RESEARCH METHODS:
Types of experiments:
The IV: What the experimenter changes.
The DV: What the experimenter is measuring.
The overall aim of an experiment is to establish a cause and effect between the IV and the DV.
Laboratory experiment: An experiment carried out in a controlled environment. Where all situational
variables are controlled (temperature, lighting).
(+)Researchers can exactly replicate the study to check the results (replication). High degree of control aver
all variables including extraneous ones giving greater accuracy. Reliably establishes cause and effect. (-)
May not represent behaviour in a real-life setting (low ecological validity). Participants are likely to know that
this is an experiment and alter their behaviour due to demand characteristics.
Field experiment An experiment carried out in the natural environment. The IV is manipulated by the
researcher and some situational variables are still controlled.
(+) In a natural environment, findings represent what you would usually find in real life with such high
ecological validity. Participants are less likely to know that they are being studied so they are less prone to
demand characteristics so more validity(-) Participants aren't aware they are being studied so gaining
consent is difficult. Extraneous variables are more likely to cause and effect may be less reliable.
Natural experiment An experiment where the IV varies naturally. The researcher doesn't manipulate the IV.
Often involves exploiting an event that is happening. E.g. studying the effect that childhood abuse hos on
forming healthy relationships.
(+) High external validity as it deals with real-life events. As the IV is something that is happening, any
changes to the DV ore more likely to be realistic, increasing validity. (-) Difficult to replicate such
experiments, the lack of control means conditions won't be the same again. The lack of control of the IV
means that there is more chance of confounding variables influencing the variables.
Quasi experiment An experiment where the IV is a naturally existing difference between people that hasn't
been manipulated as it is innate. (age, eye colour or height). (Whereas the IV in a natural experiment was at
one paint manipulated).
(+)High ecological validity. No demanding characteristics. Useful where manipulating the IV would be
impossible or unethical (-)Difficult to set up as there has to be a naturally occurring difference.
Types of obseryatjons·
• Observation techniques are useful when the researcher wants to study natural behaviour.
Naturalistic observation: Behaviour is studied in o natural situation, and the researcher does not interfere
with what is happening.
(+)Gives a realistic picture of spontaneous behaviour, it is likely to be of high ecological validity. (-) There is
little control of all the other things that ore happening, so something unknown to the observer may account
for behaviour.
Controlled observation: Some aspects of the environment are organised by the researcher, enabling them to
investigate the effects of particular situations on behaviour.
(+)An observer can focus on particular aspects of behaviour. (-) The control comes at the cost of the
environment feeling unnatural and participants' behaviour also being less natural
Overt observation: Participants ore aware that they are being observed. Researchers try to be as
unobtrusive as possible.
(+)Participants are aware that they are being observed so there are no ethical issues.(-) Are aware of being
studied which may affect the naturalness of their behaviour - demand characteristics occur.
Covert observation: Tokes place witho~t participants' awareness. Knowing that behaviour is being observed
is likely to alter the participants' behaviour.
(+)Participants ore unaware of being observed and thus their behaviour is more natural.(-) Participants
cannot give consent - ethical issue. However, retrospective consent can be given.
Nan-participant: The observer watches from a distance and does not interact with the participants.
, . be covert
(+)Observers are more objective as they are not part of the group being observed. H More likely t 0
so ethical issues can occur.
Participant observation: The observer joins the group being observed either overtly or covertly.
(+)Provide insight into behaviour from the 'inside' that may not be otherwise gained. (-) More likely to be
covert and thus have issues of participant awareness.
• If 2 or more observers record behaviour, then inter-observer reliability con be calculated. More
than 80% consistency between their data indicates a high level of reliability.
• Psychologists often make observations systematic and objective through the use of a coding
system.
Unstructured observations: The researcher records all relevant behaviour but without an observational
system.
Structured observation: Once they decide which behaviours are relevant, they plan a structured observation
using behavioural categories and sampling procedures.
Behqvioural categories: Behaviours must be operationolised by breaking them down into categories which
are ~bjective.
Sampling: The observer should record all behaviour however, If behaviours are frequent, it is impossible, so
sampling enables them to estimate. Two ways:
Event sampling: Counting the number of times a certain behaviour occurs in a period of time.
