RESEARCH ISSUES: VARIABLES
Extraneous variables
• Any variable other than the IV that changes the DV. These are unwanted and removed by the researcher
at the beginning.
• They are nuisance variables however they don’t cause a significant change in the findings (only affect 1 or
2 participants); they just make it harder to detect a result.
• There are 5 main types of extraneous variables
→ Situational
→ Participant
→ Confounding
→ Demand characteristics
→ Investigator effects
Situational extraneous variable
• Any features of the experimental situation that may affect the DV.
Participant extraneous variable
• Any individual differences between participants which may affect the DV.
Confounding variable
• Another variable that changes with the IV (becomes a second IV so we are unable to know what
caused the results).
→E.g., group with music listened to the radio whilst revising, a news alert interrupted the music
which could have distracted the participants and affected the performance of memory.
• This affects the whole group so ruins the research.
Demand characteristics
• When the participants guess or work out the aim of the research and change their behaviour
accordingly.
• This causes unnatural behaviour and the participant reactivity whereby they under or over perform to
appease or sabotage the researcher’s aim.
Investigator effects
• Researcher may influence the behaviour of the participants (unintentionally) and cause participant
reactivity.
• E.g., treatment of certain participants, selection, instructions etc.
Controlling Extraneous Variables
• If extraneous variables are left uncontrolled, they can lead to inaccurate conclusions about the
relationship between independent and dependent variables which is why we need to control them.
Standardise instructions: Reduces demand characteristics.
Randomisation: Reduces researcher bias.
Counter balancing: Reduces impact of order effects.
→ Test 2 groups on the same thing and 2 different situations then swap situations – reduces impact of order
effects as it is happening equally.
Page | 13
, TYPES OF EXPERIMENTS – EXPERIMENTAL METHODS
• All experiments involve a change in the independent variable and measure how this effects the dependent variable
however, how the independent variable changes vary depending on the type of experiment.
There are 4 main types of experiments:
Laboratory
• Highly controlled conditions – doesn’t have to be in a science lab (as a lab is an environment that the
researcher can manipulate).
• The researcher manipulates the IV so there are very few EV’s that would impact the results.
A03 – Evaluation
Strengths Limitations
➢ High control over confounding and extraneous variables. ➢ Lack ecological validity and mundane realism.
➢ High replicability – uses a standardised procedure. ➢ Lacks generalisability due to unnatural responses.
➢ Researchers can establish cause and effect relationships ➢ Higher chance of demand characteristics.
between IV and DV due to high internal validity.
➢ High reliability.
Field
• Conducting the experiment in a natural setting e.g., school, workplace etc.
• The participants may not be aware of the research.
• IV manipulated by the researcher.
A03 – Evaluation
Strengths Limitations
➢ High ecological validity – easily generalised beyond research ➢ Lack control over possible EV’s and CV’s that may influence
setting. measurement of the DV.
➢ External validity. ➢ Impossible to randomly assign participants – reduces
➢ Mundane realism – likely that the tasks in the experiment are internal validity.
usually carried out by the participant. ➢ Ethical issues – lack of consent if participants are unaware
➢ Participants unaware – less chance of demand characteristics they are being studied.
being shown.
Natural
• IV can’t be manipulated, rather happen naturally so aren’t researcher controlled.
• They are unplanned in daily life and take place in naturally occurring environments – the researchers
choose an already changing IV e.g., the effect of children being isolated from parents on development.
• The DV is measured and the EV’s and CV’s are not controlled.
A03 – Evaluation
Strengths Limitations
➢ High ecological validity. ➢ Lack of control over variables.
➢ Ethical – use of already changing IV’s. ➢ EV’s, due to no control, can’t show cause and effect
➢ High external validity and no chance of demand relationship.
characteristics. ➢ Low replicability – rare events, specific to people thus can’t
be re-tested for reliability.
Quasi
• Participants cannot be randomly assigned between levels of IV; often the level of IV is an innate
characteristic.
• “Hybrid” of a lab & natural experiment – controlled environments however naturally occurring IV.
• An experiment where the IV just exists so participants can’t be randomised, however can be
grouped e.g., study of gender, age etc.
• DV measured and other variables kept constant – the researcher can choose the IV regarding
participants but can’t change it.
A03 – Evaluation
Strengths Limitations
➢ High replicability – controlled conditions. ➢ Lacks randomisation so the causal relationships are not
➢ Only experiment that studies the effects of pre-existing demonstrated.
characteristics of people – allows for comparisons to be ➢ The confounding variables are not considered.
made between different types of people.
