Evaluating observations | AQA A-Level Psychology Revision

For 7182 specification, first teach in September 2025

AQA A-Level Psychology | Free Revision Notes

Estimated study time: 50 minutes

These Evaluating observations A-Level Psychology revision notes examine the strengths, limitations, practical problems and ethical issues associated with observational research. You will compare naturalistic and controlled, covert and overt, and participant and non-participant observations. You will also learn how to select and justify an appropriate method for an unfamiliar research scenario. Review the defining features of each approach in types of observational research before evaluating their suitability.

Learning Objectives 🎯

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

• Evaluate naturalistic and controlled observations.

• Evaluate covert and overt observations.

• Evaluate participant and non-participant observations.

• Identify practical problems affecting observational research.

• Identify ethical issues in observational studies.

• Explain how ethical issues may be managed.

• Select a suitable observational method for a research scenario.

• Justify a methodological choice using features of the proposed investigation.

Revision Notes 📚

Evaluating observations A-Level Psychology revision overview

An observation is a research technique in which psychologists watch and record behaviour.

Observations may be classified as:

• Naturalistic or controlled.

• Covert or overt.

• Participant or non-participant.

Each choice produces a different balance between:

• Control.

• Realism.

• Participant awareness.

• Ethical protection.

• Researcher involvement.

• Reliability.

• Validity.

• Practical feasibility.

There is no single observational method that is best for every investigation.

The most appropriate method depends on:

• The behaviour being studied.

• Where that behaviour normally occurs.

• Whether awareness would alter behaviour.

• Whether the researcher needs to join the group.

• The ethical risks.

• How precisely the behaviour must be recorded.

• The resources available.

Evaluation requires explanation

A strength or limitation should not be presented as an isolated label.

For example:

Naturalistic observations have high ecological validity.

This is incomplete because it does not explain why or how the feature affects the findings.

A developed evaluation would state:

A naturalistic observation records behaviour in the setting where it would ordinarily occur. Participants may therefore behave more normally than they would in an artificial research room, increasing ecological validity. However, this advantage may be reduced if participants know they are being observed and change their behaviour.

A strong evaluation point includes:

1. A feature of the method.

2. An explanation of why that feature matters.

3. Its effect on the quality of the evidence.

4. A qualification or counterpoint where appropriate.

Methodological trade-offs

Observational decisions often involve a trade-off.

For example:

• Increasing control may reduce realism.

• Concealing the observation may reduce reactivity but create ethical problems.

• Joining a group may improve understanding but reduce objectivity.

• Remaining separate may improve recording but limit insight into the situation.

• Broad behavioural categories may capture complex behaviour but reduce reliability.

• Narrow categories may improve reliability but miss meaningful context.

Selecting a method therefore involves deciding which feature matters most for the research question.

Evaluating naturalistic observations

Strength: behaviour occurs in its ordinary setting

A naturalistic observation takes place in the environment where the behaviour would normally occur.

Examples include observing:

• Pupils during an ordinary lesson.

• Customers in a real shop.

• Commuters at a railway station.

• Employees during a usual meeting.

• Children in their regular playground.

Because the context is genuine, participants encounter:

• Real social relationships.

• Ordinary environmental demands.

• Meaningful consequences.

• Familiar routines.

This may produce behaviour that is more representative of everyday life.

Strength: ecological validity

Naturalistic observations may have high ecological validity.

Ecological validity concerns whether findings represent behaviour in everyday settings.

For example, cooperation observed during an ordinary classroom activity may provide a more realistic picture than cooperation observed during a specially created puzzle task.

This can make it more appropriate to generalise the findings to similar real-life settings.

However, a natural setting does not guarantee ecological validity.

Behaviour may still be altered if:

• Participants know that they are being observed.

• Recording equipment is highly noticeable.

• The observer disrupts the setting.

• The behaviour occurs only because of an unusual event.

The whole procedure must therefore be considered.

Strength: access to spontaneous behaviour

Naturalistic observations can record behaviour that develops spontaneously.

Researchers may observe:

• Unexpected interactions.

• Responses to changing situations.

• Behaviour participants might not report accurately.

• Actions that would be difficult to reproduce in a controlled setting.

This can provide rich information about how behaviour unfolds naturally.

The researcher is not limited to behaviour produced by a highly structured task.

Strength: reduced task artificiality

Participants are not normally asked to complete a researcher-created activity.

This reduces the risk that the behaviour reflects:

• Confusion about an artificial task.

• Unusual research instructions.

• A lack of personal relevance.

• The unfamiliarity of a laboratory setting.

For example, observing actual customer choices may produce more valid evidence than asking participants to imagine what they would buy.

Limitation: limited control of extraneous variables

A major limitation of naturalistic observation is limited control over the environment.

Behaviour may be affected by:

• Noise.

• Weather.

• Time of day.

• Number of people present.

• Unexpected interruptions.

• Existing relationships.

• Features of the physical environment.

• Events occurring before the observation.

These factors may make it difficult to explain why the behaviour occurred.

For example, pupils may participate less in one lesson because the classroom is unusually hot rather than because of the teaching method being investigated.

Limitation: difficulty replicating the observation

Natural settings change.

Another researcher may be unable to recreate:

• The same people.

• The same interactions.

• The same environmental conditions.

• The same unexpected events.

• The same social atmosphere.

This can make exact replication difficult.

A researcher may repeat the observation procedure in a similar setting, but the circumstances will not be identical.

This may reduce confidence that differences between observations reflect genuine behavioural patterns rather than changes in the environment.

The importance of consistency is examined further in reliability of psychological research.

Limitation: behaviour may not occur

In a naturalistic observation, the researcher cannot guarantee that the target behaviour will happen during the observation period.

For example, a psychologist studying helping behaviour may observe a public area for several hours but record very few opportunities for helping.

This creates practical problems:

• Long periods may be required.

