Research Methods
Aims & Hypotheses
-An aim outlines the purpose of research and makes the focus of the study clear. It also explains what the research is
about.
-An hypothesis is a testable statement of what the researcher believes is true. It should be clearly defined and
operationalised (made measurable).
-A null hypothesis states that there will be no difference, this is what the researcher is setting out to disprove.
-A alternative hypothesis has two types: directional (one tailed) and non-directional (two tailed).
-A directional hypothesis states that there will be a difference and the direction of the effect.
-This is used when the findings of previous research suggest a particular outcome.
-Ex. Students who listen to music while revising will achieve a higher score in a test than those who do not listen to
music.
-A non-directional hypothesis states that there will be a difference but not the direction of the effect.
-This is used when there is no previous research or findings from previous research are contradictory.
-Ex. There will be a significant difference in test scores between students who listen to music and those who do not.
Variables
-The independent variable is manipulated.
-The dependent variable is measured.
Levels of the IV a change is made to the IV.
-There may be more than one experimental group used.
Control of variables
-There are always several unwanted factors that have the potential to affect the relationship between the IV and the
DV which may distort results.
-Psychologists are aware of this issue and have devised several ways to tackle it.
Confounding variables
-These cannot be controlled as they cannot be changed.
-Confounding variables change systematically with the IV and therefore we cannot be sure that the change in the
DV is due to the CV or the IV.
-An example of CV is personality, within research, if personality is not controlled for, we cannot be sure whether the
results are due to individual personality types or the IV.
-This will mean the outcome is meaningless for psychological research.
Extraneous variables
-Usually something can be done to control these.
-The only thing that should influence the DV is the IV, any other variable that can potentially interfere with the IV
should be controlled or removed.
-These unwanted variables are known as extraneous variables. They should, if possible, be identified at the
beginning of research and steps should be taken to minimise their influence.
-These can be straightforward to control such as lightning and heating and do not vary systematically with the IV.
Types of extraneous variables
, Type of EV Definition Control
Participant Variables Characteristics of the participant Use a repeated measures design
such as their abilities, gender, age. in a matched pairs design,
They only act as extraneous matching participants carefully and
variables where an independent allocating them to each condition
group design is used. randomly.
Participant Effects The hawthorne effect: participants Single blind technique:
perform better when they know they participants do not know the
are being observed. hypothesis and which condition they
The screw you effect: participants are in.
behave in a way to deliberately spoil This has a lack of informed consent
an experiment. and an element of deception which
Demand characteristics: cues in raises an ethical dilemma.
the environment that lead
participants to think they should
behave a particular way and
therefore differently to how they
normally would.
Social desirability bias:
participants want to be seen in the
best possible way so behave in
ways they consider socially
acceptable in that situation.
Investigator Effects/Bias The behaviour of the investigator Using standardised instructions
may affect the participants and thus and the double blind technique.
the DV. THis can be both physical Neither the participant or the
characteristics and the investigator's research assistant know the
behaviour. hypothesis or thr condition the
Also expectation effects where the participants are in. (This means the
researcher is deeply committed to research assistant must carry out
achieving a particular outcome the research on behalf of the
experimenter).
Situational Variables A feature of the research setting that Using a standardised procedure
may act as a confounding variable. and instructions as well as testing
Ex. temperature, time, lighting, every participant in quiet, lab
materials used, instructions etc. conditions
Order Effects Participants may perform differently Counterbalancing (ABBA). ½
on two conditions because of the participants do condition A then B.
order in which the conditions are ½ do condition B then A.
performed. EX. practise, bored, tired
etc
Validity
-Validity is the extent to which something measures what it sets out to measure.
-Internal validity: this refers to the extent to which the results of a study occur due to the variable being tested and
not any other variables.
, -To gain high internal validity, studies must be carefully designed and EVs controlled to ensure you are testing what
you intended to test.
-It can be improved by using tighter controls on the Evs and ensuring increased reliability in a procedure and
measurement.
-Ex, choosing a lab experiment which is highly controlled.
-External validity: this is the extent to which results can be generalised to different situations / real life (ecological
validity), to different people (population validity) and whether they are applicable to the current time period
(temporal validity).