Time sampling: Noting what the individual Is doing every 30 seconds by ticking b~hovioural categories on a
checklist.
Self-ceoart tecbniaues·
Questionnaires:
(+) Can get fully informed consent. Con be easily repeated. Can be more willing to reveal personal
information than in interviews. (-) Questions can be vogue. People may not be honest and give socially
desirable answers. The sample may be biased as only certain types of people may fill in the questionnaire.
Interviews - structured: Produces quantitoti".'e data.
(+) Easier to analyse as the questions are the same. Con be replicated due to o standardised format (-)
Participants may get nervous and not tell the truth. Interviewer's expectations may influence the answers the
interviewee may give.
Interviews - unstructured: Produces qualitative data.
(+) Gener9lly, more detailed information. Con access information that may not be revealed in predetermined
questions. (-) More likely to be affected by leading questions and interviewer bias. Requires well-trained
interviewers (can be expensive).
Designing a questionnaire:
• Questions need to be easy to analyse. (closed questions).
~ Questions should be free of bias. (no leading questions or using emotive language).
• Questions should be clear and unambiguous.
• They should contain filler questions. (reduce demand characteristics).
• Test the questionnaire in a pilot study.
• They should be objective and systematic.
.Correlqt[f?DS.
•A mathematical technique in which a researcher investigates an association or relationship
between two variables, called co-variables. A correlation investigates the association between
,, variables (rather than the couse0 and-effect relationship).
• • 0.8 represents a strong correlot1on (0.8 is great). - '·
• An experiment assesses the effect ·of one variable (IV) on another variable which is measured
(DV). Correlations assess the relationship between the co-variables.
• There are negative, positive and no correlations.
(+)Can be ~sed to test for validity ~nd re_liobility. Useful as a ~reliri:,inar'>_' technique allo':"'ing researchers to
identify a link that can be further 1nvest1gated. (-) They only identify a link they do not identify which
variable causes which. There might be a third variable present which is influencing one of the co-variabl
es.
Types of experiments:
The IV: What the experimenter changes.
The DV: What the experimenter is measuring.
The overall aim of an experiment is to establish a cause and effect between the IV and the DV.
Laboratory experiment: An experiment carried out in a controlled environment. Where all situational
variables are controlled (temperature, lighting).
(+)Researchers can exactly replicate the study to check the results (replication). High degree of control aver
all variables including extraneous ones giving greater accuracy. Reliably establishes cause and effect. (-)
May not represent behaviour in a real-life setting (low ecological validity). Participants are likely to know that
this is an experiment and alter their behaviour due to demand characteristics.
Field experiment An experiment carried out in the natural environment. The IV is manipulated by the
researcher and some situational variables are still controlled.
(+) In a natural environment, findings represent what you would usually find in real life with such high
ecological validity. Participants are less likely to know that they are being studied so they are less prone to
demand characteristics so more validity(-) Participants aren't aware they are being studied so gaining
consent is difficult. Extraneous variables are more likely to cause and effect may be less reliable.
Natural experiment An experiment where the IV varies naturally. The researcher doesn't manipulate the IV.
Often involves exploiting an event that is happening. E.g. studying the effect that childhood abuse hos on
forming healthy relationships.
(+) High external validity as it deals with real-life events. As the IV is something that is happening, any
changes to the DV ore more likely to be realistic, increasing validity. (-) Difficult to replicate such
experiments, the lack of control means conditions won't be the same again. The lack of control of the IV
means that there is more chance of confounding variables influencing the variables.
Quasi experiment An experiment where the IV is a naturally existing difference between people that hasn't
been manipulated as it is innate. (age, eye colour or height). (Whereas the IV in a natural experiment was at
one paint manipulated).
(+)High ecological validity. No demanding characteristics. Useful where manipulating the IV would be
impossible or unethical (-)Difficult to set up as there has to be a naturally occurring difference.
Types of obseryatjons·
• Observation techniques are useful when the researcher wants to study natural behaviour.
Naturalistic observation: Behaviour is studied in o natural situation, and the researcher does not interfere
with what is happening.
