the following intro- duction and instructions: Hello, I'm glad to see that you could all make it here today. You're probably wondering what this study is about. Well, as you know, various types of financial awards are given to students each year at Michigan State University and other colleges and uni- versities in Michigan. Probably yourself, a friend, or someone you know of has applied for some type of financial award at one time or another. Yet, how the awarding committee's decisions are made as to who should receive funds is unclear to most applicants, and often even the persons who sit on committees that give out financial awards are unclear about what cri- teria ought to be used. As you might well be aware, recently there has been much debate concerning this issue of what criteria should be used in awarding funds for education. To help resolve this problem, the Michigan Education- al Granting Association has decided to ask groups of students which criteria should be used and why these criteria are important when evaluating an indivi- dual's application. It is hoped that the committee, with the aid of student input, can construct better guidelines with respect to how decisions should be 14 made when evaluating applications. This means, also, that persons who do not receive an award can be given specific reasons why they did not receive a financial award. The task we have for you today consists of two parts. First, you are to develop a list of all relevant criteria to be used in evaluating applicants for a financial award. This list can be as long or as short as you want; just list all the things to be considered. Secondly, you are to rank this criteria in order of importance. For example, you may decide that 'x' criterium is to be considered ahead of and with more emphasis than 'Y' criterium. Or, you may choose to weigh all the elements equally or make groupings of elements. The main thing is that you are free to develop the list in whatever manner you choose, containing information that you wish to con- sider in evaluating applicants for financial awards. In developing your list, and later in ranking the elements on it, you should be sure to state your reasons as clearly as possible as to why 'X' is important, and why it is more important than 'Y'. The committee will be especially interested in your reasoning. Also, you must work as a group and come to a consensus decision. If, for some reason, you find that you cannot reach a consensus, then you must mark exactly what the grounds are for your disagree- ment. Next week, or whenever you decide you can get together again as a group, you will have to evaluate and rank, according to the criteria you decide upon today, some randomly chosen applicants from the organization's inactive files. Today, however, your only task is to develop a list of criteria to be used in evaluating applicants for financial awards, and then ranking that list in order of importance. Next the researcher asked the participants if they had any questions concerning the study or task. After all questions asked were answered, the researcher instructed the group to begin the task. The researcher sat in the remaining chair, in front of the divider but 15 detached from the group. Throughout the group's discus- sion of the task, the researcher refrained from initiat- ing any suggestions or conclusions. If he was asked a question, the researcher first replied by asking why that would make a difference? After the group had sufficient- ly stated their reasons, the researcher then made a decision. The few questions asked were generally of the nature of whether the group was "on the right track?" to which the reply was always “yes.” The role of the researcher at this phase was to act as a detached observer and to make sure there was no pre—determined resolution as to what the criteria for financial awards ought to be. It should also be men- tioned that the above statement was not read word-for- word to each group, nor was it memorized and repeated verbatim; rather its content was memorized and it was presented in a manner like everyday language usage. It was hoped that this would help make the formal and sterile environment of the laboratory somewhat more relaxed. Debriefingfithe Participants The debriefing session was conducted with the whole group. Questions concerning what they had done and how they had gone about doing it were asked first, such as: 16 whether they found the task easy or hard, why they found it so, and if they were satisfied with their final deci- sion. The group was also questioned as to whether or not the camera and/or the presence of the researcher was in any way intimidating. The group was asked finally if they had any suggestions about how the design of the study could be improved and if they had any additional thoughts or feelings about the study. Generally, par- ticipants were extremely cooperative in answering these post-experimental interview questions. At this point, the researcher informed the group that the applicants were fictitious and that there was no such organization as the Michigan Educational Granting Asso- ciation. The group was reassured that although the researcher had been less than open with them concerning all of the details of the study, that the task itself was meaningful: the study was concerned with how groups of individuals, or committees, went about deciding how financial awards ought to be allocated and that the pro- cedures used in this study were, in fact, how some com- mittees went about accomplishing this task. After responding to any questions or complaints that the par- ticipants had, the researcher paid the individuals and thanked them for their participation in this study. CHAPTER III RESULTS The experiments, which were videotaped, resulted in a good deal of group discussion. A transcript of the videotape of each group's discussion was coded by the researcher for initiation of Performance Outputs. The content of each separate Performance Output was then categorized into either a Problem-Solving Attempt or a Task-Organizing Procedure. Targets of these acts were not recorded. Acceptance or rejection of these Perfor- mance Outputs was also ignored. This material formed the data base for the results of the investigation which are presented in this section. Operational Referents of the Interaction Categories As defined earlier, a problem-solving attempt as a speech act is a minimal utterance capable of being evalu- ated as a unit and relevant to the task content problem. Some sub-categories of task-content relevancy are taken from Bales (1950). These include: giving information, clarifiying, explaining, analyzing, giving confirmation, giving