π Lecture 21 β Collecting and Analyzing Diagnostic Information
π Overview: This lecture covers the critical processes of collecting, analyzing, and feeding back diagnostic data in Organization Development. It compares various data collection methods, explains sampling techniques, and details both qualitative and quantitative analysis tools. Understanding these methods is essential for OD practitioners to accurately diagnose organizational problems and facilitate effective change.
ποΈ Topics Covered
This lecture begins by comparing different data collection methodsβquestionnaires, interviews, observations, and unobtrusive measuresβhighlighting their advantages and disadvantages. It then addresses sampling issues, including sample size and selection methods like random and stratified sampling. The lecture details the implementation of data collection and explores techniques for analyzing data, including qualitative tools like content analysis and force-field analysis, and quantitative tools like means, standard deviations, scattergrams, correlation coefficients, and difference tests. Finally, it discusses the critical process of feeding back diagnostic information, covering both the content and process characteristics necessary for client ownership and action.
π Lecture Summary
A Comparison of Different Methods of Data Collection
The lecture begins by comparing four major data collection methods. Questionnaires allow for quantification and easy summarization with large samples at low cost, but suffer from non-empathy, predetermined questions, over-interpretation, and response bias. Interviews are adaptive, provide rich data, and build rapport, but are expensive, prone to interviewer bias, and have coding difficulties. Observations collect data on actual behavior in real-time, but face coding issues, sampling inconsistencies, observer bias, and high expense. Unobtrusive measures are non-reactive with high face validity, but have access, validity, and coding difficulties.
Sampling
Sampling addresses questions like "How many people should be interviewed?" and "Which records should be inspected?" While sampling is not needed when data is collected from all members, it becomes crucial for large systems. Sample size depends on population size, desired confidence in data quality, and available resources. Larger, more complex populations require larger samples. Greater confidence demands a larger proportion of the population, while limited resources constrain the sample size, favoring questionnaires over interviews if high confidence is needed. Sample selection commonly uses a simple random sample, where each member has an equal chance of selection, generated via a random number table or by picking every nth name. A stratified sample is used for complex populations, where the population is divided into subpopulations (e.g., managers, workers), and a random sample is taken from each.
π Definition β Simple random sample: A sampling method where each member, behavior, or record has an equal chance of being selected. π Definition β Stratified sample: A sampling method where the population is segregated into mutually exclusive subpopulations, and a random sample is taken from each.
The Implementation of Data Collection
Data collection begins with deciding who and how many to survey. The one-to-one interview is the most common method (87% of respondents), followed by observation (60%), group interviews (52%), questionnaires (45%), and existing documents (37%). Practitioners typically rely on a variety of methods. Outside data-collection agents are recommended as they make respondents feel more secure about confidentiality. A small pilot study or beta test is crucial to correct problems before large-scale collection.
The Analysis of Data
Data analysis techniques range from simple to highly sophisticated statistical methods. Key questions to consider before collection include how data will be analyzed, by hand or computer, and how they will be coded. This is especially critical for large-scale surveys where processing is difficult. Analysis may include comparisons of different divisions or management levels. π‘ Why this matters: Properly planning for analysis before data collection is essential; if data isn't collected correctly, it cannot be meaningfully analyzed or controlled.
Evaluating the Effectiveness of Data Collection
Effectiveness is judged by criteria such as data validity (measuring what was intended), time to collect (which usually takes longer than planned), cost (including diminishing returns from too many interviews), organization culture and norms (choosing techniques suited to the climate), and the Hawthorne effect (the observer's influence on the subject's behavior).
Techniques for Analyzing Data
Data analysis falls into two classes: qualitative (easier to use and interpret) and quantitative (provides more accurate readings).
Qualitative Tools
Content analysis summarizes qualitative data (e.g., interview comments) into meaningful categories. The process involves: 1) reading responses to gain familiarity, 2) generating themes that capture recurring comments, and 3) placing each response into one of the categories. The categories with the most responses represent the most often mentioned themes.
Force-field analysis, derived from Kurt Lewin's model, organizes information into forces for change (driving forces) and forces for maintaining the status quo (restraining forces). The first step is to list all forces and then determine which are most powerful, either by ranking or rating their strength. According to Lewin, efforts to change should focus on reducing the restraining forces, which is likely to result in change with less tension than increasing the driving forces.