• Few observations may be collected.

• The data may not represent the full range of behaviour.

• Researcher time and resources may be wasted.

A controlled observation can create the same opportunity for behaviour for every participant, although this may make the situation less natural.

Limitation: recording behaviour may be difficult

Natural environments may contain:

• Several people acting at once.

• Behaviour occurring outside the observer’s view.

• Background noise.

• Movement between locations.

• Rapid or ambiguous interactions.

The observer may miss relevant behaviour or record it inaccurately.

Video or audio recording may help, but this can create:

• Ethical concerns.

• Privacy issues.

• Practical difficulties.

• Awareness of the observation.

Limitation: low internal validity

If an observation is part of an investigation comparing groups or situations, uncontrolled environmental differences may reduce internal validity.

Internal validity concerns whether the researcher’s explanation of the findings is accurate.

Suppose one playground shows more aggressive behaviour than another.

The difference might reflect:

• Playground size.

• Adult supervision.

• Number of pupils.

• Age of the children.

• Available equipment.

• School rules.

The observed setting may therefore provide realistic evidence without allowing a clear explanation of the cause.

Evaluating controlled observations

Strength: greater control

A controlled observation takes place in a situation or environment arranged by the researcher.

The researcher may control:

• The instructions.

• Materials.

• Time available.

• Seating arrangements.

• Number of people present.

• Opportunities for the target behaviour.

• Recording equipment.

This allows participants to be observed under more consistent conditions.

Strength: standardisation

A controlled observation is easier to standardise.

For example, every pair of participants might:

• Enter the same room.

• Receive the same puzzle.

• Hear identical instructions.

• Have ten minutes to complete the task.

• Be observed using the same cameras.

Standardisation reduces the possibility that differences in behaviour result from participants receiving different procedures.

It also makes the observation easier to replicate.

Strength: opportunities for behaviour can be created

Researchers can arrange situations in which the target behaviour is likely to occur.

For example, to study cooperation, the researcher could provide:

• A task requiring shared resources.

• A puzzle that needs two people.

• A fixed opportunity to ask for help.

This avoids waiting for the behaviour to arise naturally.

It also allows every participant to encounter a similar opportunity.

Strength: easier recording

A controlled setting may allow the researcher to:

• Position cameras appropriately.

• Reduce background noise.

• Observe all participants clearly.

• Use a standard recording sheet.

• Keep the observation period consistent.

• Review recordings afterwards.

This can improve the accuracy and completeness of the data.

Strength: greater replicability

Because the researcher controls the procedure, another researcher can reproduce it more closely.

They can use:

• The same room arrangement.

• The same materials.

• The same instructions.

• The same behavioural categories.

• The same observation period.

Replication helps psychologists determine whether a finding is consistent.

Limitation: artificiality

Participants may behave differently in a researcher-arranged setting.

They may find the task:

• Unfamiliar.

• Unrealistic.

• Unimportant.

• Too simple.

• Unlike situations encountered in everyday life.

For example, cooperation during a ten-minute laboratory puzzle may not represent cooperation between friends completing a meaningful real-world activity.

Limitation: lower ecological validity

The controlled environment may reduce ecological validity.

Participants may be responding to:

• The research room.

• The presence of cameras.

• Standardised instructions.

• An artificial group task.

• The knowledge that their behaviour is being assessed.

The findings may therefore be difficult to generalise to everyday situations.

However, controlled observations do not always have low ecological validity.

A carefully designed task may reproduce important features of real behaviour.

The judgement should be applied to the specific procedure.

Limitation: demand characteristics

Participants in a controlled observation often know that they are taking part in research.

They may guess:

• Which behaviour is being studied.

• What the researcher expects.

• Which responses appear socially acceptable.

They may then change their behaviour.

For example, participants who realise cooperation is being observed may share more frequently than they normally would.

This reduces the validity of the observed behaviour.

The influence of research cues is considered in participant and researcher influences.

Limitation: observer presence

Even when participants do not know the precise aim, visible observers or cameras may make them:

• Self-conscious.

• Anxious.

• More careful.

• Less spontaneous.

• More socially desirable in their behaviour.

This is known as behavioural reactivity.

The behaviour recorded may represent how people act when watched rather than how they usually act.

Naturalistic and controlled observations compared

Feature	Naturalistic observation	Controlled observation
Setting	Ordinary environment	Researcher-arranged situation
Environmental control	Lower	Higher
Standardisation	More difficult	Easier
Behavioural realism	Often higher	May be lower
Ecological validity	Often higher	May be reduced
Replication	More difficult	Easier
Opportunity for behaviour	Cannot be guaranteed	Can be arranged
Recording	May be difficult	Usually easier
Unexpected influences	More likely	Easier to reduce

A suitable choice depends on the aim.

Choose naturalistic observation when the priority is realistic everyday behaviour.

Choose controlled observation when the priority is standardisation, control and reliable recording.

Evaluating covert observations

Strength: reduced behavioural reactivity

In a covert observation, people do not know they are being observed for research.

Because they are unaware, they are less likely to:

• Guess the research aim.

• Behave in a socially desirable way.

• Try to help the researcher.

• Resist the expected behaviour.

• Become self-conscious.

This may produce more natural behaviour.

Strength: reduced demand characteristics

Demand characteristics are likely to be reduced because participants are unaware that research is taking place.

For example, commuters who do not know that helping behaviour is being studied may respond more naturally than volunteers who know their helpfulness is being assessed.

This may increase validity.

However, participants may still change their behaviour if:

• They notice the observer.

• The situation appears unusual.

• Another part of the procedure reveals that they are being studied.

Strength: access to behaviour that people may conceal

Some forms of behaviour may be difficult to investigate overtly because participants would deliberately suppress them.

Covert observation may provide access to:

• Rule-breaking.

• Prejudice.

• Dishonest behaviour.

