Group Development: Tuckman's Stages and Garland-Jones-Kolodny Model
Overview
Groups are not static entities — they evolve over time. Just as individuals go through developmental stages, groups have their own developmental trajectory. Understanding how groups develop helps leaders and members navigate the challenges of each stage, foster healthy dynamics, and achieve goals more effectively. This section covers the two most widely cited models of group development in social psychology: Tuckman's (1963) stage model and Garland, Jones and Kolodny's (1976) five-stage model.
📄 Block-4/Unit-2.pdf — Pages 30–31 📚 MPC-004 Advanced Social Psychology
1. Why Do Groups Develop Through Stages?
Groups are not ready-made entities — they are constructed through interaction. As members come together, they must:
- Learn about each other's personalities, skills and values
- Establish norms and roles through trial and error
- Build trust and cohesion gradually
- Negotiate status hierarchies
- Create a group culture through sustained interaction
This process takes time and follows recognisable patterns — which is why stage models of group development have become foundational in social psychology.
"As groups develop over time, group dynamic processes evolve." — IGNOU MPC-004 PDF
General Pattern of Group Development
Group development typically follows this arc:
Beginning stages are characterised by:
- Tentative and cautious interaction
- Establishment of norms, roles and status hierarchies
- Slow emergence of group culture
- Little conflict initially
Middle stages are characterised by:
- Members becoming more comfortable
- Resistance can develop as members assert themselves
- Members want to belong but also maintain their own identity
2. Tuckman's (1963) Four-Stage Model: Forming-Storming-Norming-Performing
Bruce Tuckman (1963) proposed the most widely used model of small group development. His model identifies four sequential stages through which every group passes.
The Four Stages
| Stage | Name | Characteristics |
|---|---|---|
| Stage 1 | Forming | Orientation, getting to know each other, dependence on the leader |
| Stage 2 | Storming | Conflict, resistance, power struggles, testing boundaries |
| Stage 3 | Norming | Cohesion develops, norms established, cooperation increases |
| Stage 4 | Performing | Functional interdependence, task-focused, high productivity |
Detailed Stage Analysis
Stage 1: Forming
- Members are polite and cautious — no one wants to rock the boat
- High dependence on the leader for direction
- Members are exploring: What is this group about? Who are these people? What are the rules?
- Anxiety and uncertainty are common
- Focus is on inclusion and belonging
Stage 2: Storming
- Conflict emerges as members assert themselves
- Power struggles over roles, leadership and direction
- Resistance to the task or to the leader
- Sub-group formation may occur
- This stage is uncomfortable but necessary — groups that avoid it often lack genuine cohesion
Stage 3: Norming
- Resolution of conflict from the storming stage
- Cohesion develops — members feel they belong
- Norms and roles become established and accepted
- Cooperation increases as trust builds
- Members begin to support each other
Stage 4: Performing
- The group is functioning effectively as a unit
- Focus shifts fully to task accomplishment
- Members are flexible and interdependent
- High productivity and creative problem-solving
- Members feel competent and confident
Tuckman's 1977 Addition: Adjourning
In 1977, Tuckman and Jensen added a fifth stage:
Stage 5: Adjourning (Mourning)
- The group concludes its work and disbands
- Members experience a sense of loss and may grieve the end of the group
- Particularly important in therapeutic groups and project teams
- The leader's role is to facilitate healthy closure and transition
The original 1963 model has four stages. The 1977 addition of "Adjourning" is often included in more recent textbooks. For IGNOU exams, focus on the original four: Forming, Storming, Norming, Performing (mnemonic: FSNP or "Freddie Sings New Pop").
3. Garland, Jones and Kolodny's (1976) Five-Stage Model
Garland, Jones and Kolodny (1976) developed a five-stage model specifically for social work and therapeutic groups. It emphasises the emotional and relational journey of group members.
