Research Methods
Experimental Method
Experimental method involves the manipulation of an IV to measure the effect on a DV. Experiments may
be laboratory, field, natural, or quasi.
Aim - a general statement of what the researcher intends to investigate- purpose of the study.
Hypothesis - a clear precise testable statement that states the relationships between the variables to be
measured. Stated at outset of the study. Can be directional or non-directional.
- Directional hypothesis – states the direction of the difference or the relationship, researchers use
directional hypotheses when findings of previous research suggest a particular outcome - “People
who … will be more…than those who...”
- Non-Directional hypothesis- states there is a difference between conditions but does not specify
the direction. Researchers use non-directional hypotheses when there is no previous research or
findings are contradictory - “People who…differ in term of … compared to….”
- Null Hypothesis – states there is no difference between conditions or co-variables.
Variables and Control of Variables
Independent variable - changes or is manipulated so the effect on the DV can be measured.
Dependent variable - measured by the researcher.
Operationalisation - clearly defining variables in terms of how they can be measured.
Extraneous variables - any variable, other than IV, that may influence the DV if it is not controlled. EV’s are
nuisance variables that are more unpredictable. Where possible, these are identified at the beginning of an
experiment to have their influence minimised. Do not vary systematically.
Cofounding variables - any variable, other than the IV, that may have affected the DV so we cannot be sure
of the true source of changes to the DV. Vary systematically with the IV, e.g. personality.
Demand characteristics - any cue from the researcher or research situation that may be interpreted by
participants as revealing true purpose of the investigation- may lead to a participant changing their
behaviour within the research situation.
Investigator effects - any effect of the investigators behaviour on the research outcome (DV). This may
include everything from the design of the study to the selection of, interaction with participants during the
research process.
Randomisation - use of chance in order to reduce the effects of extraneous/cofounding variables when
designing materials and deciding the order of conditions.
Standardisation - using the exact same formalised procedures and instructions for all participants in a
research study.
,Experimental Design
Experimental design refers to the way participants are used in experiments – how the testing is arranged in
relation to different experimental conditions.
Experimental design Evaluation
Independent Groups – when two separate Independent Groups – The biggest issue with independent
groups of participants experience two groups is that the participants who occupy the different
different conditions of the experiment. groups are not the same – it assesses the performance of
Assesses performances of two different two different groups of people. To deal with this,
groups of people researchers use random allocation. Independent designs
are less economical as each participant contributes a
single result. Twice as many participants are required.
However, order effects and demand characteristics are
less likely within this.
Repeated Measures – each participant is required to
Repeated measures – all participants complete tasks at least twice and the order of these tasks
experience both conditions to see if there is may be significant. Order effects may arise because
a difference. Guarantees we are comparing repeating tasks could create fatigue/boredom that may
‘like with like’. cause deterioration in performance on the second task.
Alternatively, performance may improve through practice,
especially on a skill-based task. Order acts as a cofounding
variable. In order to deal with this, researchers use
counterbalancing (half the participants experience
conditions in one order whilst the other half experiences
it in the opposite order). It is also more likely participants
will work out the aim of the study having carried out more
than one task. However, participant variables are more
carefully controlled and fewer participants are needed.
Matched pairs- participants are first Matched pairs – participants only take part in a single
matched on some variables that may affect condition, so order effects and demand characteristics are
the DV then one member of the pair is less of a problem. Although there is some attempt to
assigned to condition A and the other B. reduce participant variables, they can never actually be
Attempts to control cofounding variable of matched exactly. Matching may be more time consuming
participant variables. and expensive, so this is less economical than other
designs.
, Types of Experiment
Laboratory – conducted in highly controlled environments.
Strengths Limitations
-High control over extraneous variables. Means -Can lack generalisability as lab environment may
any effect on the DV is likely to be manipulation of be artificial and not like everyday life. Low external
the IV. Therefore, the researcher can be more validity.
certain about demonstrating cause and effect (high -Participants are aware they are being tested and
internal validity). may produce demand characteristics. Tasks
-Replication is more possible due to high level of participants conduct are normally meaningless.
control. Replication is vital to check to see whether
results are valid or just a ‘one off’.
Field – IV is manipulated in a natural more everyday setting.
Strengths Limitations
-Have higher mundane realism than lab -Loss of control of extraneous variables. Cause an
experiments as environment is more natural. May effect between IV and DV more difficult to
produce more valid or authentic behaviour. establish and precise replications may not be
-High external validity as participants are unaware, possible.
they are being studied. -Ethical issues arise – as participants are unaware
that they are being studied, they cannot consent to
being studied and this is an invasion of privacy.
Natural – when researcher takes advantage of pre-existing variable. The IV is natural, potentially not the
setting. The IV would have changed even if the researcher were not interested.
Strengths Limitations
-Provide opportunities for research that may not -Naturally occurring events happen rarely,
have otherwise been undertaken for practical/ reducing research opportunities. Also limits the
ethical reasons. ability to generalise findings.
-Often have high external validity as they involve -Participants may not be randomly allocated to
the study of real-life issues and problems as they experimental conditions. This means that the
arise, such as the effect of natural disaster on researcher is less sure if IV affects DV.
stress levels.
Quasi- IV based on existing difference between people- no one has manipulated this variable (i.e. age)
Strengths Limitations
Often carried out under controlled conditions and Cannot randomly allocate participants to
share strengths of lab experiments. conditions so there may be cofounding variables.