opinions, giving suggestions, and repeating. This 17 18 is not an exhaustive list of the categories of problem- solving attempts; rather, it is an attempt at definition by examples. Further, recall that a task-organizing procedure is a speech act expressing a complete thought, the content of which attempts to orient, guide, or give direction to the coordination of the group's time, behavior, and energy towards solving the primary task given the group. Some sub-categories for the relevancy to task-procedural acts are taken from McMahon et a1. (1979). These include: ordering of the task sub-units; establishing methods of allocating time and energy to these sub-units; moving from one sub-unit to the next; developing ways to deal with uncertainty and disagreement; and determining when a particular sub-task is completed. Again, this is not an exhaustive list of the types of task-organizing pro- cedures, but a definition by examples. Differentiation The participants are considered as differentiated from each other if the difference between their frequency of initiating speech acts was at least five percent. According to the particular group, this ranged from twenty to thirty acts for overall performance outputs, fourteen to twenty-three acts for problem-solving 19 attempts, and four to six acts for task-organizing pro- cedures. The status-prestige order is based on the final initiation proportion of overall performance outputs. The actor with the largest proportion of initiations was assigned the high status position; the actor with the next largest proportion of initiations, the middle status position; and the actor with the smallest proportion, the low status position. The Data The data consist of combined initiation proportions, obtained by adding together all of the individual initia- tions for each rank, cumulated over time, for the entire sample (or a subset), or for each discrete time period for the entire sample (or a subset), and then computing a proportion for each initiation rank. Three of the four groups differentiated into a status-prestige order.3 These three groups composed a differentiated subset, which could be compared with the entire sample for sig- nificant differences in performance. An alternative division of the entire sample consisted of two equal sub- sets. These two subsets were composed of those discus- sion groups which were equal in visible status charac- teristics (specifically race and gender), and those 20 groups which had visible inequalities of status charac- teristics. It was thought that this alternative division of the sample would allow for the identification of some possible sources which could confound the data. One final note: the sample is small. Hence many apparent findings may not be reliable. The reader should there- fore treat my conclusions as tentative and exercise caution in accepting them. The data are presented by two methods: cumulation over time, and discrete time periods. In the analysis cumulated over time, a general idea and the major fea- tures of trends appear, allowing the composition of a baseline structure of interaction against which signifi- cant fluctuations, resulting from either a coding arti- fact or of the interaction process, may be noted. A more precise description of the structure of this interaction can be seen in the analysis by discrete time periods. However, there is some arbitrariness in choosing a point for the division of the data. Intervals of ten minutes were chosen after it was determined that smaller inter- vals unnecessarily confounded the data. The proportions for the ranks add up to 1.00 at every interval. 21 Results Table 1 gives the final proportion of acts initiated by category for the entire sample and the three subsets of the entire sample. It is clear that most of the rele- vant acts coded were problem-solving attempts. Whether or not the differences in the proportions of the types of acts between those experiments which began with an absence of visibly unequal status characteristics and those which had visibly unequal status characteristics are meaningful cannot be determined by the data. TABLE 1. Final proportions of acts initiated by category Experiments Problem-Solving Task-Organizin% Attempts Procedures Entire Sample .79 .21 Subsets: Differentiated .79 .21 Equal Status Characteristics .77 .23 Unequal Status Characterstics .80 .20 22 Figures 1-1, 1-2, and 1-3 report the combined initiation rates cumulated over time for all ranks of the entire sample. Figure 1-1 shows the cumulated initiation proportions for overall performance outputs. It is obvious that the low status position is established within the first ten minutes of the discussion. The high and middle status positions emerge during the second ten-minute period. Once established, all status positions remain stable. Figure 1-2 indicates the cumulated initiation proportions of the status positions for the category of problem-solving attempts. The results are similar to those reported in Figure 1-1 since all initiation ranks correspond to their relative status positions, and they never overlap. Figure 1-3 presents the cumulated initiation proportions of the status positions for the category of task—organizing procedures. For the first thirty minutes of the interaction, the results are similar to those reported in Figures 1-1 and l-2, with the initiation ranks corresponding to their relative status positions. During the fourth ten-minute period, however, the low and middle initiation ranks converge to relative equality in their initiation rates. This situation remains throughout the observation period. The high status 23 position maintains the high initiation rank once established. 0 10 20 30 40 50 60 .50 .50 045 -' . '- .45 .40 g A ngh .40 ‘H C‘. ’\\ 0.9. .35 - \——-- ~~~~~ m. dgle..-- .35 83 .30 .30 as L 8.3 .25 - ”"',_______,__9y ------- - .25 04+) 020 """"""""""""""" .20 2:; ' n.m .15 y g .15 .oo 21” Jr .00 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 1-1. Performance outputs 0 10 20 30 40 50 60 .50 .50 045 -‘ High _' .45 .4 g .40 W .40 \ O O \.__.._._ . . ‘—-———_.__ Middle 543 035 "' ~~ ~.