Experimental methods | AQA A-Level Psychology Revision
Updated: Sep 9
For 7182 specification, first teach in September 2025
AQA A-Level Psychology | Free Revision Notes
Estimated study time: 45 minutes
These Experimental methods A-Level Psychology revision notes explain how psychologists manipulate an independent variable, measure a dependent variable and control possible alternative explanations. You will compare laboratory and field experiments, including their levels of control, ecological validity, replicability and susceptibility to demand characteristics. This lesson introduces the experimental method before naturally occurring and pre-existing variables and the later study of independent, dependent and extraneous variables.
The AQA specification requires knowledge of laboratory and field experiments, together with awareness of their strengths and limitations.
Learning Objectives 🎯
By the end of this revision page, you should be able to:
Define the experimental method.
Explain the roles of the independent and dependent variables.
Explain why extraneous variables are controlled.
Distinguish laboratory and field experiments.
Identify the type of experiment used in an unfamiliar investigation.
Evaluate the strengths and limitations of laboratory experiments.
Evaluate the strengths and limitations of field experiments.
Select an appropriate experimental setting for a research question.
Revision Notes đź“š
Experimental methods A-Level Psychology revision overview
The experimental method is a research method used to investigate whether one variable affects another.
The researcher:
Manipulates an independent variable.
Measures a dependent variable.
Attempts to control extraneous variables.
Compares the results from different conditions.
The basic experimental question is:
Does changing the independent variable produce a change in the dependent variable?
For example, a psychologist might investigate whether background noise affects memory performance.
Independent variable: the level of background noise.
Dependent variable: the number of words recalled.
Experimental condition: participants learn words with background noise.
Control condition: participants learn words in silence.
If the groups or conditions differ in recall, the researcher considers whether the manipulated noise level produced the difference.
What makes a study an experiment?
A study is an experiment when the researcher manipulates an independent variable and measures its effect on a dependent variable.
An experiment therefore contains:
At least two conditions or levels of the independent variable.
A measurable dependent variable.
A comparison between conditions.
Some attempt to control other influences.
The presence of numbers, a questionnaire or a group of participants does not automatically make a study an experiment.
For example:
Asking people about their stress without manipulating anything is not an experiment.
Observing playground behaviour without introducing a change is not an experiment.
Measuring the relationship between sleep and examination results is not necessarily an experiment.
Changing the amount of sleep participants receive and measuring memory would be an experiment.
The difference between experiments and studies of relationships is developed in relationships between co-variables.
The independent variable
The independent variable, abbreviated to IV, is the variable manipulated by the researcher.
It is the factor that differs between the conditions.
For example, a researcher investigates whether the presence of a mobile phone affects concentration.
The IV could be:
Mobile phone visible on the desk.
Mobile phone stored outside the room.
The researcher creates these conditions to investigate their possible effect.
A clear IV should identify exactly how the conditions differ.
Levels of the independent variable
The different conditions of the IV are called its levels.
For example:
Investigation | Level 1 of the IV | Level 2 of the IV |
Effect of noise on memory | Silence | Recorded background noise |
Effect of praise on persistence | Praise given | No praise given |
Effect of lighting on reading speed | Bright lighting | Dim lighting |
Effect of phone presence on concentration | Phone visible | Phone absent |
An IV may have more than two levels, but at least two are needed for a comparison.
The dependent variable
The dependent variable, abbreviated to DV, is the variable measured by the researcher.
It shows the outcome in each condition.
In the mobile-phone investigation, the DV might be:
The number of concentration questions answered correctly.
The time taken to complete a task.
The number of errors made.
The DV should be measured in the same way in every condition.
This allows a fair comparison.
Independent and dependent variables compared
Independent variable | Dependent variable |
Manipulated by the researcher | Measured by the researcher |
Differs between conditions | Records the outcome |
Proposed cause | Possible effect |
Must be clearly defined | Must be measured consistently |
Example: noise or silence | Example: number of words recalled |
A useful memory rule is:
IV:Â the variable the researcher changes.