π Definition β Force-field analysis: A method for analyzing qualitative data that organizes information into forces promoting change and forces resisting change. π Formula: Driving Forces β [Current State] β Restraining Forces β [Desired State] π Example: A hospital's work team identified a 6% daily absentee rate as a problem and a 3% rate as desirable. Using force-field analysis, they listed restraining forces (e.g., lack of consequences, low morale) and driving forces (e.g., supervisor pressure, job insecurity) and chose to reduce restraining forces by, for example, tying promotion policy more closely to absenteeism.
Quantitative Tools
The most common quantitative tools are means, standard deviations, frequency distributions, scattergrams, correlation coefficients, and difference tests.
A mean and standard deviation summarize the average score and the spread or variability of responses. A high standard deviation indicates considerable disagreement among respondents. A frequency distribution is a graphical method showing the number of times a particular response was given, which provides a clearer picture than the mean alone.
A scattergram visually displays the relationship between two variables by plotting each case at the intersection of its values. A correlation coefficient is a number (ranging from +1.0 to -1.0) that summarizes the data in a scattergram. +1.0 means a perfect positive relationship, -1.0 means a perfect negative relationship, and 0 means no relationship.
A difference test compares a sample group against a standard (e.g., industry average) or determines if two samples are significantly different from each other. It can also be used to determine if a group has changed its score over time by calculating a difference score (t-score or z-score).
π Definition β Correlation coefficient: A number that summarizes the data in a scattergram, indicating the strength and direction of a relationship between two variables, ranging from +1.0 to -1.0. π Formula: Correlation Coefficient (r) = value between -1.0 and +1.0 β A value of +0.8 indicates a strong positive relationship. π Example: Data on the number of change orders and conflicts per month over a year was plotted on a scattergram. The pattern showed an apparently strong positive relationship, suggesting change orders contribute to conflict between departments.
Feeding Back Diagnostic Information
This is perhaps the most critical step in the diagnostic process. The key objective is for the client to have ownership of the data. Data feedback must arouse organizational action and direct energy toward problem solving.
Determining the Content of the Feedback: Effective feedback data has nine properties: relevant, understandable, descriptive, verifiable, timely, limited, significant (focusing on what members can change), comparative (providing benchmarks), and un-finalized (acting as a stimulus for further diagnosis).
Characteristics of the Feedback Process: The feedback meeting is a forum for discussing data and devising action plans. To overcome anxiety and ensure ownership (the willingness to take responsibility for the data), the process must include: 1) motivation to work with the data, 2) structure for the meeting, 3) appropriate attendance (including those with common problems), 4) appropriate power (clarifying what the group can change), and 5) process help from the OD practitioner to keep the group focused.
π Definition β Survey feedback: A flexible and powerful technique for data feedback that arose from the wide use of questionnaires in OD work.
β Key Takeaways
This lecture is a cornerstone for any OD practitioner. You must understand the trade-offs between different data collection methods and when to use each. For sampling, remember that random samples are best for simple populations, while stratified samples are needed for complex ones. For analysis, be able to distinguish between qualitative tools (content analysis for summarizing themes, force-field analysis for identifying change drivers and resistors) and quantitative tools (mean/SD/frequency for single variables, scattergrams/correlations for relationships, and t-tests for group differences). The most critical concept is that feedback is not just about presenting data; the process must be managed to foster client ownership, meaning the data must be relevant, understandable, and fed back in a structured meeting where attendees have the power and motivation to act. Without ownership, even perfect data will not lead to change.
π§ Quick Revision Questions
- What is the difference between a simple random sample and a stratified sample, and in what kind of diagnostic situation would you choose each?
- Explain the main difference between qualitative and quantitative data analysis techniques. Give one example of each.
- According to force-field analysis by Kurt Lewin, which is the more effective change strategy: increasing driving forces or reducing restraining forces? Why?
- What does a high standard deviation indicate about a set of data, and why is a frequency distribution sometimes more informative than just the mean?
- List three of the nine properties of effective feedback content and three of the five features of a successful feedback process.
π Lecture 22 β Designing Interventions
π Overview: This lecture defines what constitutes an effective organization development (OD) intervention and explains the criteria for designing them. It explores the key contingencies that affect intervention success, including readiness for change, cultural context, and the change agentβs capabilities. The lecture also provides a comprehensive overview of the major types of OD interventionsβhuman process, techno-structural, human resources management, and strategicβand how they target different organizational issues and levels.