-This can be improved by using more realistic tests or natural settings rather than artificial conditions in controlled
settings.
-Ex. Conducting research in a real world environment such as a classroom.
Reliability
-Reliability refers to the extent to which research findings are consistent. This means that if research is repeated
similar or the same result will be obtained.
-Internal reliability refers to internal consistency and is demonstrated in the way in which a procedure is applied and
measurements used.
-External reliability refers to the extent to which a test is consistent over time and whether it consistently produces
the same result.
-A study can be reliable but not valid.
-Reliability can be improved through different measures:
-Increasing objectivity of measures will increase their consistency overtime and between people.
-Operationalizing variables will also improve reliability. Ex. defining aggression as pushing or kicking.
-Another way is to standardise procedures to ensure a study is reliable both internally and externally. This ensures
all participants have a consistent experience and the study can be replicated to test the reliability of results.
Assessing reliability
-The test-retest method: this measures external reliability. A test is given to the same participants on two or more
occasions and the results are correlated. If the results are the same it can be considered as reliable.
-Inter-observer reliability: this measures internal reliability. Several observers independently record data and the
results are correlated. If they agree then measures are reliable.
-The split-half method: this measures internal reliability of tests or questionnaires. Participants complete the test,
which is divided into two halves and each half is scored. If the test consistently measures what it is supposed to, the
scores for each half will be the same.
Pilot study
-This is a practice run for a study where the procedure is tested on a few people. These people will not then take
part in the real study.
-This allows researchers to spot any problems with the design, which can then be fixed before the study is
conducted on the whole sample.
-Ex. The instructions in an experiment may need to be clearer, or the wording on a questionnaire may be ambiguous
and need changing.
Sampling
-Sampling is the process through which participants are selected.
-Population = a group who are the focus of interest, from which a smaller sample is drawn.
-Sample = the sample is drawn from a (target) population and is assumed to be representative of the population.
-Sampling techniques = method used to select people from the population.
-Bias = this occurs when a particular group within the sample is over or under represented which can limit the extent
to which generalisations can be made.
-Generalisation = the extent to which findings and conclusions can be applied to the population. This is possible if
the sample is representative of the population.
Aims & Hypotheses
-An aim outlines the purpose of research and makes the focus of the study clear. It also explains what the research is
about.
-An hypothesis is a testable statement of what the researcher believes is true. It should be clearly defined and
operationalised (made measurable).
-A null hypothesis states that there will be no difference, this is what the researcher is setting out to disprove.
-A alternative hypothesis has two types: directional (one tailed) and non-directional (two tailed).
-A directional hypothesis states that there will be a difference and the direction of the effect.
-This is used when the findings of previous research suggest a particular outcome.
-Ex. Students who listen to music while revising will achieve a higher score in a test than those who do not listen to
music.
-A non-directional hypothesis states that there will be a difference but not the direction of the effect.
-This is used when there is no previous research or findings from previous research are contradictory.
-Ex. There will be a significant difference in test scores between students who listen to music and those who do not.
Variables
-The independent variable is manipulated.
-The dependent variable is measured.
Levels of the IV a change is made to the IV.
-There may be more than one experimental group used.
Control of variables
-There are always several unwanted factors that have the potential to affect the relationship between the IV and the
DV which may distort results.
-Psychologists are aware of this issue and have devised several ways to tackle it.
Confounding variables
-These cannot be controlled as they cannot be changed.
-Confounding variables change systematically with the IV and therefore we cannot be sure that the change in the
DV is due to the CV or the IV.
-An example of CV is personality, within research, if personality is not controlled for, we cannot be sure whether the
results are due to individual personality types or the IV.
-This will mean the outcome is meaningless for psychological research.
Extraneous variables
-Usually something can be done to control these.
-The only thing that should influence the DV is the IV, any other variable that can potentially interfere with the IV
should be controlled or removed.
-These unwanted variables are known as extraneous variables. They should, if possible, be identified at the
beginning of research and steps should be taken to minimise their influence.
-These can be straightforward to control such as lightning and heating and do not vary systematically with the IV.
Types of extraneous variables
, Type of EV Definition Control
Participant Variables Characteristics of the participant Use a repeated measures design
such as their abilities, gender, age. in a matched pairs design,
They only act as extraneous matching participants carefully and
variables where an independent allocating them to each condition
group design is used. randomly.