DV:Â the data the researcher records.
Operationalisation
Operationalisation means defining a variable in a clear and measurable way.
A vague IV might be:
Whether the room is noisy.
An operationalised IV might be:
Participants complete the task either in silence or while a recording of classroom noise is played at the same volume.
A vague DV might be:
Memory ability.
An operationalised DV might be:
The number of words correctly recalled from a list of 20 after two minutes.
Operationalisation matters because it allows researchers to:
Conduct the study consistently.
Replicate the procedure.
Decide exactly what data to record.
Judge whether the variables measure the intended concepts.
The process is covered in greater detail in identifying and defining variables.
Experimental conditions
An experimental investigation contains different conditions.
An experimental condition is a condition in which the feature of interest is present or altered.
A control condition provides a comparison.
For example:
Condition | Procedure |
Experimental condition | Participants study with notifications appearing on a phone |
Control condition | Participants study without phone notifications |
The control condition helps the researcher judge whether the manipulated feature affected the DV.
A control condition is not necessarily a condition in which “nothing happens”. Participants may complete the same task, but without the key feature being investigated.
Cause and effect
Experiments are designed to investigate cause-and-effect relationships.
The intended reasoning is:
The researcher changes the IV.
Other important factors are controlled.
The DV is measured.
A difference is found between the conditions.
The researcher concludes that the IV may have caused the difference.
The confidence of this conclusion depends on how effectively other explanations were controlled.
A difference between conditions does not automatically prove causation.
Researchers must consider whether:
The conditions differed in another important way.
The participants differed before the study.
Demand characteristics affected behaviour.
Investigator effects influenced the results.
The DV was measured validly.
Extraneous variables
An extraneous variable is any variable other than the IV that could affect the DV.
Consider an experiment investigating whether music affects reading comprehension.
Possible extraneous variables include:
The difficulty of the reading passages.
Participants’ reading ability.
The volume of the music.
Time of day.
Lighting.
Interruptions.
The amount of sleep participants had received.
The instructions given by the researcher.
If these factors vary between conditions, they could provide alternative explanations for the results.
Confounding variables
A confounding variable is an extraneous variable that varies systematically with the IV and could therefore explain the results.
Suppose:
The music condition completes a difficult reading passage.
The silent condition completes an easy reading passage.
Passage difficulty is now confounded with the IV.
If the silent condition performs better, the researcher cannot tell whether this was caused by:
The absence of music.
The easier passage.
A well-designed experiment attempts to prevent extraneous variables from becoming confounding variables.
Controlling variables
Control involves keeping possible extraneous influences constant or otherwise reducing their effect.
Control procedures may include:
Giving all participants the same instructions.
Using the same equipment.
Allowing the same amount of time.
Using the same testing room.
Presenting material in the same format.
Randomly allocating participants to conditions.
Counterbalancing the order of tasks.
Using a control group.
Randomising stimulus order.
These procedures are studied in detail in randomisation, standardisation and control groups.
Standardisation
Standardisation means keeping the procedure consistent for all participants.
Researchers may use:
Written instructions.
A fixed time limit.
The same materials.
A scripted introduction.
The same sequence of events.
A standard method of recording results.
Standardisation helps ensure that differences in the DV are not caused by participants receiving different procedures.
It also improves replicability.
Replication
Replication means repeating a study using the same or a closely similar procedure.
A procedure is easier to replicate when:
The IV is operationalised clearly.
The DV is measured precisely.
Instructions are standardised.
Materials are described.
Extraneous variables are controlled.
The stages of the investigation are recorded.
If repeated studies produce similar findings, confidence in the reliability of the result increases.
The experimental setting
Experiments may be classified according to the setting in which they are conducted.
The two forms covered in this lesson are:
Laboratory experiments
Field experiments
Both involve manipulation of an IV.
The main difference concerns where the study takes place and how much control the researcher has over the environment.
Laboratory experiments
A laboratory experiment is conducted in a highly controlled environment.
The setting may be:
A purpose-built laboratory.
