The Measurement of Group Process and Effectiveness in Educational and Work Settings (original) (raw)
Related papers
Group decision making: friend or foe?
2003
Recognized as essential to modern business, collaborative meetings are held to make decisions about product design, market definition, employee hiring, and many other factors that affect business operations. Based on the literature on group decision-making (a single collaboratively assigned selection from an array of choices), several hypothesis have been generated for further research based on three fundamental problems with group decision-making: impact of initial preference and time pressure, impact of group dynamics and politics, and the lack of an objective quantitative group preference. We suggest that group decision making should not be utilized unless there are objective criteria (i.e., money, time) that allow the group members to evaluate decision choices for determination of group preferences. Without these objective criteria, the use of group decision-making should be reserved for use in option generation (i.e. brainstorming), problem solving, or creative processes of an exploratory nature.
Group Performance and Decision Making
Annual Review of Psychology, 2004
Theory and research on small group performance and decision making is reviewed. Recent trends in group performance research have found that process gains as well as losses are possible, and both are frequently explained by situational and procedural contexts that differentially affect motivation and resource coordination. Research has continued on classic topics (e.g., brainstorming, group goal setting, stress, and group performance) and relatively new areas (e.g., collective induction). Group decision making research has focused on preference combination for continuous response distributions and group information processing. New approaches (e.g., group-level signal detection) and traditional topics (e.g., groupthink) are discussed. New directions, such as nonlinear dynamic systems, evolutionary adaptation, and technological advances, should keep small group research vigorous well into the future.
Group decision-making processes and group decision quality: moderation of mutual interest
Considering the existence of mutual interest in work groups within organisational context, we investigated three main decision-making processes: Personal task participation, negative socio-emotional behaviour and information domination. To illustrate the relationship between group decision-making processes and decision quality, we developed and tested a process model in which the final group decision quality could be predicted by abovementioned decision-making processes. Testing the moderation effect of mutual interest is considered as focus of study and its implications are discussed. Further expansion of our model involves process satisfaction as an intervening variable. The overall assessment of the group decision-making process was captured in the process satisfaction construct and its association with decision quality was analysed. The findings suggested that our 1st model supported the moderation of mutual interest with one exception. The findings of the 2nd model suggested that satisfaction of the group members about decision processes is strongly associated with decision quality.
Journal of Managerial Psychology, 2011
Purpose -Hidden profiles are decision-making tasks in which groups have the potential to outperform individual decision-makers. This paper has two purposes: first, to provide a conceptual analysis of how the group potential for solving hidden profiles can be measured; second, to empirically determine the solution rates hidden profile groups would achieve: in the absence of any group processes (i.e. the group potential); and in the absence of any dysfunctional group processes. Design/methodology/approach -The group potential was determined by averaging the group members' decision quality prior to the discussion. To determine the hidden profile solution rates in the absence of any dysfunctional group processes, the standard hidden profile procedure was modified so that nothing but the individual-level constraints could hamper the solution of hidden profiles. Findings -The actual group performance was significantly higher than the group potential, but significantly lower than the performance in the no dysfunctional group processes condition. Hence, dysfunctional group processes interfere with the realization of process gains. However, even in the absence of any dysfunctional group processes, groups did not always solve hidden profiles. Finally, the detrimental group process hampering the solution of hidden profiles does not seem to be biased information pooling favoring shared information but rather insufficient amount of information pooling. Practical implications -The results indicate that tools, which aim to facilitate the solution of hidden profiles, have to overcome both dysfunctional group processes, and individual-level constraints. Originality/value -This is the first attempt to quantify process gains in hidden profile groups.