• Socially undesirable conduct.

• Informal group interactions.

However, the sensitivity of these behaviours also increases the ethical risks.

Limitation: lack of informed consent

Participants in a covert observation cannot normally provide informed consent before the study because they do not know it is taking place.

They have not knowingly agreed to:

• Be observed.

• Have their behaviour recorded.

• Have their information analysed.

• Participate in psychological research.

This conflicts with the principle that participants should make an informed choice about participation.

Limitation: difficulty exercising the right to withdraw

Participants who do not know they are being studied cannot request to leave the research during the observation.

They may also be unable to request the removal of their data unless:

• They are identified and debriefed later.

• The researcher provides a suitable opportunity to withdraw.

This is particularly difficult when the observation involves large numbers of anonymous people.

Limitation: privacy

Covert observation may invade privacy.

The ethical significance depends partly on:

• Where the behaviour occurs.

• Whether people reasonably expect to be observed.

• Whether the behaviour is sensitive.

• Whether individuals can be identified.

• Whether recordings are made.

Observing general public behaviour in an open location may be less intrusive than secretly recording private interactions.

Researchers must consider reasonable expectations of privacy rather than assuming that any visible behaviour is available for research.

Limitation: deception

Covert observations may involve deceiving people about:

• The researcher’s identity.

• The purpose of the observer’s presence.

• Whether data are being collected.

• How the information will be used.

Deception may be especially significant in covert participant observation, where the researcher develops relationships under a false role.

Limitation: debriefing may be difficult

Researchers may try to reduce ethical problems by providing a debrief after the observation.

A debrief can:

• Reveal the true purpose.

• Explain why concealment was used.

• Offer an opportunity to ask questions.

• Allow participants to withdraw their data.

However, debriefing may be impractical when:

• Many anonymous people were observed.

• Participants have already left.

• Contact details are unavailable.

• Revealing the observation would cause distress.

• The behaviour occurred over a long period.

Covert observation should not be selected simply because it is methodologically convenient.

Evaluating overt observations

Strength: informed consent can be obtained

In an overt observation, participants know that research observation is occurring.

Researchers can explain:

• What participation involves.

• How behaviour will be recorded.

• How data will be used.

• Possible risks.

• Participants’ rights.

This makes informed consent easier to obtain.

Strength: right to withdraw

Participants can be told that they may:

• Leave the study.

• Ask the observer to stop.

• Refuse recording.

• Request removal of their data.

This provides greater participant control than covert observation.

Strength: easier ethical protection

Because the research is open, psychologists can more easily:

• Obtain permission.

• Protect confidentiality.

• Explain recording procedures.

• Debrief participants.

• Respond to concerns.

• Gain permission from relevant organisations.

Overt observation is therefore generally easier to justify ethically.

Strength: recording equipment can be used openly

Researchers may be able to position cameras or microphones openly and obtain consent for their use.

This may improve data quality because:

• Behaviour can be reviewed repeatedly.

• Several observers can code the same recording.

• Missed events can be checked.

• Inter-observer reliability can be assessed.

Limitation: behavioural reactivity

Participants may alter their behaviour because they know they are being observed.

For example, they may:

• Behave more politely.

• Follow rules more closely.

• Avoid aggression.

• Participate more actively.

• Try to present themselves positively.

The behaviour may therefore lack validity.

Limitation: demand characteristics

Participants may guess the focus of the observation.

Suppose employees know a psychologist is studying workplace communication.

They may:

• Speak more frequently.

• Avoid interrupting.

• Include quieter colleagues.

• conceal disagreements.

The observation could then measure participants’ response to the research situation rather than their usual communication.

Limitation: social desirability

Social desirability occurs when people behave in ways that present themselves favourably.

In an overt observation, participants may choose actions they believe are:

• Kind.

• Responsible.

• Cooperative.

• Acceptable.

• Expected by the researcher.

This is particularly likely when the behaviour has an obvious moral or social value.

Habituation

Behavioural reactivity may decrease through habituation.

Habituation occurs when participants become accustomed to the observer and gradually return to more normal behaviour.

A researcher might:

• Attend several sessions before collecting data.

• Leave recording equipment in place for a period.

• Spend time in the setting without actively recording.

However, habituation:

• Requires additional time.

• May not eliminate all reactivity.

• May be less effective if the observer’s purpose remains obvious.

Covert and overt observations compared

Feature	Covert observation	Overt observation
Participant awareness	Unaware	Aware
Behavioural reactivity	Usually lower	Usually higher
Demand characteristics	Often reduced	More likely
Informed consent	Difficult or absent	Easier to obtain
Right to withdraw	Difficult to exercise	Easier to protect
Privacy concerns	Often greater	Easier to manage openly
Use of recording equipment	May require concealment	Can be agreed openly
Debriefing	May be difficult	Usually easier

The choice involves a clear trade-off:

• Covert observation may increase behavioural validity.

• Overt observation offers stronger ethical protection.

Evaluating participant observations

Strength: detailed understanding

In a participant observation, the observer joins the group or activity.

Participation may help the researcher understand:

• Group routines.

• Informal rules.

• Relationships.

• Shared meanings.

• Context surrounding behaviour.

• Events outsiders might misunderstand.

This may produce a deeper and more informed interpretation.

Strength: access to restricted groups

Some groups may not allow an outside observer to watch closely.

Joining the group may provide access to:

• Private interactions.

• Informal conversations.

• Group decision-making.

• Behaviour that occurs only among accepted members.

This can produce detailed evidence unavailable to a non-participant observer.

However, gaining this access covertly may create substantial ethical problems.

Strength: participants may behave naturally

Once accepted as a member, the participant observer may become part of the ordinary social environment.

Group members may pay less attention to the observer’s presence than they would to an outsider sitting with a clipboard.

This may reduce behavioural reactivity.