Experimental Method
Experimental method involves the manipulation of an IV to measure the effect on a DV. Experiments may
be laboratory, field, natural, or quasi.
Aim - a general statement of what the researcher intends to investigate- purpose of the study.
Hypothesis - a clear precise testable statement that states the relationships between the variables to be
measured. Stated at outset of the study. Can be directional or non-directional.
- Directional hypothesis – states the direction of the difference or the relationship, researchers use
directional hypotheses when findings of previous research suggest a particular outcome - “People
who … will be more…than those who...”
- Non-Directional hypothesis- states there is a difference between conditions but does not specify
the direction. Researchers use non-directional hypotheses when there is no previous research or
findings are contradictory - “People who…differ in term of … compared to….”
- Null Hypothesis – states there is no difference between conditions or co-variables.
Variables and Control of Variables
Independent variable - changes or is manipulated so the effect on the DV can be measured.
Dependent variable - measured by the researcher.
Operationalisation - clearly defining variables in terms of how they can be measured.
Extraneous variables - any variable, other than IV, that may influence the DV if it is not controlled. EV’s are
nuisance variables that are more unpredictable. Where possible, these are identified at the beginning of an
experiment to have their influence minimised. Do not vary systematically.
Cofounding variables - any variable, other than the IV, that may have affected the DV so we cannot be sure
of the true source of changes to the DV. Vary systematically with the IV, e.g. personality.
Demand characteristics - any cue from the researcher or research situation that may be interpreted by
participants as revealing true purpose of the investigation- may lead to a participant changing their
behaviour within the research situation.
Investigator effects - any effect of the investigators behaviour on the research outcome (DV). This may
include everything from the design of the study to the selection of, interaction with participants during the
research process.
Randomisation - use of chance in order to reduce the effects of extraneous/cofounding variables when
designing materials and deciding the order of conditions.
Standardisation - using the exact same formalised procedures and instructions for all participants in a
research study.
,Experimental Design
Experimental design refers to the way participants are used in experiments – how the testing is arranged in
relation to different experimental conditions.
Experimental design Evaluation
Independent Groups – when two separate Independent Groups – The biggest issue with independent
groups of participants experience two groups is that the participants who occupy the different
different conditions of the experiment. groups are not the same – it assesses the performance of
Assesses performances of two different two different groups of people. To deal with this,
groups of people researchers use random allocation. Independent designs
are less economical as each participant contributes a
single result. Twice as many participants are required.
However, order effects and demand characteristics are
less likely within this.
Repeated Measures – each participant is required to
Repeated measures – all participants complete tasks at least twice and the order of these tasks
experience both conditions to see if there is may be significant. Order effects may arise because
a difference. Guarantees we are comparing repeating tasks could create fatigue/boredom that may
‘like with like’. cause deterioration in performance on the second task.
Alternatively, performance may improve through practice,
especially on a skill-based task. Order acts as a cofounding
variable. In order to deal with this, researchers use
counterbalancing (half the participants experience
conditions in one order whilst the other half experiences
it in the opposite order). It is also more likely participants
will work out the aim of the study having carried out more
than one task. However, participant variables are more
carefully controlled and fewer participants are needed.
Matched pairs- participants are first Matched pairs – participants only take part in a single
matched on some variables that may affect condition, so order effects and demand characteristics are
the DV then one member of the pair is less of a problem. Although there is some attempt to
assigned to condition A and the other B. reduce participant variables, they can never actually be
Attempts to control cofounding variable of matched exactly. Matching may be more time consuming
participant variables. and expensive, so this is less economical than other
designs.
, Types of Experiment
Laboratory – conducted in highly controlled environments.
Strengths Limitations
-High control over extraneous variables. Means -Can lack generalisability as lab environment may
any effect on the DV is likely to be manipulation of be artificial and not like everyday life. Low external
the IV. Therefore, the researcher can be more validity.
certain about demonstrating cause and effect (high -Participants are aware they are being tested and
internal validity). may produce demand characteristics. Tasks
-Replication is more possible due to high level of participants conduct are normally meaningless.
control. Replication is vital to check to see whether
results are valid or just a ‘one off’.
Field – IV is manipulated in a natural more everyday setting.
Strengths Limitations
-Have higher mundane realism than lab -Loss of control of extraneous variables. Cause an
experiments as environment is more natural. May effect between IV and DV more difficult to
produce more valid or authentic behaviour. establish and precise replications may not be
-High external validity as participants are unaware, possible.
they are being studied. -Ethical issues arise – as participants are unaware
that they are being studied, they cannot consent to
being studied and this is an invasion of privacy.
Natural – when researcher takes advantage of pre-existing variable. The IV is natural, potentially not the
setting. The IV would have changed even if the researcher were not interested.
Strengths Limitations
-Provide opportunities for research that may not -Naturally occurring events happen rarely,
have otherwise been undertaken for practical/ reducing research opportunities. Also limits the
ethical reasons. ability to generalise findings.
-Often have high external validity as they involve -Participants may not be randomly allocated to
the study of real-life issues and problems as they experimental conditions. This means that the
arise, such as the effect of natural disaster on researcher is less sure if IV affects DV.
stress levels.
Quasi- IV based on existing difference between people- no one has manipulated this variable (i.e. age)
Strengths Limitations
Often carried out under controlled conditions and Cannot randomly allocate participants to
share strengths of lab experiments. conditions so there may be cofounding variables.