A controlled classroom.
A quiet testing room.
Another environment arranged by the researcher.
The defining feature is not scientific equipment or a white coat.
The defining feature is the researcher’s high level of control over the setting and procedure.
Features of a laboratory experiment
A laboratory experiment usually involves:
Researcher manipulation of the IV.
Deliberate measurement of the DV.
A controlled or artificial setting.
Standardised instructions and materials.
Control over possible extraneous variables.
Participants who are usually aware that they are taking part in research.
For example:
Participants attend a controlled room and complete a memory test either in silence or while recorded speech is played through headphones.
This is a laboratory experiment because:
The IV is manipulated.
The DV is measured.
The setting is controlled.
The researcher organises the conditions.
Laboratory does not always mean a physical laboratory
A study can be a laboratory experiment even if it takes place in an ordinary room.
For example, a psychologist uses a classroom after school and:
Controls who enters.
Sets the lighting.
Provides identical laptops.
Uses scripted instructions.
Manipulates the task.
Measures performance.
This can still be classified as a laboratory experiment because the environment has been arranged and controlled by the researcher.
Field experiments
A field experiment takes place in a natural setting where the behaviour being studied would ordinarily occur.
The researcher still manipulates the IV and measures the DV.
Possible field settings include:
A school.
A workplace.
A shop.
A street.
Public transport.
A café.
A sports setting.
The setting is part of the participant’s ordinary environment rather than a room created primarily for research.
Features of a field experiment
A field experiment usually involves:
Researcher manipulation of the IV.
Measurement of the DV.
A natural setting.
Behaviour occurring in a familiar context.
Less control over environmental variables.
Participants who may or may not know the study is taking place.
For example:
A psychologist investigates whether signs encouraging stair use affect behaviour in a railway station. The sign is displayed on some days and removed on others. The number of people using the stairs is recorded.
This is a field experiment because:
The researcher manipulates the presence of the sign.
Stair use is measured.
Behaviour occurs in a natural public setting.
A field experiment still contains a manipulated IV
The natural setting does not make a field experiment a natural experiment.
In a field experiment, the researcher introduces or changes the IV.
For example:
Changing the wording of a notice.
Introducing background music.
Altering how staff greet customers.
Placing an object in a particular location.
Giving one group a particular instruction.
In a natural experiment, the IV arises without being directly manipulated by the researcher.
This distinction is developed in Natural and quasi-experiments.
Laboratory and field experiments compared
Feature | Laboratory experiment | Field experiment |
Setting | Controlled or artificial | Natural, everyday setting |
IV | Manipulated by the researcher | Manipulated by the researcher |
DV | Measured by the researcher | Measured by the researcher |
Control over extraneous variables | Usually high | Usually lower |
Standardisation | Usually easier | Often more difficult |
Replication | Usually easier | Often more difficult |
Ecological validity | May be lower | May be higher |
Demand characteristics | May be more likely | May be reduced if participants are unaware |
Ethical control | Consent and briefing are often easier | Consent, privacy or deception may be more difficult |
Confidence in causal conclusions | Often relatively high | May be reduced by uncontrolled variables |
How to identify the type of experiment
Ask the following questions.
Did the researcher manipulate the IV?
Yes: the study may be a laboratory or field experiment.
No: it is not one of these two experimental types.
Was the environment arranged and controlled by the researcher?
Yes: it is likely to be a laboratory experiment.
No: consider whether it occurred naturally.
Did the behaviour take place in its normal everyday setting?
Yes: it is likely to be a field experiment.
Were participants completing an artificial task in a controlled room?
Yes: it is likely to be a laboratory experiment.
Worked identification example
A researcher asks participants to attend a university testing room. Half complete a problem-solving task under bright lighting and half complete it under dim lighting. All other aspects of the room are kept constant.
This is a laboratory experiment because:
The researcher manipulates lighting.
Problem-solving performance is measured.
Participants attend a controlled setting.
Other features are standardised.
Worked field example
A researcher changes the level of music in a real café on different days and records how long customers remain.