The advantage is more likely when:

• The researcher has spent time with the group.

• Participation appears natural.

• Members are comfortable with the observer.

• The observer does not interrupt activities to record data.

Limitation: loss of objectivity

A participant observer may become emotionally or socially involved with the group.

They may:

• Develop friendships.

• Adopt the group’s viewpoint.

• Sympathise with particular members.

• Avoid recording negative behaviour.

• Interpret events in ways that protect the group.

This can produce observer bias.

The researcher may become less able to evaluate behaviour objectively.

Limitation: difficulty recording behaviour

The observer must participate and observe at the same time.

This may make it difficult to:

• Take notes.

• Use a checklist.

• Watch every participant.

• Record behaviour immediately.

• Remember exact sequences of events.

• Measure frequency or duration accurately.

Taking visible notes could reveal a covert observer’s role or disrupt the interaction.

Recording information later relies on memory and may reduce accuracy.

Limitation: observer influence

By joining the group, the researcher becomes part of the situation.

Their actions may influence:

• Group decisions.

• Conversation topics.

• Relationships.

• Emotional responses.

• The behaviour being recorded.

The observer may therefore change the group rather than merely observe it.

This is a form of investigator effect.

Limitation: personal risk

Participant observation may expose the researcher to:

• Stress.

• Conflict.

• Unsafe environments.

• Pressure to take part in inappropriate behaviour.

• Difficulty leaving the group.

• Emotional involvement.

Researchers need procedures that protect them as well as the participants.

Limitation: role conflict

The researcher may face a conflict between:

• Acting as a genuine group member.

• Remaining a responsible researcher.

• Protecting participants.

• Recording behaviour accurately.

• Avoiding harm.

For example, a covert participant observer might learn that someone is at risk. Remaining loyal to the research role could conflict with the responsibility to protect that person.

Evaluating non-participant observations

Strength: greater objectivity

A non-participant observer remains separate from the group or activity.

This distance may make it easier to:

• Apply behavioural categories consistently.

• Avoid emotional involvement.

• Record all group members.

• Focus on the research aim.

• Remain neutral.

The observer is less likely to adopt the group’s perspective completely.

Strength: easier recording

Because the observer does not have to take part, they can concentrate on:

• Completing a checklist.

• Counting events.

• Measuring duration.

• Making detailed notes.

• Monitoring several participants.

• Using recording equipment.

This may improve accuracy.

Strength: standardisation

A non-participant observer can follow a structured recording procedure.

For example, they may:

• Sit in the same position.

• Observe for a fixed period.

• Use predetermined categories.

• Record every occurrence of a target behaviour.

This supports reliability.

Limitation: limited insight

An observer outside the group may not understand:

• The meaning of an interaction.

• Existing relationships.

• Group-specific language.

• The history surrounding behaviour.

• Why a person responded in a particular way.

The researcher may record what happened without fully understanding its context.

Limitation: observer presence may still affect behaviour

Non-participant does not mean unnoticed.

A visible observer sitting apart from the group may attract attention and increase behavioural reactivity.

Participants may feel that they are being judged more strongly by a detached observer than by someone joining the activity.

Limitation: restricted access

The observer may be excluded from:

• Private discussions.

• Informal interactions.

• Restricted locations.

• Sensitive group events.

• Behaviour occurring away from the observation point.

The resulting data may represent only behaviour visible to outsiders.

Participant and non-participant observations compared

Feature	Participant observation	Non-participant observation
Observer role	Joins the group	Remains separate
Contextual understanding	Often greater	May be limited
Access	May reach private or informal behaviour	May be restricted
Objectivity	May be reduced by involvement	Usually easier to maintain
Recording accuracy	Often more difficult	Usually easier
Observer influence	Researcher actively enters the interaction	Presence may still influence behaviour
Practical demands	Managing two roles	Focusing mainly on recording

Choose participant observation when detailed understanding and access are especially important.

Choose non-participant observation when systematic and objective recording is the priority.

Practical issues in observational research

Defining behaviour

Researchers must decide exactly what will count as the target behaviour.

A category such as “aggression” is too vague unless it is defined through observable actions.

Possible categories might include:

• Hitting another person.

• Pushing another person.

• Taking an object by force.

• Shouting a specified insult.

Clear behavioural categories improve consistency.

Their construction is examined fully in behavioural categories and sampling behaviour.

Observer bias

Observer bias occurs when expectations affect how behaviour is interpreted or recorded.

For example, a researcher expecting one group to be more aggressive may:

• Notice more of its aggressive actions.

• Interpret ambiguous behaviour as aggressive.

• overlook similar behaviour in another group.

• Record borderline cases inconsistently.

Observer bias reduces validity.

It can be reduced through:

• Clear behavioural categories.

• Observer training.

• Blind observation where possible.

• Multiple observers.

• Video recording.

• Checking inter-observer reliability.

Inter-observer reliability

Inter-observer reliability is the level of agreement between two or more observers recording the same behaviour.

A simple procedure is:

1. Two observers watch the same behaviour.

2. Each records it independently.

3. Their records are compared.

4. The level of agreement is assessed.

High agreement suggests that:

• The behavioural categories are clear.

• Observers are applying them consistently.

• The data are less dependent on one person’s judgement.

Low agreement suggests that:

• Categories may be ambiguous.

• Observer training may be inadequate.

• The behaviour may be difficult to classify.

• Procedures may need revising.

Observer training

Observers should practise using the behavioural categories before the main study.

Training may involve:

• Reviewing category definitions.

• Coding sample videos.

• Discussing disagreements.

• Clarifying borderline examples.

• Repeating practice until agreement improves.

A pilot study may help identify problems before the full observation begins.

Behaviour occurring too quickly

Some behaviour occurs rapidly or simultaneously.

This makes it difficult for one observer to record everything accurately.

Researchers may respond by:

• Using several observers.

• Video recording the session.