Page | 14
Extraneous variables
• Any variable other than the IV that changes the DV. These are unwanted and removed by the researcher
at the beginning.
• They are nuisance variables however they don’t cause a significant change in the findings (only affect 1 or
2 participants); they just make it harder to detect a result.
• There are 5 main types of extraneous variables
→ Situational
→ Participant
→ Confounding
→ Demand characteristics
→ Investigator effects
Situational extraneous variable
• Any features of the experimental situation that may affect the DV.
Participant extraneous variable
• Any individual differences between participants which may affect the DV.
Confounding variable
• Another variable that changes with the IV (becomes a second IV so we are unable to know what
caused the results).
→E.g., group with music listened to the radio whilst revising, a news alert interrupted the music
which could have distracted the participants and affected the performance of memory.
• This affects the whole group so ruins the research.
Demand characteristics
• When the participants guess or work out the aim of the research and change their behaviour
accordingly.
• This causes unnatural behaviour and the participant reactivity whereby they under or over perform to
appease or sabotage the researcher’s aim.
Investigator effects
• Researcher may influence the behaviour of the participants (unintentionally) and cause participant
reactivity.
• E.g., treatment of certain participants, selection, instructions etc.
Controlling Extraneous Variables
• If extraneous variables are left uncontrolled, they can lead to inaccurate conclusions about the
relationship between independent and dependent variables which is why we need to control them.
Standardise instructions: Reduces demand characteristics.
Randomisation: Reduces researcher bias.
Counter balancing: Reduces impact of order effects.
→ Test 2 groups on the same thing and 2 different situations then swap situations – reduces impact of order
effects as it is happening equally.
Page | 13
, TYPES OF EXPERIMENTS – EXPERIMENTAL METHODS
• All experiments involve a change in the independent variable and measure how this effects the dependent variable
however, how the independent variable changes vary depending on the type of experiment.
There are 4 main types of experiments:
Laboratory
• Highly controlled conditions – doesn’t have to be in a science lab (as a lab is an environment that the
researcher can manipulate).
• The researcher manipulates the IV so there are very few EV’s that would impact the results.
A03 – Evaluation
Strengths Limitations
➢ High control over confounding and extraneous variables. ➢ Lack ecological validity and mundane realism.
➢ High replicability – uses a standardised procedure. ➢ Lacks generalisability due to unnatural responses.
➢ Researchers can establish cause and effect relationships ➢ Higher chance of demand characteristics.
between IV and DV due to high internal validity.
➢ High reliability.
Field
• Conducting the experiment in a natural setting e.g., school, workplace etc.
• The participants may not be aware of the research.
• IV manipulated by the researcher.
A03 – Evaluation
Strengths Limitations
➢ High ecological validity – easily generalised beyond research ➢ Lack control over possible EV’s and CV’s that may influence
setting. measurement of the DV.
➢ External validity. ➢ Impossible to randomly assign participants – reduces
➢ Mundane realism – likely that the tasks in the experiment are internal validity.
usually carried out by the participant. ➢ Ethical issues – lack of consent if participants are unaware
➢ Participants unaware – less chance of demand characteristics they are being studied.
being shown.
Natural
• IV can’t be manipulated, rather happen naturally so aren’t researcher controlled.
• They are unplanned in daily life and take place in naturally occurring environments – the researchers
choose an already changing IV e.g., the effect of children being isolated from parents on development.
• The DV is measured and the EV’s and CV’s are not controlled.
A03 – Evaluation
Strengths Limitations
➢ High ecological validity. ➢ Lack of control over variables.
➢ Ethical – use of already changing IV’s. ➢ EV’s, due to no control, can’t show cause and effect
➢ High external validity and no chance of demand relationship.
characteristics. ➢ Low replicability – rare events, specific to people thus can’t
be re-tested for reliability.
Quasi
• Participants cannot be randomly assigned between levels of IV; often the level of IV is an innate
characteristic.
• “Hybrid” of a lab & natural experiment – controlled environments however naturally occurring IV.
• An experiment where the IV just exists so participants can’t be randomised, however can be
grouped e.g., study of gender, age etc.
• DV measured and other variables kept constant – the researcher can choose the IV regarding
participants but can’t change it.
A03 – Evaluation
Strengths Limitations
➢ High replicability – controlled conditions. ➢ Lacks randomisation so the causal relationships are not
➢ Only experiment that studies the effects of pre-existing demonstrated.
characteristics of people – allows for comparisons to be ➢ The confounding variables are not considered.
made between different types of people.
Page | 14