This is a field experiment because:
Music level is manipulated.
Time spent in the café is measured.
The investigation occurs in a natural setting.
Customers are engaging in ordinary behaviour.
Possible extraneous variables might include:
Day of the week.
Weather.
Number of customers.
Staffing.
Time of day.
These would need to be controlled or considered.
Strength of laboratory experiments: control
A major strength of laboratory experiments is the high level of control over extraneous variables.
Researchers can often keep constant:
The room.
Materials.
Timing.
Instructions.
Equipment.
Task difficulty.
Noise level.
Lighting.
This reduces possible alternative explanations for differences in the DV.
If the conditions differ only in the IV, the researcher can be more confident that the IV caused the observed effect.
Internal validity
Internal validity concerns whether a study demonstrates that the IV caused the change in the DV.
Laboratory experiments may have high internal validity because:
The IV is deliberately manipulated.
Extraneous variables can be controlled.
Procedures can be standardised.
A control condition can be used.
However, laboratory location alone does not guarantee internal validity.
Poor operationalisation, demand characteristics or investigator effects may still reduce the accuracy of the conclusion.
Strength of laboratory experiments: replication
Laboratory procedures are usually easier to replicate.
The researcher can provide clear details about:
Materials.
Instructions.
Timings.
Equipment.
Conditions.
Measurements.
Another psychologist can reproduce these features closely.
Replication allows researchers to assess whether the result is consistent rather than a chance finding from one sample or occasion.
This contributes to consistency across psychological investigations.
Strength of laboratory experiments: precise measurement
The controlled environment may allow the DV to be measured precisely.
Researchers may use:
Accuracy scores.
Reaction times.
Memory totals.
Number of errors.
Behavioural frequencies.
Standardised rating scales.
Precise measurement can make differences between conditions easier to identify and analyse.
However, a precise measure is not automatically a valid measure.
For example, reaction time may be measured accurately but may not represent the complete psychological process the researcher claims to investigate.
Strength of laboratory experiments: standardised procedure
Standardisation means participants experience a consistent procedure.
This reduces the possibility that:
One group receives clearer instructions.
Some participants receive more time.
Different equipment affects performance.
The researcher provides unequal help.
Materials differ unexpectedly.
A standardised procedure therefore supports both control and replication.
Limitation of laboratory experiments: artificiality
Laboratory experiments may use tasks or settings that differ from everyday life.
Participants may be asked to:
Memorise unrelated word lists.
Respond to artificial images.
Complete repetitive computer tasks.
Sit alone in a controlled room.
Make decisions with no meaningful consequences.
Behaviour in this setting may differ from behaviour in a natural situation.
This can reduce ecological validity.
Ecological validity
Ecological validity is the extent to which findings represent behaviour in everyday settings.
A laboratory memory task may have low ecological validity if:
The material is unrealistic.
The setting is unfamiliar.
Participants know they are being tested.
The task has little personal meaning.
The behaviour has no real-world consequence.
However, not every laboratory study has low ecological validity.
A controlled task may still measure a process that also occurs outside the laboratory.
Ecological validity should be evaluated in relation to the particular procedure.
Limitation of laboratory experiments: demand characteristics
Demand characteristics are cues that lead participants to guess the aim or expected behaviour in a study.
In a laboratory experiment, participants often know:
That they are being studied.
That there are different conditions.
That the researcher expects some form of effect.
They may then:
Try to support the hypothesis.
Deliberately behave differently.
Try to appear helpful.
Resist what they think is expected.
Change their behaviour because they feel observed.
The resulting DV may reflect participants’ interpretation of the study rather than the IV alone.
Demand characteristics are developed further in Demand characteristics and investigator effects.
Limitation of laboratory experiments: investigator effects
Investigator effects occur when the researcher’s behaviour or expectations influence participants or the recording of results.
For example, a researcher might:
Use a warmer tone in one condition.
Give more encouragement to one group.
Explain instructions differently.
Record ambiguous responses in a biased way.
Unintentionally communicate the expected result.