• Observing fewer participants.

• Using a smaller number of categories.

• Selecting time or event sampling.

• Reviewing recordings more than once.

Large amounts of data

An unstructured observation may produce extensive notes that are difficult to:

• Organise.

• Compare.

• Analyse.

• Summarise objectively.

Structured behavioural categories make data more manageable but may remove important context.

The researcher must decide whether the aim requires:

• Detailed qualitative description.

• Quantitative behavioural counts.

• A combination of both.

Time and cost

Observations can be time-consuming because researchers may need to:

• Gain access to a setting.

• Train observers.

• Wait for behaviour to occur.

• Observe for long periods.

• Transcribe recordings.

• Code video repeatedly.

• Check reliability.

Participant and naturalistic observations may be particularly time-consuming.

Access to the setting

Researchers may need permission from:

• Schools.

• Employers.

• Healthcare settings.

• Community organisations.

• Parents or guardians.

• Owners of private locations.

Access may be refused if the research:

• Disrupts normal activity.

• Involves sensitive behaviour.

• Requires recording equipment.

• Creates privacy concerns.

• Places participants at risk.

Sampling behaviour

Researchers cannot always record every behaviour occurring throughout an entire day.

They may use:

• Event sampling.

• Time sampling.

These approaches can make observation manageable but may also miss relevant behaviour.

The design of these recording procedures is covered in Observational design.

Selecting a suitable time

The observation period must provide a fair opportunity for the behaviour to occur.

For example, observing playground cooperation for five minutes on one unusually quiet day may not provide representative evidence.

Researchers should consider:

• Time of day.

• Day of the week.

• Length of observation.

• Number of sessions.

• Whether the period is typical.

• Whether unusual events are occurring.

Recording equipment

Cameras and microphones may improve accuracy, but researchers must consider:

• Consent.

• Privacy.

• Secure storage.

• Visibility of the equipment.

• Areas outside the camera’s view.

• Technical failure.

• Whether recording changes behaviour.

The presence of equipment may improve reliability while reducing behavioural validity.

Ethical issues in observational research

Informed consent

Participants should normally understand what participation involves before agreeing.

This is straightforward in overt observation but difficult in covert observation.

Where full information would destroy the validity of the research, psychologists must consider whether:

• Withholding information is necessary.

• The study has sufficient value.

• Risks are minimal.

• Participants can be debriefed later.

• Data can be withdrawn after debriefing.

Ethical decision-making should be guided by ethical principles in psychological studies.

Deception

Deception occurs when participants are:

• Misled.

• Given incomplete information.

• Unaware of the researcher’s true role.

• Unaware that observation is taking place.

Covert participant observation may involve particularly extensive deception because the researcher forms relationships under a concealed identity.

Any deception should be:

• Necessary.

• Proportionate.

• unlikely to cause significant harm.

• Explained during debriefing where possible.

Privacy

People may reasonably expect privacy even when their behaviour can technically be seen.

Researchers should consider whether the observation occurs in:

• A public place.

• A semi-private organisation.

• A private home.

• An online space with restricted membership.

• A sensitive setting.

They should also consider whether:

• Individuals can be identified.

• Sensitive information is being recorded.

• Behaviour would normally be seen by strangers.

• Recording creates a more permanent intrusion than ordinary observation.

Confidentiality

Researchers must protect participants’ identities and personal information.

This may involve:

• Removing names.

• Using participant codes.

• Avoiding identifiable descriptions.

• Storing recordings securely.

• Limiting access to raw data.

• Reporting group patterns rather than individual details.

• Destroying data when appropriate.

Video and audio recordings require particular care because individuals may be recognisable.

Right to withdraw

Participants in overt observations should be told:

• They may stop taking part.

• They may ask not to be recorded.

• They may request removal of their data.

In covert observation, the right to withdraw is difficult to protect during data collection.

Where participants are debriefed, they should be given an opportunity to withdraw their data where possible.

Protection from harm

Observation should not expose participants to greater physical or psychological risk than they would ordinarily experience.

Researchers should avoid procedures that cause:

• Distress.

• embarrassment.

• conflict.

• humiliation.

• Damage to relationships.

• Identification of sensitive behaviour.

• Unsafe situations.

The researcher’s decision to observe rather than intervene may itself create an ethical issue if someone is at risk.

Debriefing

A debrief should explain:

• The true aim.

• What behaviour was recorded.

• Why any concealment was necessary.

• How the data will be used.

• Participants’ right to ask questions.

• Whether data can be withdrawn.

A debrief is not a complete solution if participants experienced serious harm or a major invasion of privacy.

Ethical problems should be reduced during the design stage rather than addressed only afterwards.

Vulnerable participants

Additional safeguards may be needed when observation involves:

• Children.

• People with reduced capacity to consent.

• Patients.

• People in dependent relationships.

• Individuals discussing sensitive experiences.

Researchers may need consent from an appropriate responsible adult or organisation as well as the participant’s agreement.

Care must be taken not to assume that organisational permission replaces individual ethical protection.

Public behaviour and consent

Some observational research may involve anonymous public behaviour.

Researchers must still consider:

• Whether the location is genuinely public.

• Whether the behaviour is sensitive.

• Whether the research intervention affects people.

• Whether participants could be identified.

• Whether recording is necessary.

• Whether consent is practical.

The absence of direct interaction does not automatically remove ethical responsibility.

Evaluating combinations of observational methods

Naturalistic, covert and non-participant observation

Example:

A researcher anonymously records queue-jumping in a public station without joining the queue.

Possible strengths:

• Behaviour occurs in a genuine setting.

• Participants are unaware.

• Demand characteristics are reduced.

• The observer can concentrate on recording.

Possible limitations:

• Environmental variables are uncontrolled.

• Informed consent is absent.

• Privacy must be considered.

• Behaviour may occur too quickly to record reliably.

