Observational design | AQA A-Level Psychology Revision

For 7182 specification, first teach in September 2025

AQA A-Level Psychology | Free Revision Notes

Estimated study time: 45 minutes

Observational design concerns how psychologists decide which behaviours to record and when to record them. This Observational design A-Level Psychology revision page explains how to construct precise behavioural categories and how event sampling and time sampling can be used to collect observational data. These choices matter because vague categories or unsuitable sampling procedures can produce incomplete or inconsistent findings. The AQA specification requires students to understand behavioural categories, event sampling and time sampling as part of scientific processes in research methods.

Learning Objectives 🎯

By the end of this revision page, you should be able to:

• Define a behavioural category.

• Construct clear, observable and operationalised behavioural categories.

• Explain how event sampling is used to record behaviour.

• Explain how time sampling is used to record behaviour.

• Compare the strengths and limitations of event sampling and time sampling.

• Select and justify an appropriate observational design for an unfamiliar research scenario.

Revision Notes 📚

What is observational design?

An observational design is the plan a researcher uses to record behaviour during an observation.

The researcher must decide:

1. Which behaviours will be recorded.

2. How each behaviour will be defined.

3. When behaviour will be recorded.

4. How the observations will be entered onto a record sheet.

These decisions are different from choosing the type of observation. For example, naturalistic, controlled, covert, overt, participant and non-participant observations concern the way the observation is conducted. You can revise these distinctions in choosing an appropriate form of observation.

The three elements required for this lesson are:

• Behavioural categories

• Event sampling

• Time sampling

Behavioural categories

A behavioural category is a clearly defined type of behaviour that an observer records.

Researchers cannot record every detail of everything that happens. They therefore identify behaviours that are relevant to the research aim and organise them into categories.

For example, a psychologist investigating cooperation during a group task might record:

• Sharing materials

• Giving another participant information

• Completing part of the task with another participant

The categories turn complex behaviour into data that can be recorded and analysed.

Operationalising behaviour

Behavioural categories must be operationalised. This means that each category is defined precisely enough for observers to recognise and record it.

A category such as “being friendly” is not sufficiently operationalised. Different observers may have different ideas about what friendliness looks like.

It could be replaced with observable categories such as:

• Smiles at another person

• Offers another person an object

• Begins a conversation with another person

These categories describe actions that can be seen or heard.

This applies the wider research-methods skill of defining variables and behaviours precisely.

Features of effective behavioural categories

Clear behavioural categories should be:

Observable

The category should describe behaviour that can be seen or heard.

An observer cannot directly record what someone is thinking or feeling. For example, “feels anxious” is not directly observable.

Observable behaviours might include:

• Taps a foot repeatedly

• Bites a fingernail

• Leaves their seat

• Asks for reassurance

These behaviours do not prove that a person feels anxious, but they can be observed and recorded consistently.

Clearly defined

Each category should explain exactly what counts as an example of the behaviour.

For example:

Vague category	Clearer operationalised category
Talks a lot	Speaks for longer than ten consecutive seconds
Is distracted	Looks away from the task materials for at least five seconds
Helps someone	Gives another person information or materials needed to complete the task
Behaves aggressively	Pushes, hits or kicks another person

The precise definition should be suitable for the research aim and setting.

Distinct from other categories

Categories should not overlap unnecessarily. If one behaviour could be placed into several categories, observers may record it differently.

For example, consider these categories:

• Talking

• Giving advice

• Helping

Giving advice involves talking and might also be interpreted as helping. The researcher must decide whether behaviours can be counted in more than one category or redesign the categories so that the distinction is clear.

Relevant to the aim

Every category should help answer the research question.

If a researcher is investigating aggressive behaviour, recording the colour of participants’ clothing would not normally be relevant.

Categories should focus on the behaviour being studied rather than every action that takes place.

Practical to use

Observers must be able to record the categories accurately while the behaviour is occurring.

A very large number of complicated categories may be difficult to use, particularly when several behaviours occur quickly or at the same time.

Constructing behavioural categories

A researcher could follow these steps:

1. State the research aim.

   
   

   Identify exactly which behaviour is being investigated.

2. Identify possible behaviours.

   
   

   Consider what observable actions are relevant to the aim.

3. Create preliminary categories.

   
   

   Turn the behaviours into a manageable set of categories.

4. Define each category.

   
   

   State precisely what the observer should record.

5. Check for overlap.

   
   

   Make sure that observers can distinguish one category from another.

6. Pilot the categories.

   
   

   Test whether observers can use them during a small-scale trial.

7. Revise the categories.

   
   

   Clarify, combine, divide or remove categories where necessary.