___" .35 .H Q‘.30 .30 +ha 6-3 -25 a ......... E 0.W_____ - .25 a.» .20 ,,,,, T .20 0L4 —————————— Lam aaaa 0494 .15 I” J; .15 .00 5:, If .00 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 1-2. Problem-solving attempts Entire Sample: Proportion of participation cumulated over time by each basic initiation rank 24 o 10 20 3o 40 so 60 .50 .50 .45 - - .45 .40 M ' .40 ‘H ‘1‘. Z \ Wr- 0-3 35 - \ \_ 35 813.30 '- \::’}£ddle 2:12: .30 '33 25 ‘ "TI Low 25 H O , - -------------- M’ b , 8.3.20 .20 SH mm.15 j, J: .15 .oo 1’ ’r .00 o 10 20 30 40 so 60 Elapsed time: ten-minute intervals Figure 1-3. Task-organizing procedures Entire Sample: Proportion of participation cumulated over time by each basic initiation rank Figures 1—4, 1-5, and 1-6 present the combined initi- ation rates for each discrete time period for all ranks of the entire sample. Figure 1-4 gives the discrete initiation proportions of the status positions for overall performance outputs. It is obvious that the low status position is differen- tiated during the first ten minutes of interaction. During the second ten-minute period, all three ranks are clearly differentiated. The initiation rate for the middle rank is reasonably constant. The rates of the high and low ranks fluctuate significantly during the 25 fourth ten-minute period when all ranks converge to rela- tive equality. The ranks then diverge again. Figure 1-5 gives the discrete initiation proportions of the status positions for the category of problem- solving attempts. The results are much the same as those reported in Figure 1-4. All initiation ranks correspond to their relative status positions, and they never over- lap. Figure 1-6 depicts the discrete initiation rates of the status positions for the category of task-organizing procedures. As shown in Figures 1-4 and 1-5, the low status position is again the low initiation rank for the first thirty minutes of interaction. Similarly, the high and middle status positions emerge during the second ten-minute period as the high and middle initiation ranks respectively. The ranks then overlap and the initiation rates fluctuate significantly. It is interesting to note the significantly increased initiation for the low status position during the fourth ten-minute period in problem- solving attempts and especially in task-organizing pro- cedures. 26 0 10 20 30 40 50 60 3 '50 Hi h 4'; .45 "" g - m 040 g / G \‘ z~\\\\\u////r ,3 ~35 " \\V// \ ,x ‘4‘ Middle /' JJ .30 I, \N‘V’ _g [’7 V‘-‘~ 4-, .25 -' I, ““~-- )- g .20 _“-—--"""'-J LOW * H .15 “'1’ 7 .00 ff fr 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 1-4. Performance outputs 0 10 20 30 40 50 60 m .50 . 3 .45 _ :x(////*\\\\\\‘\\ [/Eggh—_———_ g .40 ‘x g .35 - \\__ ___. fly . ,h- 'F'l .30 “*3:\ MILE-9W P .f e- 3 25 a I", “'x .. 53 .20 ,,,,,, -w-” Low “‘u c ...... H - .15 , 2b .00 5f 4r 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 1-5. Problem-solving attempts Entire Sample: period by each basic initiation rank. Proportion of participation by time .50 .45 .40 .35 .30 .25 .20 .15 .OO .50 .40 .35 .30 .25 .20 .15 .OO 27 O 10 20 30 4O 50 6O .50 8‘ 050 .45 d X ‘. L .45 .40 J' ‘ . LOW .40 m (U 4.) ‘ .. (u I \ oooo m .35 - (\ /\’I\ - .35 g .30 \ / .\ High .30 .3 \/ [7 \V ///"' 025 -‘ ~--~‘~...____-——- / . )— 25 (U o .3 .20 \/ Mlddle .20 -H C: 015 " 15 H ’ . .00 Ji * .00 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 1-6. Task-organizing procedures Entire Sample: Proportion of participation by time period by each basic initiation rank. Figures 2-1, 2-2, and 2-3 present the combined initiation proportions cumulated over time for all ranks of the differentiated subset. Figure 2-1 reports the cumulated initiation rates of all status positions for overall performance outputs. It is clear that the low status position is established during the first ten minutes of interaction. The high and middle status positions emerge during the second ten-minute period. Once established, the status posi- tions remain stable. Figure 2-2 presents the cumulated initiation rates of the status positions for the category of problem solving 28 attempts. The results are similar to those presented in Figure 2-1. All initiation ranks correspond to their relative status positions, and they do not overlap. Figure 2-3 shows the cumuluated initiation rates of the status positions for the category of task-organizing procedures. The results are mixed. All ranks become differentiated temporarily during the first ten-minute period. The low status position is the low initiation rank for the first thirty minutes of the discussion. There is, however, an overlapping of the rank order for the initiation rates of the high and middle status posi- tions at the end of the first ten minutes. The cumulated initiation rates for these two ranks then fluctuate sig- nificantly and converge for relative equality during the second ten-minute period. During the fourth ten-minute period, the cumulated initiation rates for all ranks con- verge to relative equality and remain undifferentiated throughout the remainder of the observation period. .50 .45 .40 .35 .30 .25 .20 Initiation Rates .15 .OO .50 .40 .35 .30 .25 .20 Initiation Rates .00 29 0 10 20 30 40 50 60 - /\ ngp' c\\ . _ \\___ ______ Midgle—u - Low b -------------------- a r 7 J4 3r 0 10 20 3O 40 50 60 Elapsed time: ten-minute intervals Figure 2-1. Performance outputs 0 10 20 30 4O 50 60 . Afr ‘\ .. \ \_ _ __ __ ___‘ Middle r- -t l— —"' LQL-tTF ;L ,-- JP" 4’ 4’ O 10 20 30 4O 50 60 Elapsed time: ten-minute intervals Figure 2-2. Problem-solving attempts Differentiated Subset: Proportion of participation cumulated over time by each basic initiation rank .50 .45 .40 .35 .30 .25 .20 .15 .00 .SO .40 .35 .30 .25 .20 .15 .OO 30 0 10 20 3o 40 50 60 .50 .50 m .45 4 B .45 3 .40 .40 £2 a .351 .35 o .30 .30 "-1 .p .3 .25 — 025 4.) .20 .20 -H 5 15 L 15 .004? 56.00 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 2-3. Task-organizingprocedures Differentiated Subset: Proportion of participation cumulated over time by each basic initiation rank Figures 2-4, 2-5, and 2-6 give the combined initia- tion proportions for each discrete time period for all ranks of the differentiated subset. Figure 2-4 presents the discrete initiation propor- tions of the status positions for overall performance outputs. It is obvious that the low status position is established during the first ten-minute period. The high and middle status positions are not differentiated until the second ten-minute period. The initiation rate for the middle rank is reasonably constant. The rates of the high and low ranks fluctuate significantly during the 31 fourth ten-minute period when all ranks converge to rela— tive equality. The ranks then diverge again. Figure 2-5 gives the discrete initiation proportions of the status positions for the category of problem- solving attempts. The results are similar to those reported in Figure 2-4. All initiation ranks correspond to their relative status positions, and they never over- lap. One difference is that only the high and middle initiation ranks converge, briefly, for relative equality in initiation rates during the fourth ten-minute period. Figure 2-6 presents the discrete initiation propor- tions of the status positions for the category of task- organizing procedures. A mere glance indicates that there is a great amount of fluctuation and overlap among the ranks and status positions throughout the observation period. It is especially interesting to note the sig- nificantly increased initiation rate for the low status position during the fourth and sixth ten-minute periods. 