Description
Observing people in a real setting and recording predefined behaviours on a checklist, with sampled times and checks on observer reliability.
Structured observation records specific, predefined behaviours rather than everything that happens. Gratton and Jones describe four planning stages: define the variables to be observed and link them to the research objectives; decide the sample, including a sample of times and locations, not just of people; decide how the variables will be recorded (pre-printed data sheet, video or dictation); and pilot the observation before the real event. Recording usually uses a simple checklist where each action by a sampled individual is ticked against categories, as in their example of coding each basketball play as 'dribbled then shot', 'immediate pass', 'lost possession' and so on. Because it is hard to record more than one variable at a time, observers should keep the variable list short, alternate between variables in set periods, use video or use several observers, in which case inter-observer reliability must be trained and checked. Typical mistakes are trying to record too much, ignoring the observer's effect on behaviour, not sampling times or locations and relying on memory. In UX it suits counting behaviours in physical or shared settings (kiosks, retail, hospital wards, control rooms) or coding session videos, giving frequencies that complement interviews and analytics.
Attributes
- Discipline
- Field & Ethnographic
- Loss type
- Vestigial
- Stage
- Discovery, Evaluation
- Data
- Quantitative
- Effort
- Days
Steps
- Define the behaviours or events to record and write clear category definitions tied to your research questions.
- Choose who to observe and sample times and locations (for example peak and off-peak sessions on different days).
- Build a checklist or coding sheet, or a video coding scheme, keeping the number of variables manageable.
- Pilot the sheet in the real setting and refine categories that are ambiguous or never used.
- Train observers and, where more than one observer is used, double-code a sample to check inter-observer agreement.
- Collect the data, noting context and any signs that observers are affecting behaviour.
- Tally frequencies and compare across times, locations or groups, and triangulate with other data.
Sources
- Gratton, C. & Jones, I., Research Methods for Sports Studies (2010)
- Veal, A. J. & Darcy, S., Research Methods in Sport Studies and Sport Management (2014)
Also known as
- Systematic observation
- Observation checklist
- Non-participant structured observation
- Coded observation schedule
- Structured observational counts
- Observation recording sheet method
- Observational methodology (Anguera)
Categories
- User Research
- Field & Ethnographic