Choosing Structured or Unstructured for a New Study
- This skill asks you to pick structured or unstructured observation for a study you have not seen before.
- The choice turns on one question: do you already know which behaviours you are looking for?
- It also depends on whether the study wants countable numbers or a rich description.
- Know the behaviours and want counts? Reach for structured observation.
- Still exploring and want detail? Reach for unstructured observation.
A Worked Example
- Imagine you are asked to study how Year 7 pupils behave during their first week in a busy secondary-school canteen.
- No one has mapped these behaviours yet, so you cannot list them in advance.
- Begin with an unstructured observation and note everything the pupils do.
- Those notes reveal common actions, such as queuing, saving seats and helping a friend carry a tray.
- Unstructured first: write a running account of what happens across several lunchtimes.
- Structured next: turn the actions you spotted into a tally chart and count them.
Justifying Your Choice in the Answer
- If the behaviours are already clear, say structured observation fits and explain why.
- If the study is opening up something new, say unstructured observation fits.
- Always state what kind of data each choice would hand you.
- Do not pick structured observation when you cannot yet name the behaviours to record.
- Do not pick unstructured observation when the study needs simple, comparable counts.
A Step-by-Step Routine
- Ask whether the scenario already names the behaviours to look for.
- Ask whether it wants countable numbers or a rich description.
- Name your approach and justify it from those two answers; the justification tied to the study scores separately from the name.
- Which approach suits a study exploring a behaviour for the first time?
- Which approach gives countable, comparable data?
- For counting how often pupils jump the canteen queue, which would you choose and why?
- How can an unstructured observation help you plan a structured one?