Participant Effects The hawthorne effect: participants Single blind technique:
perform better when they know they participants do not know the
are being observed. hypothesis and which condition they
The screw you effect: participants are in.
behave in a way to deliberately spoil This has a lack of informed consent
an experiment. and an element of deception which
Demand characteristics: cues in raises an ethical dilemma.
the environment that lead
participants to think they should
behave a particular way and
therefore differently to how they
normally would.
Social desirability bias:
participants want to be seen in the
best possible way so behave in
ways they consider socially
acceptable in that situation.
Investigator Effects/Bias The behaviour of the investigator Using standardised instructions
may affect the participants and thus and the double blind technique.
the DV. THis can be both physical Neither the participant or the
characteristics and the investigator's research assistant know the
behaviour. hypothesis or thr condition the
Also expectation effects where the participants are in. (This means the
researcher is deeply committed to research assistant must carry out
achieving a particular outcome the research on behalf of the
experimenter).
Situational Variables A feature of the research setting that Using a standardised procedure
may act as a confounding variable. and instructions as well as testing
Ex. temperature, time, lighting, every participant in quiet, lab
materials used, instructions etc. conditions
Order Effects Participants may perform differently Counterbalancing (ABBA). ½
on two conditions because of the participants do condition A then B.
order in which the conditions are ½ do condition B then A.
performed. EX. practise, bored, tired
etc
Validity
-Validity is the extent to which something measures what it sets out to measure.
-Internal validity: this refers to the extent to which the results of a study occur due to the variable being tested and
not any other variables.
, -To gain high internal validity, studies must be carefully designed and EVs controlled to ensure you are testing what
you intended to test.
-It can be improved by using tighter controls on the Evs and ensuring increased reliability in a procedure and
measurement.
-Ex, choosing a lab experiment which is highly controlled.
-External validity: this is the extent to which results can be generalised to different situations / real life (ecological
validity), to different people (population validity) and whether they are applicable to the current time period
(temporal validity).
-This can be improved by using more realistic tests or natural settings rather than artificial conditions in controlled
settings.
-Ex. Conducting research in a real world environment such as a classroom.
Reliability
-Reliability refers to the extent to which research findings are consistent. This means that if research is repeated
similar or the same result will be obtained.
-Internal reliability refers to internal consistency and is demonstrated in the way in which a procedure is applied and
measurements used.
-External reliability refers to the extent to which a test is consistent over time and whether it consistently produces
the same result.
-A study can be reliable but not valid.
-Reliability can be improved through different measures:
-Increasing objectivity of measures will increase their consistency overtime and between people.
-Operationalizing variables will also improve reliability. Ex. defining aggression as pushing or kicking.
-Another way is to standardise procedures to ensure a study is reliable both internally and externally. This ensures
all participants have a consistent experience and the study can be replicated to test the reliability of results.
Assessing reliability
-The test-retest method: this measures external reliability. A test is given to the same participants on two or more
occasions and the results are correlated. If the results are the same it can be considered as reliable.
-Inter-observer reliability: this measures internal reliability. Several observers independently record data and the
results are correlated. If they agree then measures are reliable.
-The split-half method: this measures internal reliability of tests or questionnaires. Participants complete the test,
which is divided into two halves and each half is scored. If the test consistently measures what it is supposed to, the
scores for each half will be the same.
Pilot study
-This is a practice run for a study where the procedure is tested on a few people. These people will not then take
part in the real study.
-This allows researchers to spot any problems with the design, which can then be fixed before the study is
conducted on the whole sample.
-Ex. The instructions in an experiment may need to be clearer, or the wording on a questionnaire may be ambiguous
and need changing.
Sampling
-Sampling is the process through which participants are selected.
-Population = a group who are the focus of interest, from which a smaller sample is drawn.
-Sample = the sample is drawn from a (target) population and is assumed to be representative of the population.
-Sampling techniques = method used to select people from the population.
-Bias = this occurs when a particular group within the sample is over or under represented which can limit the extent
to which generalisations can be made.
-Generalisation = the extent to which findings and conclusions can be applied to the population. This is possible if
the sample is representative of the population.
