ZeePedia buy college essays online


Organization Development

<<< Previous Diagnosing Groups and Jobs:Design Components, Outputs Next >>>
 
img
Organization Development ­ MGMT 628
VU
Lesson 18
Diagnosing Groups and Jobs
Diagnosis is the second major phase in the model of planned change. Based on open-systems theory, a
comprehensive diagnostic framework for organization-, group-, and job-level systems was discussed. The
organization-level diagnostic model was elaborated and applied. After the organization level, the next two
levels of diagnosis are the group and job. Many large organizations have groups or departments that are
themselves relatively large. Diagnosis of large groups can follow the dimensions and relational fits
applicable to organization-level diagnosis. In essence, large groups or departments operate much like
organizations, and their functioning can be assessed by diagnosing them as organizations.
Small departments and groups, however, can behave differently from large organizations and so they need
their own diagnostic models to reflect those differences. In the first section, we discuss the diagnosis of
work groups. Such groups generally consist of a relatively small number of people working face-to-face on
a shared task. Work groups are prevalent in all sizes of organizations. They can he relatively permanent and
perform an ongoing function, or they can be temporary and exist only to perform a certain task or to make
a specific decision.
Finally, we describe and apply a diagnostic model of individual jobs--the smallest unit of analysis in
organizations. An individual job is constructed to perform a specific task or set of tasks. How jobs are
designed can affect individual and organizational effectiveness.
Group-Level Diagnosis:
Figure 24 replicates the comprehensive model discussed earlier but highlights the group- and individual-
level models. It shows the inputs, design components, outputs, and relational fits for group-level diagnosis.
The model is similar to other popular group-level diagnostic models, such as Hackman and Morris's task
group design model, McCaskey's framework for analyzing groups, and Ledford, Lawler, and Mohrman's
participation group design model.
Inputs:
Organization design is clearly the major input to group design. It consists of the design components
characterizing the larger organization within which the group is embedded: technology, structure,
measurement systems, and human resources systems, as well as organization culture. Technology can
determine the characteristics of the group's task; structural systems can specify the level of coordination
required among groups. The human resources and measurement systems, such as performance appraisal
and reward systems, play an important role in determining team functioning. For example, individually
based performance appraisal and reward systems tend to interfere with team functioning because members
may be more concerned with maximizing their individual performance to the detriment of team
performance. Collecting information about the group's organization design context can greatly improve the
accuracy of diagnosis.
Design Components:
Figure 24 (B) shows that groups have five major components: goal clarity, task structure, group
composition, group functioning, and performance norms.
Goal clarity involves how well the group understands its objectives. In general, goals should be moderately
challenging; there should be a method for measuring, monitoring, and feeding back information about goal
achievement; and the goals should be clearly understood by all members.
img
Organization Development ­ MGMT 628
VU
Figure 24: Comprehensive Model for Diagnosing Organizational Systems
Task structure is concerned with how the group's work is designed. Task structures can vary along two key
dimensions; coordination of members' efforts and regulation of their task behaviors. The coordination
dimension involves the degree to which group tasks are structured to promote effective interaction among
group members. Coordination is important in groups performing interdependence tasks, such as surgical
teams and problem-solving groups. It is relatively unimportant, however, in groups composed of members
who perform independent tasks, such as a group of telephone operators or salespeople. The regulation
dimension involves the degree to which members can control their own task behaviors and be relatively
free from external controls such as supervision, plans, and programs. Self-regulation generally occurs when
members can decide on such issues as task assignments, work methods, production goals, and membership.
Composition concerns the membership of groups. Members can differ on a number of dimensions having
relevance to group behavior. Demographic variables, such as age, education, experience, and skills and
abilities, can affect how people behave and relate to each other in groups. Demographics can determine
whether the group is composed of people having task-relevant skills and knowledge, including
img
Organization Development ­ MGMT 628
VU
interpersonal skills. People's internal needs also can influence group behaviors. Individual differences in
social needs can determine whether group membership is likely to be satisfying or stressful.
Group functioning is the underlying basis of group life. How members relate to each other is important in
work groups because the quality of relationships can affect task performance. In some groups, for example
interpersonal competition and conflict among members result in their providing little support and help for
each other. Conversely groups may become too concerned about sharing good feelings and support and
spend too little time on task performance. In organization development considerable effort has been
invested to help work group members develop healthy interpersonal relations, including ability and a
willingness to share feelings and perceptions about members' behaviors so that inter-personal problems
and task difficulties can he worked through and resolved. Group functioning therefore involves task-related
activities, such as giving and seeking information and elaborating, coordinating, and evaluating activities;
and the group-maintenance function, which is directed toward holding the group together as a cohesive
team and includes encouraging, harmonizing, compromising, setting standards, and observing.