• The observer may not understand the context of each incident.

Naturalistic, overt and non-participant observation

Example:

A psychologist openly observes pupils during an ordinary lesson without joining the class.

Possible strengths:

• Behaviour occurs in a familiar environment.

• Consent can be obtained.

• The observer can use a structured checklist.

• Ethical procedures are easier to manage.

Possible limitations:

• Pupils may change their behaviour.

• The observer may miss behaviour outside their view.

• Classroom conditions are difficult to control.

• The researcher may lack insight into relationships between pupils.

Naturalistic, covert and participant observation

Example:

A researcher secretly joins an existing social group.

Possible strengths:

• Access to genuine group behaviour.

• Reduced behavioural reactivity.

• Detailed understanding of group culture.

• Access to interactions hidden from outsiders.

Possible limitations:

• Serious consent and deception issues.

• Difficulty recording accurately.

• Risk of becoming emotionally involved.

• Possible danger to the researcher.

• Loss of objectivity.

• Difficulty leaving or debriefing the group.

Controlled, overt and non-participant observation

Example:

Participants complete a planned team task in a research room while an observer records behaviour.

Possible strengths:

• Strong procedural control.

• Consistent opportunities for behaviour.

• Easy use of recording equipment.

• Replication is possible.

• Consent and withdrawal can be managed.

Possible limitations:

• Artificial behaviour.

• Demand characteristics.

• Socially desirable responding.

• Reduced ecological validity.

Controlled, overt and participant observation

Example:

A researcher openly joins participants during a standardised group problem-solving task.

Possible strengths:

• The task is consistent across groups.

• The researcher gains direct insight into interaction.

• Ethical procedures can be explained.

• The same situation can be repeated.

Possible limitations:

• The researcher may influence the group.

• Participation may make accurate recording difficult.

• Behaviour may be artificial.

• Participants may guess the aim.

Selecting an appropriate observational method

Begin with the research aim

The research aim should determine the method.

Ask:

• Does the researcher need natural everyday behaviour?

• Must every participant receive the same opportunity?

• Would awareness seriously alter behaviour?

• Can consent be obtained without damaging the research?

• Does the observer need to understand the group from within?

• Is objective behavioural counting the main priority?

• Is the setting safe and accessible?

Choosing naturalistic observation

Naturalistic observation may be appropriate when:

• The behaviour depends heavily on its everyday context.

• An artificial task would distort the behaviour.

• The researcher wants realistic behaviour.

• The natural setting is accessible.

• Environmental variation can be tolerated.

Example:

Investigating how pupils form friendship groups during their usual break.

The ordinary playground context may be essential to the behaviour.

Choosing controlled observation

Controlled observation may be appropriate when:

• Every participant needs the same opportunity to behave.

• Behaviour is rare or difficult to observe naturally.

• Standardisation is important.

• Recording must be precise.

• The study must be replicable.

Example:

Comparing cooperation when pairs receive the same puzzle and limited resources.

The researcher can ensure that every pair encounters the same task.

Choosing covert observation

Covert observation may be methodologically useful when:

• Awareness is likely to change behaviour substantially.

• The behaviour occurs in a genuinely public setting.

• The observation involves minimal risk.

• Identifiable information is unnecessary.

• The scientific value justifies concealment.

Ethical acceptability must still be considered carefully.

Choosing overt observation

Overt observation may be appropriate when:

• Informed consent is essential.

• The behaviour is sensitive.

• Recording equipment is required.

• Participants need to exercise withdrawal rights.

• The setting is private.

• The researcher can allow time for habituation.

Choosing participant observation

Participant observation may be appropriate when:

• Group membership is necessary for access.

• The meaning of behaviour depends on group culture.

• Informal interactions are central to the aim.

• An external observer would be excluded.

• Rich contextual understanding is required.

Choosing non-participant observation

Non-participant observation may be appropriate when:

• Accurate counting is the priority.

• The observer needs to use a checklist.

• Several people must be monitored.

• Researcher influence should be reduced.

• Joining the activity is unnecessary or inappropriate.

A decision framework for scenarios

Use the following steps.

Step 1: Identify the target behaviour

What exactly does the researcher want to observe?

For example:

• Helping.

• Aggression.

• Cooperation.

• Classroom participation.

• Customer choice.

Step 2: Identify the normal setting

Where does the behaviour ordinarily occur?

Choose naturalistic observation if the genuine context is essential.

Choose controlled observation if identical conditions are more important.

Step 3: Consider awareness

Would knowing about the observation alter behaviour significantly?

If yes, covert observation may produce more valid behaviour, but ethical problems must be addressed.

If ethical protection is the priority, overt observation may be preferable.

Step 4: Consider observer involvement

Does the researcher need to become part of the group to understand the behaviour?

If yes, participant observation may be suitable.

If systematic recording is more important, non-participant observation may be preferable.

Step 5: Consider recording requirements

Ask:

• Can the observer see all relevant behaviour?

• Are several observers needed?

• Is video recording necessary?

• Are the categories clear?

• Will the behaviour occur often enough?

Step 6: Consider ethics

Ask:

• Can informed consent be obtained?

• Is privacy protected?

• Can participants withdraw?

• Could anyone be harmed?

• Is debriefing possible?

• Will recordings be kept confidential?

Step 7: Justify the final choice

A justification must link the method to the scenario.

For example:

A naturalistic, overt, non-participant observation would be appropriate because the researcher wants to record pupils’ normal classroom participation without altering the lesson. Overt observation would allow consent to be obtained, and remaining separate would allow the researcher to count hand raises accurately. However, the pupils may initially alter their behaviour because they know they are being observed, so the observer could attend several lessons to allow habituation.

Applying method selection

Scenario: playground aggression

A psychologist wants to investigate aggressive behaviour during ordinary school break times.

A naturalistic observation may be suitable because:

• The behaviour is studied in its normal setting.