Testing categories before the main observation links directly to using a trial run to identify weaknesses.

Worked example: constructing behavioural categories

A researcher wants to observe off-task behaviour during an independent classroom activity.

The initial categories are:

• Distracted

• Not working

• Disruptive

These categories are subjective and overlap. A participant who talks to another student might be described using all three.

The researcher could replace them with:

Behavioural category	Operational definition
Talks about something unrelated to the task	Speaks to another student about a subject not connected to the activity
Leaves seat	Moves away from their allocated chair without the teacher’s permission
Looks away from task	Looks away from the worksheet or relevant equipment for at least five consecutive seconds
Uses an unrelated item	Uses an object that is not required for the activity

The revised categories describe observable actions rather than interpretations.

Observation schedules

An observation schedule is a record sheet containing the categories that the observer will use.

A simple schedule might contain:

Behavioural category	Tally	Total frequency
Shares materials	`
Gives task-related information	`
Completes a joint action	`

A schedule helps the observer record behaviour consistently and produces data that can be analysed.

The layout should be practical. Observers should be able to find each category quickly and enter a tally without missing later behaviour.

Event sampling

Event sampling involves recording every occasion on which a particular behaviour or event occurs during the observation period.

The observer usually makes a tally each time one of the behavioural categories occurs.

For example, a researcher observing aggressive behaviour in a playground might record every occasion on which a child:

• Pushes another child

• Hits another child

• Kicks another child

• Takes an object from another child

If pushing occurs five times, the observer records a frequency of five.

How event sampling works

A researcher using event sampling would:

1. Create operationalised behavioural categories.

2. Decide when the observation begins and ends.

3. Watch for each target behaviour throughout the observation.

4. Record every occurrence of each behaviour.

5. Calculate the frequency of each category.

Event sampling focuses on the occurrence of behaviour rather than on fixed intervals of time.

Strengths of event sampling

It provides a frequency count

The researcher can calculate exactly how often each recorded behaviour occurred.

This allows categories or participants to be compared.

It can record infrequent but important behaviour

If a behaviour occurs only occasionally, event sampling allows the observer to record it whenever it happens.

A time-sampling procedure might miss an uncommon event if it occurs between the selected recording points.

It can provide detailed information about target behaviours

Where the number of categories is manageable, every occurrence can be recorded rather than only a sample of behaviour.

Limitations of event sampling

Frequent behaviour may be difficult to record

If events happen rapidly, the observer may not be able to record every occurrence accurately.

This is particularly difficult if several target behaviours happen at the same time.

Continuous behaviour may be difficult to divide into separate events

Some behaviours do not have an obvious beginning and end.

For example, it may be difficult to decide whether a participant who speaks continuously for 30 seconds has displayed one event or several events.

The category must therefore specify what counts as a new occurrence.

The observer may miss other relevant behaviour

Focusing on particular events means that behaviour outside the predetermined categories is not recorded.

Time sampling

Time sampling involves recording behaviour at predetermined intervals during an observation period.

The observation is divided into units of time. The observer records whether a target behaviour occurs at the selected times or within the selected intervals.

For example, a researcher could observe a classroom for 20 minutes and record behaviour during every one-minute interval.

The researcher might record whether each participant is:

• Looking at the task

• Writing

• Talking about the task

• Engaged in unrelated behaviour

How time sampling works

A researcher using time sampling would:

1. Decide the total observation period.

2. Divide the period into fixed intervals.

3. Create clear behavioural categories.

4. Record the relevant behaviour at each predetermined interval.

5. Calculate how many intervals contained each behaviour.

The length of the interval should suit the behaviour being studied. If the interval is too long, important changes may be missed. If it is too short, the observer may struggle to record the behaviour accurately.

Worked example: time sampling

A psychologist observes a student for 12 minutes. Behaviour is recorded once during every 30-second interval.

Number of intervals:

12 minutes × 2 = 24 intervals

The student is recorded as on-task during 18 intervals.

Percentage of intervals containing on-task behaviour:

Percentage=(number of on-task intervals ÷ total number of intervals) × 100

Percentage=(18÷24)×100

Percentage=75%

The student was therefore recorded as on-task during 75% of the sampled intervals.

This does not necessarily mean the student was on-task for exactly 75% of the entire observation. It describes the proportion of sampled intervals in which the behaviour was recorded.

Strengths of time sampling

It makes observations manageable

The observer does not have to record every behaviour continuously.

This can be useful when behaviour occurs frequently or when several participants are being observed.

It provides a structured procedure

Predetermined intervals create a standardised system for recording behaviour.

Different observers can use the same categories and time intervals.

It can show patterns across an observation period

Researchers can examine whether behaviour is recorded more often during particular intervals.