Performance norms are member beliefs about how the group should perform its task and include
acceptable levels of performance. Norms derive from interactions among members and serve as guides to
group behavior. Once members agree on performance norms, either implicitly or explicitly, then members
routinely perform tasks according to those norms. For example, members of problem-solving groups often
decide early in the life of the group that decisions will be made through voting; voting then becomes a
routine part of group task behavior.
Outputs:
Group effectiveness has two dimensions: performance and quality of work life. Performance is measured in
terms of the group's ability to control or reduce costs, increase productivity, or improve quality. This is a
"hard" measure of effectiveness. In addition, effectiveness is indicated by the group member's quality of
work life. It concerns work satisfaction, team cohesion, and organizational commitment.
Fits:
The diagnostic model in Figure 24(B) shows that group design components must fit inputs if groups are to
be effective in terms of performance and the quality of work life. Research suggests the following fits
between the inputs and design dimensions:
1.
Group design should be congruent with the larger organization design. Organization structures
with low differentiation and high integration should have work groups that are composed of highly
skilled and experienced members performing highly interdependent tasks. Organizations with
differentiated structures and formalized human resources and information systems should spawn
groups that have clear, quantitative goals and support standardized behaviors. Although there is
little direct research on these fits, the underlying rationale is that congruence between organization
and group designs support overall integration within the company. When group designs are not
compatible with organization designs, groups often conflict with the organization. They may
develop norms that run counter to organizational effectiveness, such as occurs in groups
supportive of horseplay, goldbricking, and other counterproductive behaviors.
2.
When the organization's technology results in interdependent tasks, coordination among members
should be promoted by task structures, composition, performance norms, and group functioning.
Conversely when technology permits independent tasks, the design components should promote
individual task performance. For example, when coordination is needed, task structure might
physically locate related tasks together; composition might include members with similar
interpersonal skills and social needs; performance norms would support task-relevant interactions;
and healthy interpersonal relationships would be developed.
3.
When the technology is relatively uncertain and requires high amounts of information processing
and decision making, group task structure, composition, performance norms, and group
functioning should promote self-regulation. Members should have the necessary freedom,
information, and skills to assign members to tasks, to decide on production methods, and to set
performance goals. When technology is relatively certain, group designs should promote
standardization of behavior, and groups should be externally controlled by supervisors, schedules,
and plans. For example, when self-regulation is needed, task structure might be relatively flexible
and allow the interchange of members across group tasks; composition might include members
with multiple skills, interpersonal competencies, and social needs; performance norms would
support complex problem solving; and efforts would be made to develop healthy interpersonal
relations.
img
Organization Development ­ MGMT 628
VU
Application 3: Top-Management Team at Ortiv Glass Corporation
The Ortiv Glass Corporation produces and markets plate glass for use primarily in the construction and
automotive indulines. The multiplant company has been involved in OD for several years and actively
supports participative management practices and employee involvement programs. Ortiv's organization
design is relatively organic, and the manufacturing plants are given freedom and encouragement to develop
their own organization designs and approaches to participative management. It recently put together a
problem-solving group made up of the top-management team at its newest plant.
The team consisted of the plant manager and the managers of the five functional departments reporting to
him: engineering (maintenance), administration, human resources, production, and quality control. In
recruiting managers for the new plant, the company selected people with good technical skills and
experience in their respective functions. It also chose people with some managerial experience and a desire
to solve problems collaboratively, a hallmark of participative management. The team was relatively new,
and members had been working together for only about five months.
The team met formally for two hours each week to share pertinent information and to deal with plant wide
issues affecting all of the departments, such as safety procedures, interdepartmental relations, and personnel
practices. Members described these meetings as informative but often chaotic in terms of decision making.
The meetings typically started late as members straggled in at different times. The latecomers generally
offered excuses about more pressing problems occurring elsewhere in the plant. Once started, the meetings
were often interrupted by "urgent" phone messages for various members, including the plant manager, and
in most cases the recipient would leave the meeting hurriedly to respond to the call.
The group had problems arriving at clear decisions on particular issues. Discussions often rambled from
topic to topic, and members tended to postpone the resolution of problems to future meetings. This led to
a backlog of unresolved issues, and meetings often lasted far beyond the two-hour limit. When group
decisions were made, members often reported problems in their implementation. Members typically failed
to follow thorough on agreements, and there was often confusion about what had actually been agreed
upon. Everyone expressed dissatisfaction with the team meetings and their results.