• Real peer groups are present.

• Everyday playground conditions are preserved.

An overt approach may be ethically preferable because children are involved.

A non-participant role may allow the observer to record behaviour using a checklist.

However:

• Awareness may reduce aggression.

• Parental or guardian permission may be needed.

• Behavioural categories must be clear.

• The observer must consider protection from harm.

Scenario: cooperation under limited resources

A researcher wants every pair of participants to encounter exactly the same opportunity to cooperate.

A controlled observation may be appropriate because:

• The same materials can be provided.

• Instructions can be standardised.

• Every pair receives the same amount of time.

• Cooperation can be compared fairly.

An overt, non-participant procedure may allow:

• Informed consent.

• Clear recording.

• Use of cameras.

• Inter-observer reliability checks.

The limitation is that cooperation in an artificial task may not represent everyday relationships.

Scenario: customer responses to queue-jumping

A psychologist wants to study how customers respond when someone moves ahead in a public queue.

A naturalistic field setting may increase realism.

A covert observation may reduce behavioural reactivity, but the researcher must consider:

• Whether the manipulation creates conflict.

• Whether participants may be distressed.

• Whether consent can be waived ethically.

• Whether debriefing is practical.

• Whether the behaviour is recorded anonymously.

A non-participant observer may record responses without becoming part of the queue interaction.

Scenario: culture within a restricted group

A researcher wants to understand the unwritten rules followed by members of a private hobby group.

Participant observation may provide:

• Access to group interactions.

• Direct understanding of routines.

• Insight into group-specific meanings.

An overt participant role may be ethically safer than covert membership.

However:

• Members may alter behaviour.

• The researcher may become biased.

• Participation may interfere with accurate recording.

Writing an effective method justification

A suitable justification might state:

A controlled, overt, non-participant observation would be appropriate. The researcher could provide every group with the same materials and instructions, ensuring that each group has an equal opportunity to display cooperation. Remaining separate would allow the observer to record behaviour accurately using a checklist. An overt procedure would allow informed consent, although awareness of the observation might lead participants to behave more cooperatively than usual.

This response:

• Selects three classifications.

• Links each one to the scenario.

• Identifies a strength.

• Recognises a limitation.

• Includes an ethical consideration.

Weak method justification

I would use a controlled observation because it is reliable.

This is weak because it does not explain:

• What would be controlled.

• Why control is useful for the scenario.

• How reliability would be improved.

• What limitation might remain.

Improved method justification

A controlled observation would allow every pair to receive the same puzzle, instructions and time limit. This standardisation would make differences in cooperation easier to compare and would allow the procedure to be replicated. However, the artificial task may reduce ecological validity.

Overall evaluation

Each form of observation offers a different balance of strengths and limitations.

Naturalistic observations

Strong for:

• Realistic behaviour.

• Everyday context.

• Ecological validity.

• Spontaneous actions.

Limited by:

• Low control.

• Replication difficulties.

• Unpredictable behaviour.

• Recording problems.

Controlled observations

Strong for:

• Standardisation.

• Control.

• Precise recording.

• Replication.

• Consistent opportunities.

Limited by:

• Artificiality.

• Demand characteristics.

• Behavioural reactivity.

• Reduced ecological validity.

Covert observations

Strong for:

• Reduced participant awareness.

• Fewer demand characteristics.

• More natural behaviour.

Limited by:

• Lack of informed consent.

• Privacy concerns.

• Deception.

• Difficulty protecting withdrawal rights.

Overt observations

Strong for:

• Informed consent.

• Ethical transparency.

• Right to withdraw.

• Open use of recording equipment.

Limited by:

• Reactivity.

• Demand characteristics.

• Socially desirable behaviour.

Participant observations

Strong for:

• Detailed understanding.

• Access.

• Insight into group culture.

• Contextual interpretation.

Limited by:

• Observer bias.

• Loss of objectivity.

• Recording difficulties.

• Researcher influence.

Non-participant observations

Strong for:

• Systematic recording.

• Greater objectivity.

• Use of checklists.

• Easier comparison.

Limited by:

• Reduced contextual understanding.

• Restricted access.

• Possible reactivity to an outside observer.

The best method is the one that provides evidence suitable for the research aim while protecting participants and producing data that can be recorded accurately.

Key Words 🔑

Key word	Student-friendly definition	How it may be used in an exam
Naturalistic observation	Observation in the setting where the behaviour would ordinarily occur.	Evaluate realism, ecological validity and lack of control.
Controlled observation	Observation in a situation or setting arranged by the researcher.	Evaluate standardisation, replication and artificiality.
Covert observation	Observation in which people do not know that they are being studied.	Evaluate natural behaviour and ethical concerns.
Overt observation	Observation in which participants know they are being studied.	Evaluate consent and behavioural reactivity.
Participant observation	Observation in which the researcher joins the group or activity.	Evaluate insight, access and observer involvement.
Non-participant observation	Observation in which the researcher remains separate.	Evaluate systematic recording and limited insight.
Ecological validity	The extent to which findings represent behaviour in everyday settings.	Evaluate naturalistic and controlled observations.
Internal validity	The extent to which the researcher’s explanation of the findings is accurate.	Evaluate uncontrolled influences and observer bias.
Behavioural reactivity	A change in behaviour caused by awareness of being observed.	Evaluate overt observations.
Demand characteristics	Cues that lead participants to guess the research aim and alter their behaviour.	Explain reduced validity in overt or controlled observations.
Observer bias	The influence of an observer’s expectations on what is recorded.	Evaluate objectivity and interpretation.
Inter-observer reliability	The level of agreement between observers recording the same behaviour.	Explain how consistency can be assessed.
Behavioural category	A clearly defined observable action recorded by an observer.	Explain how observations can be made more reliable.
Standardisation	Keeping a research procedure consistent.	Explain a strength of controlled observations.
Replication	Repeating a study using the same or similar procedure.	Evaluate controlled and naturalistic observations.
Habituation	Becoming accustomed to the observer so that behaviour returns towards normal.	Suggest how reactivity may be reduced.
Informed consent	Agreement to participate after receiving sufficient information.	Evaluate covert and overt observations.
Privacy	A person’s reasonable expectation that behaviour will not be observed or recorded.	Evaluate observation in public or private settings.
Confidentiality	Protecting participants’ identities and personal information.	Explain how observational data should be handled.
Right to withdraw	The right to leave a study or request removal of data.	Evaluate covert observations.
Protection from harm	The responsibility to avoid unnecessary physical or psychological risk.	Evaluate observational procedures and researcher intervention.
Debriefing	Explaining the study fully after participation.	Suggest how deception may be addressed.
Social desirability	Behaving in a way that creates a favourable impression.	Explain altered behaviour in overt observations.
Generalisability	The extent to which findings apply beyond the observed people and setting.	Evaluate restricted settings and observation periods.