Limitations of time sampling

Behaviour may be missed

An important event may occur between the selected intervals and therefore not be recorded.

The data provide a sample rather than a complete record of all behaviour.

Infrequent behaviour may be underrepresented

A rare event may not happen during any of the selected intervals, even though it occurs during the full observation period.

The choice of interval may influence the findings

Different interval lengths may produce different records.

Researchers should choose intervals that are appropriate for the frequency and duration of the behaviour.

Comparing event sampling and time sampling

Feature	Event sampling	Time sampling
What is recorded?	Every occurrence of a target behaviour	Behaviour at predetermined time intervals
Main type of data	Frequency of events	Number or proportion of intervals containing behaviour
Particularly useful for	Clear, separate events	Frequent or ongoing behaviour
Main strength	Can record every occurrence of a target event	Makes a busy observation more manageable
Main limitation	Rapid or simultaneous events may be missed	Behaviour occurring between intervals may be missed
Treatment of rare behaviour	May capture it whenever it occurs	May miss it if it falls outside the sampled intervals

Choosing between event sampling and time sampling

There is no sampling method that is always superior. The most appropriate choice depends on the research aim and behaviour.

Event sampling may be appropriate when:

• The target behaviour has a clear beginning and end.

• The researcher needs a frequency count.

• The behaviour is not occurring too rapidly.

• Infrequent events are important.

Time sampling may be appropriate when:

• Behaviour occurs frequently or continuously.

• Continuous recording would be impractical.

• The researcher wants a structured sample across a longer period.

• Several participants or categories must be observed.

Scenario: choosing event sampling

A psychologist wants to record how often customers ask a shop assistant for help.

Event sampling would be suitable because asking for help is a clear and separate event. The observer could record each occasion on which a customer asks a task-related question.

Time sampling could miss requests made between the selected recording intervals.

Scenario: choosing time sampling

A researcher wants to estimate how regularly students remain focused during a 40-minute lesson.

Time sampling could be suitable because concentration-related behaviour may be ongoing and difficult to divide into separate events.

The researcher could record whether each student is looking at or working with the relevant lesson materials at predetermined intervals.

Behavioural categories and reliability

Two observers should be able to use the same categories and record behaviour consistently.

Vague categories may lead observers to classify identical behaviour differently. Clear operational definitions make agreement more likely.

For example, observers may disagree about whether a participant is “interested”, but they are more likely to agree about whether the participant asks a question about the task.

Researchers can compare the observations recorded by different observers. This is studied more fully when revising measuring consistency between observers.

Clear categories do not guarantee agreement, but they reduce the need for subjective interpretation.

Behavioural categories and validity

Behavioural categories must represent the behaviour identified in the research aim.

A category can be easy to record but still provide a poor measure of the intended behaviour.

For example, looking at a worksheet may be used as a category of on-task behaviour. However, a participant could look at the worksheet without concentrating on it.

Researchers should therefore consider whether the chosen observable behaviours provide a reasonable measure of the concept being investigated.

Improving an observational design

An observational design may be improved by:

• Replacing subjective categories with observable actions.

• Defining precisely what counts as one occurrence.

• Removing overlapping categories.

• Adding an important behaviour that was not covered.

• Reducing the number of categories where observers cannot record them accurately.

• Changing from event sampling to time sampling when behaviour occurs too rapidly.

• Changing from time sampling to event sampling when rare events are being missed.

• Revising the length of the time intervals.

• Piloting the observation schedule.

• Training observers to apply the definitions consistently.

These decisions should be considered when planning the method and procedure of a complete study.

Applying observational design in an examination

Questions may ask you to construct categories, explain a sampling method or recommend an observational design for a scenario.

Pay attention to the command word:

• Identify means name the category or sampling method.

• Describe means state how the method would be carried out.

• Explain means show how or why the method works.

• Construct means create usable categories or a record sheet.

• Suggest means propose a method that fits the scenario.

• Justify means explain why the suggestion is appropriate.

A strong behavioural category should contain an action and a clear definition.

For example:

Sharing materials: voluntarily passes an item required for the task to another participant.

A weak category would be:

Cooperative behaviour

The weak category does not explain what the observer should actually record.

