7 Studying Groups and Organizations
Learn how to observe group interaction, map relationships, study formal and informal organizational practices, and interpret findings responsibly.
Studying collective life
Studying a group or organization involves examining both what people do together and how their relationships and setting shape what happens. Researchers use observation, interviews, documents, and network data to describe interaction patterns, identify formal and informal roles, and understand how work or decisions move through a system. No single method captures every part of group life, so comparing different kinds of evidence can strengthen findings.
Observing group interaction
involves taking part in a group’s activities while observing them. focuses on watching without joining. Both approaches can reveal routines, spoken and unspoken norms, status differences, and how people respond to particular situations. Observation may be open-ended, guided by a research question, or structured around a coding scheme.
A practical observation plan specifies:
The question: For example, “How are concerns raised during team meetings?”
The setting and sample: Which groups, activities, people, and time periods will be observed? Include ordinary routines and, where relevant, less common events.
The unit of observation: This might be a speaking turn, decision, interruption, task handoff, or meeting.
What to record: Describe actions and context before interpreting their meaning. Note who acted, what happened, when it happened, and how others responded.
Field notes may include brief jottings made during an event, fuller descriptions written soon afterward, and separate analytic memos that record questions or emerging interpretations. Keeping observation separate from interpretation helps prevent assumptions from being mistaken for evidence. Interviews and documents can help check what an observed action meant to participants and how it fits the wider setting.
Recording communication patterns
classifies communication into categories so researchers can compare patterns across members or meetings. A simpler interaction map can record who communicates with whom, the sequence of exchanges, and their approximate duration. These approaches can help show whether a meeting is dominated by task coordination, whether some members are rarely heard, or whether disagreement is followed by clarification.
A person’s behavior in one meeting should not, by itself, be treated as a fixed role or personality trait. Interaction patterns describe behavior in a particular setting; interpretation should consider the context and evidence beyond an isolated event.
Defining and mapping networks
In , people or organizations are represented as nodes, and their relationships as ties. A might represent advice-seeking, collaboration, communication, referrals, or resource exchange. Ties can be directed, as when A seeks advice from B, or undirected, as when A and B collaborate; they may also vary in strength or frequency.
A network study begins by defining its boundary: who or what counts as part of the network, and which relationships matter to the research question. Researchers may collect network information through surveys, observation, interviews, administrative records, or communication data. Each source has limits. For example, survey responses may omit relationships that people do not recall or do not feel comfortable reporting.
Interpreting network measures
Network measures answer different questions:
counts an actor’s direct ties. It can identify people with many reported connections, but does not by itself show influence or effectiveness.
measures how often an actor lies on shortest paths between other actors. High betweenness may indicate a bridge between otherwise separated parts of a network.
describes how many of the possible ties in a network are present. A dense network has many connections, but is not automatically a sign of better performance.
reveal connected subgroups and actors with few or no recorded ties.
Network diagrams make patterns easier to inspect, but they are not explanations on their own. A central person may be a useful coordinator, an overloaded bottleneck, or both. Researchers interpret measures in light of the type of , organizational context, and relevant evidence from observation or interviews.
Studying organizational practice
Organizations have , such as job titles, reporting lines, rules, and procedures, as well as , including trusted advice relationships, workarounds, and unofficial coordination. An organization chart describes authority but may not show how information or practical help actually flows. Organizational research therefore examines everyday work alongside documents, policies, meeting practices, and communication patterns.
Ethnographic observation can reveal gaps between written procedures and practice. For example, a researcher studying hospital handoffs might observe several shifts, record how information is passed, examine handoff forms, and interview staff about interruptions or missing details. Comparing these sources can help distinguish a poorly designed procedure from difficulties in how it is used. Network mapping might then show whether coordination depends heavily on one person or whether some teams have weak connections to others.
Using findings responsibly
Research findings can inform practical decisions in workplaces, schools, health services, and community partnerships. They may help with meeting facilitation, onboarding, referral pathways, cross-team communication, or workload distribution. Findings should be presented as evidence about patterns and conditions, not as simplistic rankings of individuals.
Ethics and research quality
Research involving groups and organizations can expose private relationships, sensitive conversations, or perceived performance problems. Researchers should seek appropriate consent and review, explain how information will be used, protect identities, limit access to identifiable data, and consider whether feedback could harm participants. Network diagrams require particular care because a person may be identifiable even when names are removed.
Quality improves when the research question, sampling decisions, and coding rules are explicit; observation records distinguish description from inference; and findings are checked against more than one data source. Researchers should also reflect on how their presence, organizational position, and assumptions may shape what they observe and how they interpret it.