32 0 10 20 30 40 50 60 .50 A f U) 045 ‘ High I" B .40 g .\ ,_ V 035 '- / . l— g .30 \\ / .. \\Mfdle .,..| \f IL Ts‘ '\ 4.) I, ‘ \i .3 '25 ‘ v’ 1:9" +3 020 :4”'~~.“ 7/, .2 ',I ‘s‘~’ H 015 j? b .00 if if 0 10 20 30 40 50 60' Elapsed time: ten-minute intervals Figure 2-4. Performance outputs 0 10 20 30 40 50 60 .60 . .55 1 High _ .50 /\ m 045 '1 / \ / .- £3 .40 A m .35 fl \ //V\ . - C: \ /- \Middle 0 030 g z \‘ fl ‘\’/’ €::; .g .25 1 I’A~“ '- 4.) .20 L 2” \‘Low -r-l "a' ““ [I \~ .5 15 "H “I", \“‘ t .00 jf 4 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 2-5. Problem-solving attempts Differentiated Subset: Proportion of participation by time period by each basic initiation rank .50 .45 .40 .35 .30 .25 .20 .15 .00 .60 .55 .50 .45 .40 .35 .30 .25 .20 .15 .OO 33 10 20 30 40 50 69 .60 3’ .60 055 -' "I - .55 .50 [ .50 45 I” ‘\ LOW of 45 m . d A I, “ I, l- . 3 '40 \ A / \_ j ‘s, .40 g 4/:X< / \f “ X .35 ,I’ - .35 8 ’30 A \d/ / \L Afligh ~30 -~ "\\/’ \ l. y _______ .25 --" .25 .‘E. .20 I Mida1e\ .20 E 015 L " .15 H j V .00 4? 3’ .00 0 10 20 30 40 50 60 Figure 2-6. Elapsed time: ten-minute intervals Task-organizingprocedures Differentiated Subset: Proportion of participation by time period by each basic initiation rank Figures 3-1, 3-2, and 3-3 indicate the combined initiation proportions cumulated over time for all ranks of the initially equal status characteristics subset. Figure 3—1 shows the cumulated initiation rates of the status positions for overall performance outputs. The low status position is once more the first to be differentiated. The high and middle status positions emerge during the second ten-minute period. Once estab- lished, these positions remain stable. Figure 3-2 presents the cumulated initiation rates of the status positions for the category of problem-solving 34 attempts. The results are much the same as those reported in Figure 3-1. One dissimilarity is noted. During the first ten minutes of interaction, all three ranks become differentiated according to their respective status positions. Again, once established, the ranks remain stable. The cumulated initiation rates of the status posi- tions for the category of task-organizing procedures seen in Figure 3-3 indicate some differences in the kinds of participation among the status positions. The ranks for the high and middle status positions do not correspond to their respective status-prestige order. In addition, the ranks for these two status positions do not remain clearly differentiated. Only the low status position, differentiated during the second ten-minute period as the low initiation rank, remains stable. Initiation Rates Initiation Rates .50 .40 .35 .30 .25 .20 .15 .00 .50 .40 .35 .30 .25 .20 .15 .00 35 0 10 20 30 40 50 60 u N911 - “993’ L J“ v df 1 2r 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 3-1. Performance outputs o 10 20 3o 40 so 60 "l Nh b ‘x i \t ,_ Middle L \‘ _ _ .7" ’ - ,--..-----.1.-.<.>F I jb t 4 J4 o 10 20 3o 40 so 60 Elapsed time: ten-minute intervals Figure 3-2. Problem-solvingkattempts Equal Status Characteristics Subset: Proportion of participatiOn cumulated over time by each basic initiation rank. .50 .45 .40 .35 .30 .25 .20 .15 .00 .50 .45 .40 .35 .30 .25 .20 .15 .00 36 0 1o 20 3o 40 so 60 .50 **\ .50 8 .45 ‘ \\ / r- .45 g ~40 \ l!’ “-._ Middle .40 m ‘\/" "‘--—- .35 fl <_l o“ — 35 C1 . o .2 .30 /\ High .30 .p ........ .. .g 025 ' z --------------- LOW - .25 u .20 .20 '2 H 015 “L P; .15 .oo 2’ c} .00 o 10 20 30 4o 50 60 Elapsed time: ten-minute intervals Figure 3-3. Task-organizing!procedures Equal Status Characteristics Subset: Proportion of participation cumulated over time by each basic initiation rank. Figures 3-4, 3-5, and 3-6 indicate the combined initiation proportions for each discrete time period for all ranks of the equal status characteristics subset. Figure 3-4 gives the discrete initiation rates of the status positions for overall performance outputs. There is significant fluctuation in the initiation rates for all ranks and they do not remain clearly differentiated. During the fourth ten-minute period, the ranks converge for relative equality in initiation rates. The middle status position, however, is momentarily distinguished by a slight margin as the high initiation rank. 37 Figure 3-5 shows the discrete initiation rates of the status positions for the category of problem-solving attempts. The results are much the same as those reported in Figure 3-4, with two exceptions. Firstly, during the first ten minutes of the discussion, all ranks are distinct and correspond to their respective status positions. Secondly, the high initiation rank during the fourth ten-minute period is clearly the middle status position. In addition, it is interesting to note that when the high initiation rank corresponds to the high status position, it is sharply differentiated from the other ranks. Figure 3-6 presents the discrete initiation propor- tions of the status positions for the category of task- organizing procedures. Again, as in Figure 2—6, there is a great amount of fluctuation and overlap among the ranks and status positions throughout the observation period. It is worth pointing out the sharp increase in the initiation rate of the high status position for both categories during the second ten-minute period. .60 .55 .50 .45 .40 .35 .30 .25 .20 Initiation Rates .15 .00 .60 .55 .50 .45 .40 .35 .30 .25 .20 Initiation Rates .15 .00 Equal Status Characteristics Subset: 38 0 10 20 30 40 50 60 4 / \ /High _ \\ /,\ \ // f- \ . R / I. v ‘eungddle \ // I” LOW V a ”'—_J‘,‘ ...... 1’ .. 0 10 20 30 40 SO 60 Elapsed time: ten-minute intervals Figure 3-4. Performance outputs 0 10 20 3O 40 50 60 - P //// High \ A / -l \ r- \i // i .. A. ,” ~ '“Low d \ -_7A—/’ .\ ,— ,«” /’ Middle w l tr 4’ 0 10 20 30 40 50 6O Elapsed time: ten-minute intervals Problem-solvingiattempts Proportion of participation by time period by each basic initiation rank. .60 .55 .50 .45 .40 .35 .30 .25 .20 .15 .00 .60 .55 .50 .45 .40 .35 .30 .25 .20 .15 .00 39 0 10 20 30 40 50 60 .60 .60 055 - f '- .55 .50 A , \ .50 T I \ 3 .45 . \ / b .45 j; .40 \ \ .40 m / \ I ‘ igh : .35 "' \ ‘ \ .ddl 035 ,3 .30 ‘J/ .4\ ~ {M1 e .30 I .S‘. .25 1 ....... -‘ X, v/ 0" - .25 1:: .20 ‘\- L’ .20 015 -L )b 015 .00 A? If .00 D 10 20 30 40 SO 60 Elapsed time: ten-minute intervals Figure 3-6. Task-organizingprocedures Equal Status Characteristics Subset: Proportion of participation by time period by each basic initiation rank. Figures 4-1, 4-2, and 4-3 present the combined initiation rates cumulated over time for all ranks of the unequal status characteristics subset. Figure 4-1 reports the cumulated initiation rates of the status positions for overall performance outputs. The low status position is differentiated within the first ten minutes of interaction. The high and middle status positions, however, are not clearly distinguished until after more than thirty minutes of interaction. Once established, all status positions remain stable. 