(+)Gives a realistic picture of spontaneous behaviour, it is likely to be of high ecological validity. (-) There is
little control of all the other things that ore happening, so something unknown to the observer may account
for behaviour.
Controlled observation: Some aspects of the environment are organised by the researcher, enabling them to
investigate the effects of particular situations on behaviour.
(+)An observer can focus on particular aspects of behaviour. (-) The control comes at the cost of the
environment feeling unnatural and participants' behaviour also being less natural
Overt observation: Participants ore aware that they are being observed. Researchers try to be as
unobtrusive as possible.
(+)Participants are aware that they are being observed so there are no ethical issues.(-) Are aware of being
studied which may affect the naturalness of their behaviour - demand characteristics occur.
Covert observation: Tokes place witho~t participants' awareness. Knowing that behaviour is being observed
is likely to alter the participants' behaviour.
(+)Participants ore unaware of being observed and thus their behaviour is more natural.(-) Participants
cannot give consent - ethical issue. However, retrospective consent can be given.
Nan-participant: The observer watches from a distance and does not interact with the participants.
, . be covert
(+)Observers are more objective as they are not part of the group being observed. H More likely t 0
so ethical issues can occur.
Participant observation: The observer joins the group being observed either overtly or covertly.
(+)Provide insight into behaviour from the 'inside' that may not be otherwise gained. (-) More likely to be
covert and thus have issues of participant awareness.
• If 2 or more observers record behaviour, then inter-observer reliability con be calculated. More
than 80% consistency between their data indicates a high level of reliability.
• Psychologists often make observations systematic and objective through the use of a coding
system.
Unstructured observations: The researcher records all relevant behaviour but without an observational
system.
Structured observation: Once they decide which behaviours are relevant, they plan a structured observation
using behavioural categories and sampling procedures.
Behqvioural categories: Behaviours must be operationolised by breaking them down into categories which
are ~bjective.
Sampling: The observer should record all behaviour however, If behaviours are frequent, it is impossible, so
sampling enables them to estimate. Two ways:
Event sampling: Counting the number of times a certain behaviour occurs in a period of time.
Time sampling: Noting what the individual Is doing every 30 seconds by ticking b~hovioural categories on a
checklist.
Self-ceoart tecbniaues·
Questionnaires:
(+) Can get fully informed consent. Con be easily repeated. Can be more willing to reveal personal
information than in interviews. (-) Questions can be vogue. People may not be honest and give socially
desirable answers. The sample may be biased as only certain types of people may fill in the questionnaire.
Interviews - structured: Produces quantitoti".'e data.
(+) Easier to analyse as the questions are the same. Con be replicated due to o standardised format (-)
Participants may get nervous and not tell the truth. Interviewer's expectations may influence the answers the
interviewee may give.
Interviews - unstructured: Produces qualitative data.
(+) Gener9lly, more detailed information. Con access information that may not be revealed in predetermined
questions. (-) More likely to be affected by leading questions and interviewer bias. Requires well-trained
interviewers (can be expensive).
Designing a questionnaire:
• Questions need to be easy to analyse. (closed questions).
~ Questions should be free of bias. (no leading questions or using emotive language).
• Questions should be clear and unambiguous.
• They should contain filler questions. (reduce demand characteristics).
• Test the questionnaire in a pilot study.
• They should be objective and systematic.
.Correlqt[f?DS.
•A mathematical technique in which a researcher investigates an association or relationship
between two variables, called co-variables. A correlation investigates the association between
,, variables (rather than the couse0 and-effect relationship).
• • 0.8 represents a strong correlot1on (0.8 is great). - '·
• An experiment assesses the effect ·of one variable (IV) on another variable which is measured
(DV). Correlations assess the relationship between the co-variables.
• There are negative, positive and no correlations.
(+)Can be ~sed to test for validity ~nd re_liobility. Useful as a ~reliri:,inar'>_' technique allo':"'ing researchers to
identify a link that can be further 1nvest1gated. (-) They only identify a link they do not identify which
variable causes which. There might be a third variable present which is influencing one of the co-variabl
es.