Common Mistakes ⚠️

Mistake: Stating that naturalistic observations always have high ecological validity.

Why this is incorrect:Participants may know they are being observed or the observation period may be unusual.

How to improve:Evaluate the realism of the full procedure rather than relying only on its label.

Mistake: Saying controlled observations remove every extraneous variable.

Why this is incorrect:Participants and situations may still differ in ways that affect behaviour.

How to improve:State that controlled observations usually provide greater control, not complete control.

Mistake: Saying covert observation is automatically the best method because behaviour is natural.

Why this is incorrect:Covert observation may involve serious problems with consent, privacy and withdrawal.

How to improve:Balance behavioural validity against ethical protection.

Mistake: Saying overt observations cannot produce valid data.

Why this is incorrect:Participants may habituate to the observer, and some behaviours may be recorded accurately despite awareness.

How to improve:Describe reactivity as a possible limitation rather than an inevitable result.

Mistake: Treating participant observation as more objective than non-participant observation.

Why this is incorrect:Joining the group may produce emotional involvement and observer bias.

How to improve:Explain that participation may improve insight while reducing objectivity.

Mistake: Saying a non-participant observer has no effect on behaviour.

Why this is incorrect:A visible outside observer may still make participants self-conscious.

How to improve:Distinguish remaining separate from being unnoticed.

Mistake: Listing “high ecological validity” without applying it to the scenario.

Why this is incomplete:Application requires an explanation of how the actual setting represents everyday behaviour.

How to improve:Refer to the specific participants, activity and location.

Mistake: Describing observer bias as participants changing their behaviour.

Why this is incorrect:Observer bias concerns the researcher’s expectations affecting recording or interpretation.

How to improve:Use behavioural reactivity or demand characteristics for changes in participant behaviour.

Mistake: Assuming two observers automatically produce reliable data.

Why this is incorrect:The observers may disagree if behavioural categories are unclear.

How to improve:Compare their recordings and calculate or assess inter-observer agreement.

Mistake: Using vague behavioural categories such as “being friendly”.

Why this is incorrect:Observers may interpret the category differently.

How to improve:Define specific observable actions, such as smiling, offering help or beginning a conversation.

Mistake: Saying observation directly measures thoughts or feelings.

Why this is incorrect:Observation records behaviour from which internal processes may be inferred.

How to improve:Separate the observable action from its possible psychological meaning.

Mistake: Ignoring the right to withdraw in covert research.

Why this is incorrect:Unaware participants cannot exercise an informed right to withdraw during the observation.

How to improve:Consider debriefing and opportunities to remove data where possible.

Mistake: Assuming behaviour in a public place can always be recorded ethically.

Why this is incorrect:People may still have privacy expectations, and the behaviour may be sensitive or identifiable.

How to improve:Consider location, sensitivity, identifiability and the method of recording.

Mistake: Selecting a method without justifying each classification.

Why this is incomplete:The examiner needs to understand why the method suits the particular aim.

How to improve:Explain the choice of setting, awareness and observer role separately.

Mistake: Recommending covert observation only to avoid gaining consent.

Why this is incorrect:Avoiding consent for convenience is not an adequate ethical justification.

How to improve:Explain why awareness would seriously damage the research and how risks would be minimised.

Exam-Style Questions ✍️

Question 1

Which one of the following is a likely strength of a controlled observation?

A. Behaviour must occur spontaneouslyB. The procedure can be standardisedC. Participants are always unaware of the studyD. The observer must join the group

[1 mark]

Question 2

Explain one limitation of a naturalistic observation.

[3 marks]

Question 3

Explain one strength and one ethical limitation of covert observation.

[4 marks]

Question 4

Explain why an overt observation may produce less valid behaviour than a covert observation.

[4 marks]

Question 5

A researcher joins an existing community group and takes part in its activities while observing decision-making.

Explain one strength and one limitation of participant observation in this investigation.

[6 marks]

Question 6

Two observers use the same behavioural categories to record playground aggression. Their records differ considerably.

Explain what this suggests about inter-observer reliability and suggest one way the researchers could improve it.

[4 marks]

Question 7

A psychologist wants to investigate how often pupils contribute during ordinary classroom discussions. The school agrees to the research, and the psychologist plans to sit at the back of each lesson.

Select an appropriate form of observation and justify your choice.

[6 marks]

Question 8

A researcher wants every pair of participants to have the same opportunity to display cooperative behaviour.

Explain why a controlled, overt, non-participant observation might be suitable.

[6 marks]

Question 9

A psychologist secretly records helping behaviour in a public railway station.

Discuss two ethical issues raised by this investigation and explain how one of them might be addressed.

[6 marks]

Question 10

Evaluate different forms of observation.

Refer to practical and ethical issues and the selection of an appropriate method in your answer.

[8 marks]