40 Figure 4-2 presents the cumulated initiation propor- tions of the status positions for the category of problem-solving attempts. As in the other subsets, the low status position becomes the low initiation rank with- in the first ten minutes of interaction and remains at that rank throughout the observation period. The high and middle initiation ranks are also distinguished within the first time period. These two initiation ranks, how— ever, do not correspond to the status-prestige order and do not remain clearly differentiated. Figure 4-3 indicates the cumulated initiation rates of the status positions for the category of task-organiz- ing procedures. The results are different from those reported in Figures 4-1 and 4-2. The high initiation rank is differentiated first, within the first ten minutes of the discussion. This rank is maintained by the high status position until the fifth ten-minute period. The middle and low initiation ranks do not emerge until the fourth ten—minute period, and they do not correspond to their respective status positions. By the end of the observation period, the high and middle initiation ranks have converged to relative equality. 41 0 10 20 30 40 50 60 m .50 B 045 - h g .40 -__:==‘:,_, <‘~“ High 8 035 "" \-_-—_._——'b 'H .30 Middle 4.) m '3 .25 q ,, """""""""""" L -H .20 ,,,, Low C: ¥ _______________ H .15 t .00 k ~ fir '6’ 10 20 30 4O 50 60 Elapsed time: ten-minute intervals Figure 4-1. Performance outputs 0 10 20 30 40 50 60 .50 h_ m .45 -' ‘~~'_-- '- 3 .40 \\~ Efihc: m g £__~_-_--— m .35 - / Middle L 8 .30 '13 m .25 4 Low ” 1:: .20 4a”--' -------------- J .3 ------- "‘_-.‘_..-"7' H .15 3V " J; .00 ,{ if 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 4-2. Problem-solvingattempts Unequal Status Characteristics Subset: Proportion of .50 .45 .40 .35 .30 .25 .20 .15 .00 .50 .45 .40 .35 .30 .25 .20 .15 .00 participation cumulated over time by each basic initiation rank. 42 0 10 20 30 40 50 60 .50 #y .50 g 045 d '—r \ - 045 g .40 High .40 c: 035 -------------- "' .35 ,x’ Low '9' 030 1‘,’ .30 4" :.::—\ __, -J” .‘3 .25 d “""’ ‘~"“~— ————— ‘T' -25 .4: .20 Middle .20 C H 015 “V b .15 .oo ’4 . Jr .00 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 4-3. Task-organizingprocedures Unequal Status Characteristics Subset: Proportion of participation cumulated over time by each basic initiation rank. Figures 4-4, 4-5, and 4-6 give the combined initia- tion proportions for each discrete time period for all ranks of the unequal status characteristic subset. Figure 4-4 shows the discrete initiation rates of the status positions for overall performance outputs. The ranks for the status positions are not constant, they do not remain distinct, and there is some overlapping. It is obvious that the low status position is established during the first time period. The high and middle status positions are not differentiated until the fourth ten- minute period. The high initiation rank, however, does not become separated from the other ranks until the fifth 43 ten-minute period. Moreover, all ranks are not clearly separated until the sixth ten-minute period when they correspond to their status positions. Only the initia- tion rate for the high status position is reasonably stable. Figure 4-5 gives the initiation proportions of the status positions for the category of problem-solving attempts. The results are roughly comparable to those in Figure 4-4, although there are some noteworthy dif- ferences. All ranks differentiate during the first time period, but only the low initiation rank corresponds to its status position. The high initiation rank emerges as the high status position during the fourth ten-minute period. It is not until the fifth ten-minute period when all ranks become distinct and correspond to their respec- tive status positions. Again, the initiation rate for the high status position is reasonably constant. Figure 4-6 presents the discrete initiation proportions of the status positions for the category of task—organizing procedures. The ranks for the status positions are not constant and they overlap. Although the initiation rank changes, the initiation rate for the middle status posi- tion is reasonably constant. It is especially interest- ing to note the change in initiation ranks of the high and low status positions during the fourth ten-minute 44 period as a result of sharp fluctuations in their initiation rates. 0 10 20 30 40 50 60 .50 m .45 - Hi h ‘ a“? ’ “~\\/ C: 035 J \ ”I s‘\ Midd-lze .—:- .3 .30 ’h’ ‘§<\ +1 . \xz’ ‘~~. g .25 d 1” Low \‘NT" 44 .20 - 4,-" w-i ...... 7 5 15 4 M” 0 v F .00 3’ X 0 10 20 30 40 SO 60 Elapsed time: ten-minute intervals Figure 4-4. Performance outputs 0 10 20 30 40 SO 60 .50 - . 3 - \ \g M 4f; 040 N~‘ m V 35 1 / \ Middle , C: ' \ ~~—._ .3 .30 “o"\ / 4.) ’I’ ‘/"\‘ .2 .25 - an“ .. f, .20 I,” Low ---- a ~~~ ’I” H .15 a «a... - .00 ZI’ 51’ 0 10 20 30 40 50 60 Elapsed time: ten-minute intervals Figure 4-5. Problem-solvingattempts Unequal Status Characteristics Subset: Proportion of participation by time period by each basic initiation rank. .50 .40 .35 .30 .20 .15 .00 .50 .45 .40 .35 .30 .25 .20 .15 .00 45 0 10 20 30 40 50 60 .60 ,5. 055 - I, “\ F" 050 A I, ‘.‘IIOW m ’1' "x g .45 d I” ‘\‘ F- m .40 I ‘ER a 35 l‘\\ l O o d I, Hi h P 2:: .30 4I” k .3 r— —\ I”, // JJ .25 d ........ 1’ \ / .— .El .20 \ 4’ ; 4/ H f», f ‘ J .15 a Middle _ .00 5f 3? O 10 20 30 40 50 60' Elapsed time: ten-minute intervals Figure 4-6. Task-organizingprocedures Unequal Status Characteristics Subset: Proportion of participation by time period by each basic initiation rank. .60 .55 .50 .45 .40 .35 .30 .25 .20 .15 .00 CHAPTER IV DISCUSSION AND CONCLUSIONS The central problem of this study was to describe the development of a status-prestige order and its relation- ship to the emergence of procedural rules in certain small groups. An examination of the data presented here for the entire sample and all subsets suggests that such a relationship exists; yet the exact nature of that asso- ciation cannot be established empirically with these data. Nevertheless, the results reported here point to several additional interesting features. Ranking The figures for the cumulated initiation rates clearly indicate that the status-prestige order develops early in the discussion4 and remains stable throughout the observation period. The low status position is differentiated first, within the first ten minutes of interaction. Then, during the second ten—minute period, there is a marked increase in the initiation rate, both in problem-solving attempts and in task-organizing procedures,5 of the emerging high status position. The status-prestige order is fully developed and stable from this point. 