ποΈ Topics Covered
The lecture begins by defining effective interventions and their three main criteria: fit with organizational needs, causal knowledge of outcomes, and transfer of change-management competence. It then covers how to design interventions by considering contingencies related to the change situation (readiness, capability, culture, change agent skills) and the target of change (organizational issues and levels). A detailed overview of four major intervention types follows: human process, techno-structural, human resources management, and strategic interventions, with specific methods listed under each category.
π Lecture Summary
What are effective interventions?
An intervention is a set of sequenced, planned actions intended to help an organization increase effectiveness. Interventions purposely disrupt the status quo. In OD, three major criteria define an effective intervention: (1) it fits the needs of the organization; (2) it is based on causal knowledge of intended outcomes; and (3) it transfers change-management competence to organization members.
The first criterion requires that interventions be relevant, based on valid information about the organization, provide opportunities for free and informed choice, and gain members' internal commitment. Valid information reflects what members perceive and feel. Free and informed choice means members actively participate in decisions affecting them and can choose not to participate. Internal commitment means members accept ownership and responsibility for implementing the intervention.
The second criterion involves knowledge of outcomes. Interventions must be based on valid knowledge that specific results can be produced. Currently, knowledge of intervention effects is in a rudimentary stage in OD, with much evaluation research lacking sufficient rigor. Few attempts have been made to compare the effectiveness of different OD techniques.
The third criterion concerns enhancing the organization's capacity to manage change. Members should gain knowledge and skill in managing change from active participation. Competence in change management is essential in today's rapidly changing environment.
How to design effective interventions:
Designing OD interventions requires careful attention to the needs and dynamics of the change situation. Two major sets of contingencies affect intervention success: those related to the change situation and those related to the target of change.
Contingencies Related to the Change Situation:
Several situational factors affect intervention success. These include individual differences among members (e.g., needs for autonomy), organizational factors (e.g., management style, technical uncertainty), and dimensions of the change process (e.g., degree of top-management support). Four generic contingencies apply to all interventions:
Readiness for Change: Intervention success depends on the organization being ready for planned change. Indicators of readiness for change include sensitivity to pressures for change, dissatisfaction with the status quo, availability of resources, and commitment of significant management time. When readiness is low, interventions must first focus on increasing willingness to change.
Capability to Change: Managing planned change requires specific knowledge and skills, including the ability to motivate change, lead change, develop political support, manage the transition, and sustain momentum. If members lack these capabilities, a preliminary training intervention may be needed.
Cultural Context: The national culture within which the organization operates can exert a powerful influence on members' reactions to change. Interventions must be modified to fit local cultural values and assumptions, particularly when OD practices developed in one culture are applied in another.
Capabilities of the Change Agent: Many OD failures result from change agents applying interventions beyond their competence. Practitioners should assess their experience and expertise against requirements. The ethical guidelines require full disclosure and practitioners should intervene within their capabilities or recommend someone more suited.
Contingencies Related to the Target of Change:
Two key contingencies related to change targets can affect intervention success: the organizational issues the intervention is intended to resolve and the level of organizational system at which the intervention is expected to have a primary impact.
Organizational Issues:
Organizations must address four interrelated issues to operate effectively, which are key targets of OD interventions:
- Strategic issues β decisions about products, services, markets, and how to relate to environments. OD methods are called strategic interventions (e.g., integrated strategic change, mergers and acquisitions, trans-organizational development, organization learning).
- Technology and structure issues β how to divide work, coordinate departments, deliver products, and link people to tasks. Methods are called techno-structural interventions (e.g., organization design, employee involvement, work design).
- Human resources issues β attracting competent people, goal setting, performance appraisal, rewards, career development, and stress management. Methods are human resources management interventions.
- Human process issues β social processes like communication, decision making, leadership, and group dynamics. Methods are human process interventions (e.g., conflict resolution, team building).
π‘ Why this matters: These organizational issues are interrelated. Decisions about one issue must fit with choices about others. For example, a goal-setting intervention may need to be integrated with a reward system linking pay to goal achievement.
Organizational Levels:
Organizations function at different levelsβindividual, group, organization, and trans-organization. Interventions primarily affect one level but also have secondary impacts on others. For example, structural design mainly affects the organization level but indirectly affects groups and individuals. Practitioners must think systemically, addressing the possibility of cross-level effects.
Overview of interventions:
Human Process Interventions:
These interventions focus on people and the processes through which they accomplish goals, including communication, problem solving, group decision making, and leadership. They derive from psychology, social psychology, group dynamics, and human relations.