Relationships among team members were cordial yet somewhat strained, especially when the team was
dealing with complex issues in which members had varying opinions and interests. Although the plant
manager publicly stated that he wanted to hear all sides of the issues, he often interrupted the discussion or
attempted to change the topic when members openly disagreed in their views of the problem. This
interruption was typically followed by an awkward silence in the group. In many instances when a solution
to a pressing problem did not appear forthcoming, members either moved on to another issue or they
informally voted on proposed options, letting majority rule decide the outcome. Members rarely discussed
the need to move on or vote; rather, these behaviors emerged informally over time and became acceptable
ways of dealing with difficult issues.
Analysis:
Application 3 presents an example of applying group-level diagnosis to a top-management team engaged in
problem solving.
The group is having a series of ineffective problem-solving meetings. Members report a backlog of
unresolved issues, poor use of meeting time, lack of follow through and decision implementation, and a
general dissatisfaction with the team meeting. Examining group inputs and design components and how
the two fit can help explain the causes of those group problems.
The key issue in diagnosing group inputs is the design of the larger organization within which the group is
embedded. The Ortiv Glass Corporation's design is relatively differentiated. Each plant is allowed to set up
its own organization design. Similarly, although no specific data are given, the company's technology,
structure, measurement systems, human resources systems, and culture appear to promote flexible and
innovative behaviors at the plant level. Indeed, freedom to innovate in the manufacturing plants is probably
an outgrowth of the firm's OD activities and participative culture.
In the case of decision-making groups such as this one, organization design also affects the nature of the
issues that are worked on. The team meetings appear to be devoted to problems affecting all of the
functional departments. This suggests that the problems entail high interdependence among the functions;
consequently high coordination among members is needed to resolve them. The team meetings also seem
to include many issues that are complex and not easily solved, so there is probably a relatively high amount
of uncertainty in the technology or work process. The causes of the problems or acceptable solutions are
not readily available. Members must process considerable information during problem solving, especially
when there are different perceptions and opinions about the issues.
img
Organization Development ­ MGMT 628
VU
Diagnosis of the team's design components answers the following questions:
1.
How clear are the group's goals? The team's goals seem relatively clear: they are to
solve
problems. There appears to be no clear agreement, however on the specific problems to be addressed. As a
result, members come late because they have "more
pressing" problems needing attention.
2.
What is the group's task structure? The team's task structure includes face-to-face
interaction during the weekly meetings. That structure allows members from different
functional departments to come together physically to share information and to solve
problems mutually affecting them. It facilitates coordination of problem solving among the
departments in the plant. The structure also seems to provide team members with the freedom necessary to
regulate their task behaviors in the meetings. They can adjust their behaviors and interactions to suit the
flow of the discussion and problem-solving process.
3.
What is the composition of the group? The team is composed of the plant manager and
managers of five functional departments. All members appear to have task-relevant skills and experience,
both in their respective functions and in their managerial roles. They also seem to be interested in solving
problems collaboratively. That shared interest suggests that members have job-related social needs and
should feel relatively comfortable in group problem-solving situations.
4.
What are the group's performance norms? Group norms cannot be observed directly but must
be inferred from group behaviors. The norms involve member beliefs about how  the  group  should
perform its task, including acceptable levels of performance. A useful way to describe norms is to list
specific behaviors that complete the sentences "A good group member should...." and "its okay to...."
Examination of the team's problem-solving behaviors suggests the following performance norms are
operating in the example:
·
"It's okay to come late to team meetings."
·
"It's okay to interrupt meetings with phone messages."
·
"It's okay to leave meetings to respond to phone messages."
·
"It's okay to hold meetings longer than two hours."
·
"A good group member should not openly disagree with others' views."
·
"It's okay to vote on decisions."
·
"A good group member should be cordial to other members."
·
"It's okay to postpone solutions to immediate problems."
·
"It's okay not to follow through on previous agreements."
5.
What is the nature of team functioning in the group? The case strongly suggests that
interpersonal relations are not healthy on the management team. Members do not seem to
confront differences openly. Indeed, the plant manager purposely intervenes when conflicts
emerge. Members feel dissatisfied with the meetings, but they spend little time talking about those
feelings. Relationships are strained, but members fail to examine the underlying causes.
The problems facing the team can now be explained by assessing how well the group design fits the inputs.
The larger organization design of Ortiv is relatively differentiated and promotes flexibility and innovation in
its manufacturing plants. The firm supports participative management, and the team meetings can be seen
as an attempt to implement that approach at the new plant. Although it is too early to tell whether the team
will succeed, there does not appear to be significant incongruity between the larger organization design and
what the team is trying to do. Of course, team problem solving may continue to be ineffective, and the
team might revert to a more autocratic approach to decision making. In such a case, a serious mismatch
between the plant management team and the larger company would exist, and conflict between the two
would likely result.