46 47 It is clear that this status-prestige order corres- ponds to the rank-order of problem-solving attempts, but not to that of task-organizing procedures. In addition, the rank-order for problem-solving attempts, but not for task-organizing procedures, remains stable once estab- lished. It is also interesting to note that the low status position does not remain stable in the low initia- tion rank for task-organizing procedures. Some curious variations are indicated by these results. The data, when examined more closely in the figures for discrete time periods, reveal significant fluctuations in the initiation rates. These fluctuations from relatively constant rates are particularly apparent, but not limited to, the duration of the fourth ten-minute interval for the high and low status positions. At the risk of relying upon personal knowledge of the transcripts, a brief explanation is offered to help shed light on these patterns of interaction disclosed by the data. Phases One interpretation of this variability found in the data can be thought of roughly in terms of phases in the overall conversational organization and in the notion of 48 topicality within these different phases. I draw briefly here on the notion of phases in group problem solving by Bales and Strodtbeck (1951) and to a lesser extent, on the idea of topicality for the ethnomethodologists (Schegloff and Sacks, 1974:242). By “phases,” Bales and Strodtbeck (1951:386) mean ”qualitatively different subperiods within a total con- tinuous period of interaction in which a group proceeds from initiation to completion of a problem involving group decision.” They posit a phase hypothesis concern- ing the overall conversational organization which states ”. . . that under [certain] conditions groups tend to move in their interaction from a relative emphasis upon problems of orientation, to problems of evaluation, and subsequently to problems of control . . .' (1951:387. Emphasis original.) Since many of the categories for problem-solving attempts and task-organizing procedures utilized in the present study were developed with Bales' categories in mind, a similar change in the relative emphasis on the initiation of certain speech acts can be seen in the data. Using a similar analysis, it seems that what is hap- pening in these groups is that the relative emphasis of the different types of speech acts (problem-solving attempts and task—organizing procedures) is changing over 49 time with the change in topic and with the development of a status-prestige order. In particular, Bales' cate- gories for problems of orientation and evaluation approx- imately correspond to the present categories of problem- solving attempts, while the categories associated with problems of control most closely resemble task-organizing procedures. Further, as the topical structure changes from one topic to the next during what can be understood as the second phase, there is more emphasis on problems of control over how to determine when one item of discus- sion is finished and the group is ready to move on to the next one. This also can be thought of in the more general terms of a larger structure concerning issues, which, in turn, are composed of several items. The reader will recall that the instructions given each group were to develop a list of criteria 22g to rank those criteria according to importance. A careful peru— sal of the transcripts suggests that the groups seem to have divided their assigned task into two distinct parts or sub-tasks and then discussed each part separately. Because of the logic and content entailed, these two phases can be called descriptively the Reality Construc- tion Phase and the Book Keeping Phase. During the Reality Construction Phase, the partici- pants talk in a general manner about what kind of person 50 they think is deserving of money and what kind is not deserving. In short, they construct two ideal types of people. It is during this first phase (in which the second ten-minute interval is included),6 that the status-prestige order develops.7 The Book Keeping Phase begins after the group has made some decisions about what the criteria should be for the allocation of money. Ranking those criteria takes place during this phase of the discussion. Since these two phases are concerned with two dif- ferent types of tasks, it can further be conjectured that different modes of thinking would be necessary to corres- pond to these two phases. By dividing the overall struc- ture of the discussion into two such distinct parts or phases, the data can be examined for any relative emphas- is given to different types of speech acts during time periods tentatively corresponding to these two phases. By thinking in terms of qualitatively different phases in the overall structure of the conversation, we can further speculate that there are also differences in types of task-organizing procedures corresponding to the primary functions or purposes of these different phases. If this reasoning is sound, then we would expect to find that different types of task-organizing procedures are more prevalent during different phases of the discussion. 