Standardised instructions and objective measurement can reduce these effects.
Limitation of laboratory experiments: participant awareness
Knowing that they are participating in research can alter behaviour.
Participants may:
Concentrate more than usual.
Become anxious.
Try harder.
Become self-conscious.
Respond in a socially desirable way.
Search for the purpose of the study.
This is sometimes called reduced mundane realism, because the research experience differs from ordinary life.
Participant awareness contributes to the concern that laboratory findings may not represent natural behaviour.
Limitation of laboratory experiments: generalisation
A highly controlled investigation may use a restricted sample, such as students or volunteers who can attend the laboratory.
The findings may not generalise to:
Other age groups.
People in different environments.
People who would not volunteer.
Everyday situations.
Different cultural contexts.
This is primarily a sampling issue rather than an unavoidable feature of all laboratory experiments.
The influence of sample selection is considered in Populations and samples.
Strength of field experiments: natural behaviour
A major strength of field experiments is that participants behave in a setting where the behaviour would normally occur.
For example:
Customers behave in a real shop.
Pupils work in a real classroom.
Commuters make choices in a real station.
Employees complete tasks in a real workplace.
The behaviour may therefore be more natural than behaviour produced by an artificial laboratory task.
Strength of field experiments: ecological validity
Field experiments may have higher ecological validity because:
The setting is realistic.
The task is part of ordinary life.
Behaviour has genuine consequences.
Participants respond to natural environmental demands.
The context in which behaviour normally occurs is preserved.
This may make it more appropriate to generalise the findings to similar real-world situations.
However, field experiments do not automatically have high ecological validity.
Participants might still complete an artificial task within the natural environment.
Strength of field experiments: reduced demand characteristics
Participants in a field experiment may be unaware that research is taking place.
If they do not know they are participants, they are less likely to:
Guess the aim.
Search for clues.
Change their behaviour deliberately.
Try to please the investigator.
Behave as they believe research participants should.
This may increase the validity of the behaviour recorded.
Demand characteristics are reduced rather than necessarily eliminated.
The manipulation itself might still be noticed.
Strength of field experiments: practical applications
Field experiments investigate behaviour in realistic settings.
The findings may therefore be useful when deciding how to change:
School procedures.
Workplace environments.
Public information.
Customer experiences.
Health messages.
Transport behaviour.
The researcher can observe the effect of an intervention in the type of context where it might later be used.
However, practical usefulness depends on whether the findings are valid and generalisable.
Limitation of field experiments: reduced control
Field settings contain many variables that are difficult to control.
For example, a study in a school may be affected by:
Different teachers.
Noise from nearby rooms.
Time of day.
Pupil attendance.
Previous lessons.
Classroom temperature.
Unexpected interruptions.
These variables may affect the DV.
If they differ systematically between conditions, they may become confounding variables.
Limitation of field experiments: weaker causal conclusions
Field experiments still manipulate an IV, so they can investigate possible causal effects.
However, reduced control means that the researcher may be less certain that the IV caused the results.
Suppose a shop plays slow music on Monday and fast music on Saturday.
Any difference in customer behaviour could be caused by:
The music.
The day of the week.
The number of customers.
Different staff.
Different types of shopper.
The setting therefore increases realism but may reduce internal validity.
Limitation of field experiments: difficulty replicating
A field setting cannot always be recreated exactly.
Possible changes include:
Weather.
Noise.
Number of people present.
Staff behaviour.
Public events.
Time of day.
Unexpected interruptions.
Another researcher may repeat the basic procedure but encounter different conditions.
This makes exact replication more difficult and may reduce reliability.
Limitation of field experiments: ethical issues
Some field experiments involve participants who do not know they are being studied.
This may create ethical concerns involving:
Informed consent.
Deception.
Privacy.
Confidentiality.
The right to withdraw.
Protection from harm.
For example, a person cannot exercise an informed right to withdraw during the procedure if they do not know that research is taking place.
Researchers must balance:
The value of natural behaviour.
The need to avoid demand characteristics.