The team's issues are highly interdependent and often uncertain, and meetings are intended to resolve plant
wide problems affecting the various functional departments. Those problems are generally complex and
require the members to process a great deal of information and create innovative solutions. The team's task
structure and composition appear to be the nature of team issues. The face-to-face meetings help to
coordinate problem solving among the department managers, and except for the interpersonal skills,
members seem to have the necessary task-relevant skills and experience to drive the problem-solving
process. There appears, however, to be a conflict in the priority between the problems to be solved by the
team and the problems faced by individual managers.
More important, the key difficulty seems to be a mismatch between the team's performance norms and
interpersonal relations and the demands of the problem-solving task. Complex, interdependent problems
require performance norms that support sharing of diverse and often conflicting kinds of information. The
img
Organization Development ­ MGMT 628
VU
norms must encourage members to generate novel solutions and to assess the relevance of problem-solving
strategies in light of new issues. Members need to address explicitly how they are using their knowledge and
skills and how they are weighing and combining members' individual contributions.
In our example, the team's performance norms fail to support complex problem solving; rather, they
promote a problem-solving method that is often superficial, haphazard, and subject to external disruptions.
Members' interpersonal relationships reinforce adherence to the ineffective norms. Members do not
confront personal differences or dissatisfactions with the group process. They fail to examine the very
norms contributing to their problems. In this case, diagnosis suggests the need for group interventions
aimed at improving performance norms and developing healthy interpersonal relations.
Table of Contents:
  1. The Challenge for Organizations:The Growth and Relevance of OD
  2. OD: A Unique Change Strategy:OD consultants utilize a behavioral science base
  3. What an “ideal” effective, healthy organization would look like?:
  4. The Evolution of OD:Laboratory Training, Likert Scale, Scoring and analysis,
  5. The Evolution of OD:Participative Management, Quality of Work Life, Strategic Change
  6. The Organization Culture:Adjustment to Cultural Norms, Psychological Contracts
  7. The Nature of Planned Change:Lewin’s Change Model, Case Example: British Airways
  8. Action Research Model:Termination of the OD Effort, Phases not Steps
  9. General Model of Planned Change:Entering and Contracting, Magnitude of Change
  10. The Organization Development Practitioner:External and Internal Practitioners
  11. Creating a Climate for Change:The Stabilizer Style, The Analyzer Style
  12. OD Practitioner Skills and Activities:Consultant’s Abilities, Marginality
  13. Professional Values:Professional Ethics, Ethical Dilemmas, Technical Ineptness
  14. Entering and Contracting:Clarifying the Organizational Issue, Selecting an OD Practitioner
  15. Diagnosing Organizations:The Process, The Performance Gap, The Interview Data
  16. Organization as Open Systems:Equifinality, Diagnosing Organizational Systems
  17. Diagnosing Organizations:Outputs, Alignment, Analysis
  18. Diagnosing Groups and Jobs:Design Components, Outputs
  19. Diagnosing Groups and Jobs:Design Components, Fits
  20. Collecting and Analyzing Diagnostic information:Methods for Collecting Data, Observations
  21. Collecting and Analyzing Diagnostic information:Sampling, The Analysis of Data
  22. Designing Interventions:Readiness for Change, Techno-structural Interventions
  23. Leading and Managing Change:Motivating Change, The Life Cycle of Resistance to Change
  24. Leading and managing change:Describing the Core Ideology, Commitment Planning
  25. Evaluating and Institutionalizing Organization Development Interventions:Measurement
  26. Evaluating and Institutionalizing Organization Development Interventions:Research Design
  27. Evaluating and Institutionalizing Organization Development Interventions
  28. Interpersonal and Group Process Approaches:Group Process
  29. Interpersonal and Group Process Approaches:Leadership and Authority, Group Interventions
  30. Interpersonal and Group Process Approaches:Third-Party Interventions
  31. Interpersonal and Group Process Approaches:Team Building, Team Building Process
  32. Interpersonal and Group Process Approaches:Team Management Styles
  33. Organization Process Approaches:Application Stages, Microcosm Groups
  34. Restructuring Organizations:Structural Design, Process-Based Structures
  35. Restructuring Organizations:Downsizing, Application Stages, Reengineering
  36. Employee Involvement:Parallel Structures, Multiple-level committees
  37. Employee Involvement:Quality Circles, Total Quality Management
  38. Work Design:The Engineering Approach, Individual Differences, Vertical Loading
  39. Performance Management:Goal Setting, Management by Objectives, Criticism of MBO
  40. Developing and Assisting Members:Career Stages, Career Planning, Job Pathing
  41. Developing and Assisting Members:Culture and Values, Employee Assistance Programs
  42. Organization and Environment Relationships:Environmental Dimensions, Administrative Responses
  43. Organization Transformation:Sharing the Vision, Three kinds of Interventions
  44. The Behavioral Approach:The Deep Assumptions Approach
  45. Seven Practices of Successful Organizations:Training, Sharing Information