51 In addition, there is expected to be a differential distribution among the status positions in the initiation of the different task-organizing procedures. Differential Distribution of Types of Acts The present coding scheme did make distinctions of various types of task-organizing procedures; however, these distinctions were not recorded. This content dis— tinction can be seen in the following (hypothetical) task-organizing procedures: (1) ”Would you repeat that please?”; or, ”I don't understand what you mean"; (2) ”What's next?'; and (3) "First we should decide on all the criteria we are going to consider and then we should rank them." Clearly, the content and relative importance of each of these types of separate speech acts cannot be considered equal; they were however, each counted as a single task—organizing procedure. The first task- organizing procedures are concerned with individual requests for a performance from another actor. The second task-organizing procedure is a kind of speech marker indicating that the previous item of discussion has ended and the next subject for consideration has not been determined. The last act is qualitatively different in that it is proposing a "rule” or general principle for organizing the group's efforts. It seems logical that 52 more of this latter type of task-organizing procedure would be found near the beginning of the discussion. This would allow for greater efficiency in completing the assigned task. It is beyond the scope of this report to further analyze the parameters of the various types of task-organizing procedures that occur. The figures for the discrete time periods do indicate a differential distribution of problem-solving attempts and task—organizing procedures among the status positions during different time intervals of the observation period. More specifically, and of greater interest here, an examination of the initiation rate of task-organizing procedures for discrete time intervals reveals signifi- cant fluctuations. The increase in the initiation rate for task-organiz- ing procedures (and for problem-solving attempts) during the second ten-minute interval (the Reality Construction Phase) for the emerging high status-prestige position is followed by its establishment in that position. If the distinctions in task-organizing procedures had been recorded, I would expect to find those which act as general procedural "rules” or general principles for organizing the group discussion to occur early in the interaction. These acts would be concerned with ordering 53 of the task sub-units and establishing methods for allo- cating time and energy to the task. It has already been pointed out that the low status- prestige position does not remain stable in the low ini- tiation rank for task-organizing procedures. The data show a pointed increase in the initiation rate for the low status-prestige position in task-organizing proce- dures (and in problem-solving attempts), about half-way through the observation period, usually during the fourth ten-minute period which corresponds to the Book Keeping Phase. One possible explanation is that the low status position at this point in the interaction is acting as a sort of discussion facilitator by keeping the discussion going: that is, in moving from one item of discussion to the next. Again, if these differences in task-organizing pro- cedures had been recorded, I would expect to find those which deal with moving from one item to the next thereby determining when a particular item of discussion is com- pleted. It seems fair to say that the low status- prestige position plays a special role in the group dis- cussion, but a more precise description of that role is unclear from these data. l ‘11.: I.- 54 Structural Change Along with the above consideration, the decrease in performance outputs (task-organizing procedures and problem-solving attempts) for the high status-prestige position during the fourth ten-minute interval (the Book Keeping Phase) must also be addressed. It can be specu- lated that the high status position is shared with the other positions.8 It should be pointed out that the status-prestige order was previously established and does remain stable according to the data for the cumulated initiation rates. Although this project does not adhere to the strict conditions for the terms of Expectation States Theory, Berger and Conner (1974:104) offer an explanation for a change in status structures which may be of similar interest here. In discussing the maintenance of expecta- tions, they assume that if the expectation structure changes, there must be an alteration of a scope condition or the introduction of a new condition. They suggest that if the ”task focus begins to weaken and the group becomes more process-oriented, the whole expectation structure may undergo change.” It seems that a similar explanation may help to understand this great fluctuation in relative initiation rates during this time. Since the 55 task of the group has changed from deciding on the cri- teria to ranking it, it is not inconceivable that the structure of the group interaction may have undergone a change reflecting this new condition. Issues The first issue to be commented on is that only the initiator, and not the target, of a performance output was recorded. This procedure is not without precedent. 1! Fisek (1974:72) identifies the high status position by the frequency of performance outputs initiated and ignores the targets of those acts. Similarly, McMahon et al. (1979) record only the frequency of certain acts initiated by each actor. When the target of these per- formance outputs is ignored, however, certain types of information and data become unavailable. One additional point needs to be mentioned concerning the data in relation to the coding scheme utilized in this study. If influence is a result of acceptance of performance outputs by the other group members (see Berger and Conner, 1974:88), then not all performance outputs initiated are equal in meaning. Obviously, if this is the case, a performance output that is accepted will have more weight that one which is rejected. The 56 present coding scheme did not recognize acceptance or rejection of performance outputs. Finally, it was previously remarked that neither gender nor race was completely controlled. This resulted in breaking down the entire sample into two subsets according to equality or inequality of visible status characteristics. An examination of the results from the unequal status characteristics subset shows that the dif- ferentiation of the high and middle status positions and initiation ranks did not take place until significantly later than the subset for equal status characteristics. The high and middle status positions seemed to emerge, according to the data, during the fourth ten-minute period. In addition, the degree of differentiation between these two positions is not as pronounced as it is in the equal status characteristics subset. It is also questionable whether or not the high and middle initia- tion ranks did differentiate in the category of problem- solving attempts. In the category of task-organizing procedures, it appears that the high and low status posi- tions share the high initiation rank at the end of the observation period. Further, a comparison of the unequal subset to the equal subset shows that the latter com- pleted their discussion in less than 60 minutes. This seems to indicate that these differences in visible j 57 status characteristics do have an effect on the structure of the discussion. CHAPTER V SUMMARY This pilot study began with a secondary analysis