Participants’ ethical rights.
These issues are developed in Ethics in psychological research.
Privacy in field experiments
Natural settings may be public or private.
Observing behaviour in a public location does not automatically remove every ethical concern.
Researchers should consider:
Whether people reasonably expect to be observed.
Whether sensitive behaviour is being recorded.
Whether individuals could be identified.
Whether recording devices are used.
Whether the manipulation could cause distress.
How data will be stored and reported.
Results should normally be recorded and presented without identifying individual participants.
Limitation of field experiments: practical difficulties
Field experiments may be difficult to organise.
Researchers may need:
Permission from an organisation.
Cooperation from staff.
Access to a real setting.
Procedures for recording behaviour unobtrusively.
Arrangements to keep conditions consistent.
Plans for unexpected events.
The researcher may also be unable to manipulate a variable in a natural setting because doing so would be unsafe, disruptive or unethical.
The control-validity trade-off
Laboratory and field experiments often involve a trade-off.
Greater laboratory control | Greater field realism |
Stronger control of extraneous variables | Behaviour occurs in a natural setting |
Easier standardisation | Greater ecological validity may result |
Easier replication | Demand characteristics may be reduced |
Stronger causal conclusions | Findings may be more applicable to everyday behaviour |
Greater artificiality | More uncontrolled influences |
Participant awareness may increase | Ethical and practical control may be harder |
Neither method is always superior.
The most appropriate method depends on the research question.
Selecting a laboratory experiment
A laboratory experiment may be appropriate when the researcher needs:
Precise manipulation of the IV.
Precise measurement of the DV.
Strong control over extraneous variables.
Standardised testing.
An easily replicated procedure.
A clear comparison between conditions.
For example, a laboratory setting may be useful for measuring reaction time because equipment, instructions and stimulus presentation can be controlled closely.
Selecting a field experiment
A field experiment may be appropriate when the researcher needs:
Natural behaviour.
A realistic context.
Reduced participant awareness.
Evidence about an intervention in an everyday setting.
A task that would be distorted by bringing participants into a laboratory.
For example, investigating whether a sign changes littering behaviour may be more valid in a real public space than in a simulated room.
The research question determines the method
Consider two different questions.
Question 1
Does the volume of background speech affect immediate word recall under controlled conditions?
A laboratory experiment may be suitable because precise control and measurement are important.
Question 2
Does background music affect how long customers remain in a café?
A field experiment may be suitable because the behaviour needs to occur in a genuine café.
The choice should be justified by linking the method with the requirements of the investigation.
Experimental method and experimental design
The type of experiment and the experimental design are different decisions.
Type of experiment
This concerns how and where the IV is created.
Examples:
Laboratory.
Field.
Natural.
Quasi.
Experimental design
This concerns how participants are allocated to conditions.
Examples:
Independent groups.
Repeated measures.
Matched pairs.
A laboratory experiment could use any of the three designs.
A field experiment could also use different designs depending on the investigation.
These allocation methods are examined in Experimental designs.
Experimental method and sampling
The type of experiment does not tell you how participants were selected.
A laboratory experiment might use:
Volunteer sampling.
Opportunity sampling.
Random sampling.
Systematic sampling.
Stratified sampling.
A field experiment may include people already present in the setting, which may resemble an opportunity sample.
Sampling affects representativeness and generalisation, while the experimental type concerns the setting and manipulation.
Experimental method and data type
Experiments may produce either quantitative or qualitative data.
Most experimental DVs are quantitative, such as:
Scores.
Times.
Frequencies.
Ratings.
However, an experiment could also collect qualitative responses, such as participants’ written descriptions after each condition.
The presence of numerical data does not define a study as an experiment.
Manipulation of the IV is the essential feature.
Applying laboratory evaluation
Consider this investigation:
Participants complete a computer attention task in a controlled room. In one condition, phone notifications appear on the screen. In the other condition, no notifications appear. The researcher records the number of targets detected.
Possible strengths include:
Notification conditions are controlled.
The same computer task can be used.