Organizational Behavior and Human Decision Processes, 1996
In response to growing demands for efficiency and This study examines the role of group composition flexibility, organizations are implementing teams to do and information distribution on group process and demuch of the work traditionally accomplished by indicision making. Three-person groups performed a decividuals (Boyett & Conn, 1992; Katzenbach & Smith, sion task that involved solving a murder mystery (Stas-1993). In part, this strategy is based on the assumption ser & Stewart, 1992). Groups were composed of (a) that decisions made by groups of employees with diverthree individuals familiar to each other, (b) two familsified expertise will be higher in quality than those iar individuals and a stranger, or (c) three strangers. made by employees with more homogenous back-Prior to group discussion, evidence bearing on the grounds, or by any one employee (e.g., a manager) who case was either fully shared (all members possessed might have access to the same knowledge but whose identical information) or partially shared (each member possessed several unique clues to which no other single organizational perspective would be more limmember had access). The results indicate that allited (Jackson, 1992). stranger groups were most likely to identify the cor-In practice, unfortunately, these efforts are not alrect suspect when information was fully shared, howways fruitful (Hackman, 1990). In groups that form ever, all-familiar and 2 familiar/1 stranger groups were through natural selection, the most common bases of most likely to identify the correct suspect when critimember attraction are similarity (Ancona & Caldcal clues remained unshared. Group process analysis well, 1992; Newcomb, 1961), proximity (Festinger, reveals that this pattern of results was due to an ''ag-Schachter & Bach, 1950), and prior acquaintance gregation strategy'' on the part of strangers and an (Mannix, Goins, & Carroll, 1996; Tenbrunsel, Wade-''information pooling strategy'' on the part of groups composed of familiar individuals. ᭧ 1996 Academic Press, Inc. Benzoni, Moag, & Bazerman, 1994). These processes, while maximizing relationship potential, often minimize the potential for learning. The knowledge and peron the faculty of the Graduate School of Business at the University problem-solving effectiveness (Jackson, 1992). of Chicago; the third author, a doctoral student at Northwestern University; and the fourth author, a faculty member in the Kellogg A common organizational response to this problem Graduate School of Management at Northwestern University. Adis to design cross-functional teams, combining repredress reprint requests to Professor Deborah Gruenfeld, Department sentatives of different organizational functions to inof Organization Behavior, Kellogg Graduate School of Management, sure diversity in knowledge and perspectives. But
Organizational Behavior and …, 1976
Norms controlling how members deal with performance strategies were altered experimentally in small task-oriented groups. The basic task required assembly of small electrical components. In one task condition (equal information) all task-relevant information was provided to each group member; in another (unequal information) it was spread unevenly among members, requiring exchange of information for optimum group performance. In the unequal information condition, an intervention inducing explicit discussion of task performance strategies improved group performance. In the equal information condition, effectiveness was increased by an intervention that reinforced existing norms against explicit discussion of performance strategies. Spontaneous discussion of strategy did not take place in control groups for either task condition, and control groups were lowest in performance effectiveness. Measures of interaction process and of member reactions to the group were affected substantially by the experimental interventions.
Analyzing group work processes: Towards a conceptual framework and systematic statistical analyses
2001
Each action during group work has many effects, and their effects can vary over time. Thus, a systematic analysis of group processes requires a multi-dimensional conceptual framework and dynamic statistical tools. Past research on group work has typically focused on either cognitive or socio-emotional aspects using coarse levels of analyses. In contrast, this chapter presents a new framework of individual actions that combines cognitive and socio-emotional aspects. This framework organizes each action along five dimensions. These are evaluation (agree, disagree, ignore), knowledge (contribution, repetition, null), validity (right, wrong, unrelated) invitation (command, question, statement), and politeness (give face, neutral, threaten face). In particular, evaluations, repetitions, invitations and politeness link actions together to create coherent interactions. As a result, they help measure how well a group works together. Dynamic analysis of group processes faces at least six difficulties. First, coding difficulties reduce inter-coder reliability and the precision of parameter estimates. Second, effects can differ for each group. Third, effects can change over time. Fourth, we must identify time periods with stable effects. Fifth, the outcome variables can be discrete rather than continuous. Lastly, serial correlation of the residuals can occur. This chapter presents a new method that addresses all of these issues: multilevel Logit with time series analyses. A study shows how group processes are modeled using this framework and this method.
A field experiment investigated whether group performance is improved when the subjects a) individually conduct an evaluation of task difficulty before entering the process of group decision making and/or b) know about a result from group decision making research which suggests that conformity has a positive impact in easy tasks but negative impact in difficult tasks (Grofman, 1978; Thomas & Fink, 1961). Twenty-eight groups of three or four persons worked on judgmental tasks related to an environmental planning project in which they were participating for several months. Group judgements were significantly improved when subjects were advised to evaluate the task difficulty before entering group decision making compared to a set of control groups. The interpretation focuses on cognitive aspects of the evaluation procedure, which might counteract anchoring effects and enhance the generation of arguments. Though the trend was in line with the hypothesis, there was no significant result with respect to the information on the differential effects of conformity pressure in easy versus difficult tasks. A post-experimental questionnaire revealed that most of the subjects did not know about the differential effects of group conformity but considered a task difficulty analysis to be helpful and supportive for group decision making.