of existing transcripts from a laboratory project originat- ing from an ethnomethodological perspective. The concern here, however, was with the development of a status- prestige order and its relationship to the emergence of general principles for discussion organization. This idea stemmed largely from some of the literature on Expectation States Theory by Berger, Conner, and Fisek (1974). Although there were some obvious difficulties with the fit of the experimental design to the problem at hand, some tentatively interesting results did become apparent. Simple initiation proportions for all ranks revealed four main points. First, the status-prestige order develops early in the discussion and the low status posi- tion is identified first. Second, an increase in the initiation rate for task-organizing procedures in the second ten-minute period is followed by the establishment of the high status position. The status-prestige order then remains stable. Third, the status-prestige order 58 59 corresponds to that of problem-solving attempts, but not task-organizing procedures. Fourth, the low status posi- tion does not remain stable in the low initiation rank for task-organizing procedures. What is not clear is a more precise description of the relationship between the emergence of the high status position and the develOpment of some kind of general organizing principles. Similarly, the role of the low status position in regard to the issue of topicality or topic talk (Schegloff and Sacks, 1974:242) is unclear. In addition, differences in interaction patterns for groups with visible inequalities in status characteris- tics when compared to groups lacking those same visible inequalities in status characteristics were noted. What needs to be done is to gather additional evi- dence in a more rigorous manner. The concepts used for this study can easily be modified and refined to fit more closely those of Expectation States Theory. This would include an expanded code book, taking into account not only more categories of speech acts, but also the targets of those acts. In addition, this notion of topicality or topic talk, needs to be examined not only from the social psychological perspective presented here, but also from a phenomenological-ethnomethodological perspective. This 60 would help us gain a better understanding of those trans- formation points where the (status expectation) structure changes. NOTES 1. It should be noted that although the terms used here are borrowed mainly from Berger and Conner (1969: 1974) and Conner (1975), they are freely interpreted and adapted for present purposes. I make no claim that the terms defined here necessarily correspond in a strict sense to the types of analysis that Berger and Conner had in mind in relation to Expectation States Theory. Such a strict adherence to the conditions for the terms of Expectation States Theory is beyond the scope of this project. 2. Since the research project reported in this study is concerned only with the interaction in Part One, the procedures for the second part will not be related. 3. The members of the undifferentiated groups, for identification purposes only, were designated by the same criteria. 4. The subset of unequal status characteristics does not develop a high status position until significantly later in the discussion. More on this aspect will be mentioned at the end of this section. 5. There is one exception. In the category of task- organizing procedures for the unequal status characteris- tics subset, the high initiation rank maintained the already sharply differentited initiation rate of .45. 6. Any future study would have to determine exactly the boundary markers of these phases. Because there was no fixed time schedule such as a definite end point in clock-time for the discussants to reach an agreement, this phase lasted different lengths of time, depending upon the conflict or agreement among the discussants. 7. Except for one group of the unequal status characteristics subset. 8. Fisek (1974:58) reminds us that the status-prestige order is assumed to be a deference order rather than a dominance order. It is plausible that the high status position is giving opportunities to talk to the other positions. One example of how this can happen is by 61 62 pausing. Pausing practices or silence can be considered a speech act. Indeed, Fisek (1974:57) states that silence “may be the most common kind of action oppor- tunity: an actor who pauses during a speech, or simply remains silent, is offering the other actors in the group a chance to perform . . .' Any future study would have to take pausing practices and silence into account for a more complete understanding of changes in status struc- tures. LIST OF REFERENCES LIST OF REFERENCES Anderson, Bo and Richard Hurst. 1977. "Formulating Agreements About Rewards in Small Groups.” Paper presented at the meetings of the North Central Sociological Association in Pittsburgh. Bales, R.F. 1950. Interaction Process Analysis: A Method for the Study of Small Groups. Cambridge, Mass.: Addison-Wesley Press. Berger, Joseph. 1974. "Expectation States Theory: A Theoretical Research Pogram.” In Expectation States Theory: A Theoretical Research Program, edited by J. Berger, T.L. Conner, and M.H. Fisek. Cambridge, Mass.: Winthrop Publishers. Berger, Joseph and T.L. Conner. 1969. “Performance Expectations and Behavior in Small Groups.” Acta Sociologica, v. 12, no. 4:186-98. . 1974. ”Performance Expectations and Behavior in Small Groups: A Revised Formulation.‘I In Expectation States Theory: A Theoretical Research Program, edited by J. Berger, T.L. Conner, and M.H. Fisek. Cambridge, Mass.: Winthrop Publisher. Berger, Joseph, B.P. Cohen, and M. Zelditch, Jr. 1966. ”Status Characteristics and Expectation States.” In Sociological Theoriesjin Prggress, edited by J. Berger, M. Zelditch, Jr. and B. Anderson. Volume 1. Boston: Houghton Mifflin Co. Conner, T.L. 1975. "Revision of the Behavior Categories of Expectation States Theory," photocopy, Department of Sociology: Michigan State University. Fisek, M.H. 1974. "A Model for the Evolution of Status Structures in Task-Oriented Discussion Groups." In Expectation States Theory: A Theoretical Research Program, edited J. Berger, T.L. Conner, and M.H. Fisek. Cambridge, Mass.: Winthop Publisher. 63 64 McMahon, A.M., P. Barchas, E. Cohen, P. Hildebrand, and M. Fennel. 1979. "Organizational Task Performance in Male and Female Groups,” photocopy, Department of Sociology, Stanford University. Scheflen, A.E. 1974. How Behavior Means. New York: J. Aronson. Schegloff, E. and H. Sacks. 1974. ”Opening up Closings.” In Ethnomethodology, edited by R. Turner. Baltimore: Penguin Books.