Targets detected provide a precise DV.
The procedure can be replicated.
A causal effect can be investigated.
Possible limitations include:
The task may not represent everyday concentration.
Participants may guess the aim.
Being observed may alter effort.
Phone notifications on a research screen may differ from personally meaningful messages.
Applying field evaluation
Consider another investigation:
A researcher changes the wording of posters placed beside recycling bins and records the percentage of waste placed in the correct bin.
Possible strengths include:
Recycling occurs in a natural setting.
Behaviour is genuine rather than hypothetical.
Participants may be unaware of the study.
Findings may be useful for designing public messages.
Possible limitations include:
Different people may use the area on different days.
Bin use may be affected by how full the bins are.
The amount and type of waste may vary.
Weather or nearby events may affect behaviour.
Unaware participants cannot provide informed consent in the usual way.
A method for evaluating an experiment
Use the following structure.
1. Identify a feature
For example, the laboratory setting provides high control.
2. Apply it to the investigation
Identify the specific variable that can be controlled.
3. Explain the consequence
State how this affects validity, reliability or causal conclusions.
4. Add a qualification
Explain why the feature does not guarantee a perfect study.
For example:
A strength of the laboratory experiment is that background noise can be kept constant in both conditions. This reduces the possibility that noise rather than phone notifications affected attention, increasing internal validity. However, participants may guess that the study concerns distraction and alter their effort, so demand characteristics could still affect the results.
Writing an effective comparison
A strong comparison might state:
Both laboratory and field experiments involve manipulation of an independent variable and measurement of a dependent variable. Laboratory experiments take place in controlled settings, allowing extraneous variables to be managed and procedures to be replicated. Field experiments take place in natural settings, so behaviour may have greater ecological validity and demand characteristics may be reduced. However, field researchers have less control over environmental variables, making it harder to establish that the independent variable caused the result.
This answer:
Gives a clear similarity.
Distinguishes the settings.
Compares control.
Compares validity.
Links each method to causal inference.
Writing an effective evaluation paragraph
A developed evaluation paragraph could state:
Laboratory experiments usually allow researchers to control extraneous variables. For example, task difficulty, instructions and testing time can be kept constant across conditions. This increases internal validity because a difference in the dependent variable is more likely to have resulted from the manipulated independent variable. However, the high level of control may create an artificial situation, so participants’ behaviour may not represent how they would respond in everyday life.
The paragraph moves through:
Methodological feature.
Applied explanation.
Effect on research quality.
Counterargument.
Overall comparison
Laboratory experiments are particularly strong when the priority is:
Control.
Standardisation.
Replication.
Precise measurement.
Confidence in causal conclusions.
Field experiments are particularly strong when the priority is:
Natural behaviour.
Realistic settings.
Ecological validity.
Reduced participant reactivity.
Direct application to everyday contexts.
The central judgement is:
Laboratory experiments often gain control at the cost of realism, while field experiments often gain realism at the cost of control.
This is a general comparison rather than an absolute rule. The quality of any investigation depends on its particular procedure.
Common Mistakes ⚠️
Mistake: Saying that any investigation conducted by a psychologist is an experiment.
Why this is incorrect:An experiment requires the researcher to manipulate an independent variable and measure a dependent variable.
How to improve:Identify the manipulated conditions before classifying a study as experimental.
Mistake: Confusing the independent and dependent variables.
Why this is incorrect:The independent variable is manipulated, whereas the dependent variable is measured.
How to improve:Ask: “What did the researcher change?” and “What data did the researcher record?”
Mistake: Describing the aim instead of operationalising the variables.
Why this is incorrect:Terms such as “stress”, “memory” or “noise” do not state exactly how the variables were created or measured.
How to improve:State the precise conditions and the numerical or observable DV.
Mistake: Calling every other variable a confounding variable.
Why this is incorrect:An extraneous variable becomes confounding only when it varies systematically with the IV and could explain the results.
How to improve:Explain how the variable differed between conditions and affected the DV.
Mistake: Saying that a laboratory experiment must take place in a scientific laboratory.