Key Words 🔑

Key word	Student-friendly definition	How it may be used in an exam
Observational design	The plan for deciding which behaviours will be recorded and when they will be recorded.	Explain or design a procedure for an observational study.
Behavioural category	A defined type of observable behaviour that an observer records.	Construct categories suitable for a research scenario.
Operationalisation	Defining a behaviour precisely so that it can be identified and recorded.	Improve vague or subjective behavioural categories.
Observation schedule	A record sheet containing the categories used during an observation.	Design a practical method for recording behaviour.
Event sampling	Recording every occurrence of a target behaviour during the observation.	Select a method when the frequency of a clear event is required.
Time sampling	Recording behaviour at predetermined intervals during an observation.	Select a method when continuous observation would be difficult.
Interval	A fixed unit of time used in a time-sampling procedure.	Explain how the researcher will organise the recording period.
Frequency	The number of times a behaviour is recorded.	Interpret or compare event-sampling results.
Tally	A mark made each time a behaviour occurs.	Show how event-sampling data could be recorded.
Inter-observer reliability	The extent to which different observers produce consistent records of the same behaviour.	Explain why categories must be clear and operationalised.

Common Mistakes ⚠️

Mistake: Using feelings or personality traits as behavioural categories.

Why this is incorrect:Thoughts, feelings and personality traits cannot be observed directly. Different observers may make different assumptions about them.

How to improve:Replace the inferred state with a visible or audible action, such as smiling, speaking or leaving a seat.

Mistake: Creating categories that are too vague.

Why this is incorrect:A category such as “misbehaves” does not explain exactly what the observer should record.

How to improve:State the precise action, such as “throws an object” or “speaks while another participant is speaking”.

Mistake: Using overlapping categories.

Why this is incorrect:Observers may place the same behaviour into different categories, reducing consistency.

How to improve:Clarify the boundaries between categories or combine categories that cannot be distinguished reliably.

Mistake: Confusing event sampling with time sampling.

Why this is incorrect:Event sampling records each occurrence of a target behaviour. Time sampling records behaviour at predetermined time intervals.

How to improve:Ask whether the recording is triggered by the behaviour occurring or by a particular time being reached.

Mistake: Claiming that time sampling records every behaviour.

Why this is incorrect:Behaviour occurring outside the selected intervals may not be recorded.

How to improve:Describe time sampling as producing a sample of behaviour across the observation period.

Mistake: Assuming event sampling is always more accurate.

Why this is incorrect:If behaviour occurs rapidly or several events happen simultaneously, the observer may be unable to record every occurrence.

How to improve:Choose the procedure according to the frequency, duration and clarity of the target behaviour.

Mistake: Naming a sampling method without applying it to the scenario.

Why this is incorrect:An application question requires an explanation of why the method fits the behaviour being investigated.

How to improve:Refer directly to whether the behaviour is rare, frequent, continuous or easy to divide into separate events.

Mistake: Giving categories without operational definitions.

Why this is incorrect:A category title alone may not tell observers what should count as an occurrence.

How to improve:Include a precise description of the visible or audible action that must occur.

Exam-Style Questions ✍️

Question 1

Define a behavioural category. (1 mark)

Question 2

State one difference between event sampling and time sampling. (2 marks)

Question 3

A psychologist wants to record how often passengers offer their seat to another person on a bus.

Explain why event sampling would be appropriate for this observation. (2 marks)

Question 4

A researcher creates the following behavioural categories for an observation of a group activity:

• Helpful

• Friendly

• Interested

Explain one problem with these behavioural categories. (2 marks)

Question 5

A psychologist wants to observe cooperative behaviour during a group problem-solving task.

Construct two suitable behavioural categories. Each category must include a clear operational definition. (4 marks)

Question 6

A researcher observes a participant for 15 minutes and records behaviour once every 30 seconds.

The participant is recorded as looking at the task during 21 of the sampled intervals.

Calculate the percentage of sampled intervals during which the participant was recorded as looking at the task. Show your working. (3 marks)

Question 7

A psychologist uses event sampling to observe behaviour in a busy playground. During the pilot observation, several target behaviours occur at the same time and the observer is unable to record them all.

Explain one limitation of the researcher’s event-sampling procedure and suggest one improvement. (4 marks)

Question 8

A researcher wants to investigate off-task behaviour during lessons. The following categories are proposed:

Category	Definition
Not concentrating	The student does not appear to be concentrating
Disruptive	The student behaves in a disruptive way
Talking	The student speaks

Explain two weaknesses of the proposed categories and suggest an improvement for each weakness. (6 marks)

Question 9

A psychologist wants to investigate how frequently staff members communicate with residents in a care setting. The psychologist is considering event sampling and time sampling.

Compare the two sampling procedures and recommend the more appropriate procedure for this investigation. Justify your answer. (6 marks)

Question 10

A researcher plans to observe social interaction in a college common room.

Design an observational procedure for this investigation. In your answer, include:

• Two operationalised behavioural categories

• A suitable sampling method

• An explanation of how behaviour would be recorded

• One way of improving the consistency of the observations

(8 marks)