Interpersonal and group process interventions:
- T-group β provides experiential learning about group dynamics, leadership, and interpersonal relations. Brings ten to fifteen strangers together with a professional trainer.
- Process consultation β focuses on interpersonal relations and social dynamics in work groups. A consultant helps members diagnose group functioning and devise solutions.
- Third-party intervention β a form of process consultation aimed at dysfunctional interpersonal relations. Helps people resolve conflicts through problem solving, bargaining, and conciliation.
- Team building β helps work groups become more effective, including examination of group tasks, member roles, and strategies.
System-wide human process interventions:
- Organization confrontation meeting β mobilizes members to identify problems, set action targets, and begin working on problems. Applied when organizations are experiencing stress.
- Inter-group relations β improves interactions among groups. Includes microcosm group intervention and inter-group conflict model.
- Large-group interventions β involve a broad variety of stakeholders in large meetings to clarify values, develop new ways of working, or solve problems.
- Grid organization development β a normative, packaged program with standardized instruments and specific procedures.
Techno-structural Interventions:
These interventions focus on technology (task methods, job design) and structure (division of labor, hierarchy). They derive from engineering, sociology, psychology, and socio-technical systems.
Restructuring interventions:
- Structural design β concerns the organization's division of labor, moving from traditional structures (functional, self-contained-unit, matrix) to integrative forms (process-based, network-based).
- Downsizing β reduces costs and bureaucracy through personnel layoffs, organization redesign, and outsourcing.
- Reengineering β radically redesigns core work processes to create tighter linkage and coordination. Often uses new information technology. Often fails if it ignores basic OD principles.
Employee involvement (EI) β aims to improve employee well-being and effectiveness by moving knowledge, power, information, and rewards downward. Includes parallel structures, high-involvement plants, and total quality management.
Work design β concerns designing work for groups and individual jobs. Includes engineering, motivational, and socio-technical systems approaches producing enriched jobs and self-managing teams.
Human Resources Management Interventions:
These interventions focus on personnel practices used to integrate people into organizations, including career planning, reward systems, goal setting, and performance appraisal. They derive from economics, labor relations, and personnel practices.
Performance management interventions:
- Goal setting β sets clear and challenging goals to improve fit between personal and organizational objectives.
- Performance appraisal β a systematic process of jointly assessing work-related achievements, strengths, and weaknesses.
- Reward systems β design of organizational rewards to improve satisfaction and performance.
Development and assistance interventions:
- Career planning and development β helps people choose organizations, career paths, and attain career objectives.
- Managing workforce diversity β makes HR practices responsive to diverse individual needs (women, ethnic minorities, disabled).
- Employee wellness β includes employee assistance programs (EAPs) for substance abuse, mental health issues, and stress management programs.
Strategic Interventions:
These link internal functioning to the larger environment and transform organizations to keep pace with change. They derive from strategic management, organization theory, open-systems theory, and cultural anthropology.
Managing organization-environment relationships:
- Integrated strategic change β describes how planned change contributes to strategic management. A strategic change plan helps manage transitions.
- Trans-organization development β helps organizations enter alliances, partnerships, and joint ventures for complex tasks.
- Merger and acquisition integration β assists organizations to form a new entity, addressing strategic, leadership, and cultural issues.
Transforming organizations:
- Culture change β helps organizations develop cultures appropriate to their strategies and environments.
- Self-designing organizations β gains capacity to alter fundamentally through participative processes involving multiple stakeholders.
- Organization learning and knowledge management β enhances capability to acquire new knowledge (OL) and use it to improve performance (KM), enabling continuous improvement.
β Key Takeaways
This lecture establishes that effective OD interventions must fit organizational needs, be based on causal knowledge of outcomes, and transfer change-management competence to members. Success depends heavily on situational contingencies including readiness for change, change capability, cultural context, and change agent skills. The four major intervention typesβhuman process, techno-structural, human resources management, and strategicβtarget different but interrelated organizational issues and levels. Crucially, because organizational issues and levels are connected, interventions often need to be integrated with one another to achieve systemic impact. Students must remember that interventions must be designed carefully based on diagnosis, not applied generically, and that change agents must work within their capabilities.
π§ Quick Revision Questions
- What are the three major criteria for an effective OD intervention?
- List four generic situational contingencies that can affect intervention success.
- What are the four interrelated organizational issues that are key targets of OD interventions?
- Name one human process intervention for interpersonal/group dynamics and one for system-wide change.
- Why must OD interventions be integrated with one another rather than applied in isolation?