Why this is incorrect:A laboratory experiment is defined by its controlled environment, not by specialist equipment or the room’s name.
How to improve:Focus on how far the researcher arranged and controlled the setting.
Mistake: Saying that a field experiment has no independent variable.
Why this is incorrect:The researcher manipulates the IV in both laboratory and field experiments.
How to improve:Identify what the researcher introduced or changed in the natural setting.
Mistake: Confusing a field experiment with a natural experiment.
Why this is incorrect:In a field experiment, the researcher manipulates the IV. In a natural experiment, the change occurs without direct researcher manipulation.
How to improve:Ask who created the difference between conditions.
Mistake: Saying field experiments have no control at all.
Why this is incorrect:Researchers can standardise parts of a field procedure, although they usually have less control than in a laboratory.
How to improve:Describe control as relatively lower rather than completely absent.
Mistake: Saying laboratory experiments always have low ecological validity.
Why this is incorrect:Ecological validity depends on the realism and relevance of the particular task and setting.
How to improve:Apply the evaluation to what participants actually did.
Mistake: Saying field experiments always have high ecological validity.
Why this is incorrect:A study may take place in a natural setting while still requiring an artificial task.
How to improve:Explain whether the behaviour and context genuinely represent everyday life.
Mistake: Saying demand characteristics are impossible in field experiments.
Why this is incorrect:Participants may notice the manipulation or realise they are being observed.
How to improve:State that demand characteristics may be reduced, particularly when participants are unaware.
Mistake: Listing “high control” without explaining why it matters.
Why this is incomplete:Evaluation requires a link between the feature and the quality of the conclusions.
How to improve:Explain that control reduces alternative explanations and can increase internal validity.
Mistake: Saying replication proves that the original conclusion is correct.
Why this is incorrect:Replication assesses consistency, but a consistently repeated study could still use an invalid measure.
How to improve:Separate reliability from validity.
Mistake: Treating laboratory and field experiments as experimental designs.
Why this is incorrect:They are types of experiment. Repeated measures, independent groups and matched pairs are experimental designs.
How to improve:Separate the research setting from the allocation of participants to conditions.
Mistake: Giving generic evaluation that is not linked to the scenario.
Why this loses application marks:The examiner needs to see how the methodological issue affects the particular investigation.
How to improve:Name the exact extraneous variable, demand characteristic or artificial feature involved.
Exam-Style Questions ✍️
Question 1
Which one of the following is a defining feature of the experimental method?
A. Measuring two co-variables without manipulating either
B. Manipulating an independent variable and measuring a dependent variable
C. Observing behaviour without introducing any change
D. Collecting only qualitative data
[1 mark]
Question 2
Define the term dependent variable.
[2 marks]
Question 3
A psychologist investigates whether background music affects concentration.
Participants complete a proofreading task either in silence or while instrumental music is played. The psychologist records the number of errors correctly identified.
Identify:
a)Â The independent variable.
[1 mark]
b)Â The dependent variable.
[1 mark]
Question 4
Explain why an extraneous variable may reduce the internal validity of an experiment.
[3 marks]
Question 5
A researcher invites participants into a controlled room. They complete a memory task either under bright lighting or dim lighting. The researcher uses the same word list and time limit in both conditions.
Explain why this is a laboratory experiment.
[4 marks]
Question 6
A psychologist places different messages beside recycling bins in a school canteen and records the amount of waste placed in the correct bin.
Explain why this is a field experiment.
[4 marks]
Question 7
Explain one strength and one limitation of laboratory experiments.
[6 marks]
Question 8
Explain one strength and one limitation of field experiments.
[6 marks]
Question 9
A researcher investigates whether background music affects how long customers remain in a café. Slow music is played on weekday mornings and fast music is played on Saturday afternoons.
Identify one possible confounding variable and explain how it could affect the results.
[4 marks]
Question 10
Compare laboratory and field experiments.
Refer to control, ecological validity and demand characteristics in your answer.
[8 marks]
Comments