Observation in Research: Participant, Structured, Overt and Covert

What is observation as a research method?

Observation is the systematic watching, listening to and recording of behavior, interactions and settings. It differs from everyday watching in three ways: it follows a research question, it is planned, and it is documented in a way others can follow.

Observation is valuable whenever there is a gap between what people say and what they do. Teachers may report giving all students equal attention; an observation of who gets called on may show something different. It is also the method of choice when participants cannot easily report on their behavior, for example young children or routine actions people are not aware of.

Types of observation

Observation designs vary along several dimensions. Describe your choice on each in your methods chapter.

Observer roles (Gold, 1958)

A classic typology by Raymond Gold distinguishes four roles along a continuum from full participation to pure observation. Most student projects use one of the middle roles.

Structured observation: building a coding scheme

In structured observation you define in advance what you will record, how and when. A coding scheme lists categories that are mutually exclusive and clearly defined, with examples. Sampling can be by time (e.g. record what happens every 30 seconds) or by event (record every instance of a behavior).

• Define each category with an inclusion rule and an example.
• Train observers and pilot the scheme in one or two sessions.
• Check agreement between observers on a subset.
• Report frequencies, rates per hour and agreement statistics.

Unstructured observation and field notes

In qualitative and ethnographic observation, you record what you see and experience in detailed field notes. Separate description from interpretation: write what happened, then your reflections in a separate column or marked section.

• Jotted notes during or right after the observation: key words, quotes, times.
• Full field notes the same day: detailed description of setting, people, actions, conversations.
• Analytic memos: first interpretations, questions, ideas for the next session.
• Reflexive notes: your own feelings, role and possible influence on the situation.

Validity: the observer effect and other problems

People often behave differently when they know they are being watched, an effect sometimes called reactivity. Reduce it by spending time in the field before recording data, being unobtrusive and observing over several sessions. Other threats include selective perception (seeing what you expect), going native in long participant observation (losing critical distance) and fatigue in long sessions. Combine observation with interviews or documents to triangulate.

Ethics: overt vs covert observation

Overt observation requires informed consent from those observed, or at least from gatekeepers and those directly involved, depending on the setting. Covert observation deceives participants and is only acceptable in exceptional cases, for example when open observation would make the research impossible and the topic has high public interest, and only with explicit ethics approval. Observing in public spaces still raises questions about privacy and recording. Never record audio or video without consent. Your university’s research ethics committee, or IRB in the US, will decide; their guidelines and your department’s rules take precedence.

Analysis and tools

Structured observation data are analyzed statistically (frequencies, comparisons, agreement statistics). Field notes are analyzed qualitatively, for example with thematic analysis. At /codieren you can code field notes like interview transcripts and keep your coding scheme with definitions and examples. The methods tool helps you decide whether observation fits your question. Upload your university’s guidelines to Professor Digital and the advice is tailored to them.