The Five Stages
| Stage | Name | Key Focus |
|---|---|---|
| Stage 1 | Pre-Affiliation | Should be affected to the group; approach/avoidance conflict |
| Stage 2 | Power and Control | Should be able to have some influence over other members |
| Stage 3 | Intimacy | There must be certain closeness and intimacy among members |
| Stage 4 | Differentiation | Members differentiate their personal goals from group goals |
| Stage 5 | Separation | Each member has a separate identity despite being part of the group |
Detailed Stage Analysis
Stage 1: Pre-Affiliation
- Members experience an approach-avoidance conflict — they want the group's benefits but fear vulnerability
- Behaviour is tentative and exploratory
- Members test whether the group is safe and worth joining
- The leader's role is to reduce anxiety and make membership attractive
Stage 2: Power and Control
- Members begin asserting their need for influence and autonomy
- Testing of the leader's authority and group norms
- Members determine their place in the hierarchy
- Conflict may emerge over decision-making and leadership
- Similar to Tuckman's Storming stage but with stronger emphasis on power
Stage 3: Intimacy
- Members develop genuine closeness beyond surface politeness
- Self-disclosure increases — members share more personal information
- Trust and mutual understanding deepen
- The group begins to feel like a community
- Members are invested in each other's growth and wellbeing
Stage 4: Differentiation
- A more mature stage where members can distinguish between:
- Their personal goals (what they want as individuals)
- Group goals (what the group is collectively trying to achieve)
- Members appreciate the group while maintaining individual identity
- The group can handle disagreement without fragmenting
- Members can challenge the leader and each other constructively
Stage 5: Separation
- Preparation for the end of the group
- Members develop a separate identity — they carry the group's lessons forward as individuals
- They recognise they are part of the group AND distinct from it
- Healthy termination allows members to apply their growth in new contexts
- Contrast with unhealthy termination: abandonment, avoidance or premature exit
4. Comparing the Two Models
| Dimension | Tuckman (1963) | Garland, Jones & Kolodny (1976) |
|---|---|---|
| Number of stages | 4 (+ Adjourning in 1977) | 5 |
| Primary focus | Task group development | Therapeutic/social work groups |
| Emphasis | Group task productivity | Emotional and relational journey |
| Stage terminology | Forming, Storming, Norming, Performing | Pre-Affiliation, Power/Control, Intimacy, Differentiation, Separation |
| Best applied to | Work teams, project groups, committees | Therapy groups, support groups, social work |
| Handling conflict | Storming stage = necessary conflict | Power & Control stage = authority testing |
| Ending | Adjourning | Separation |
Which Model to Use?
- Tuckman: Best for work teams, educational groups, project teams where task achievement is primary
- Garland et al.: Best for therapeutic groups, support groups, community development groups where emotional process is central
Both models agree that:
- Group development is sequential and progressive
- Conflict is normal and necessary (Storming/Power & Control)
- Cohesion builds over time (Norming/Intimacy)
- Healthy termination requires deliberate attention (Adjourning/Separation)
5. Structural Characteristics and Group Development
Many factors affect group development, but structural characteristics have a particularly important impact:
Key Structural Factors
| Structural Factor | Impact on Group Development |
|---|---|
| Group size | Smaller groups develop more quickly; larger groups have slower cohesion |
| Group composition | Homogeneous groups develop faster; heterogeneous groups develop more slowly but may produce better outcomes |
| Group purpose | Closed vs. open groups develop differently; therapeutic vs. task groups have different trajectories |
| Leadership style | Directive leaders can speed early stages; facilitative leaders support later stages |
| Duration | Short-term groups may not progress beyond early stages; long-term groups can reach full development |
| Meeting frequency | More frequent meetings accelerate development |
Group Structure and Social Integration
As groups develop, structure and social integration become increasingly sophisticated:
- Members develop positions that stand in relationship to one another
- Reciprocal role patterns become established
- Expectations and attitudes become predictable
- This structure gives groups stability and predictability
6. Group Development in the Indian Context
Indian Examples of Group Development Stages
Self-Help Groups (SHGs) in rural India provide excellent illustrations:
| Stage | SHG Example |
|---|---|
| Pre-Affiliation / Forming | Women initially uncertain whether to join; NGO worker builds trust and allays fears |
| Power & Control / Storming | Debates over who becomes treasurer or group leader; testing of facilitator's rules |
| Intimacy / Norming | Women begin sharing personal financial problems; mutual support develops |
| Differentiation / Performing | Individual members develop separate income-generating projects while supporting the group |
| Separation / Adjourning | Women graduate from micro-finance group; many form autonomous cooperatives |
Cultural Factors in Indian Group Development
- Collectivist orientation: Indian groups may skip or minimize the Storming/Power & Control phase due to cultural pressure to maintain harmony — but this can lead to suppressed conflict that emerges later
- Authority deference: Hierarchical respect for elders/leaders may make the Power & Control stage less visible but no less present
- In-group preference: Indian groups often develop strong in-group cohesion quickly but struggle with integrating out-group members