[Link]
ly/pmt-edu
CAIE Psychology
A-level
Research Methods
Notes
Author: Remika Sirikulthada
[Link] [Link]
This work by PMT Education is licensed [Link] [Link]
under CC BY-NC-ND 4.0
Variables
Experiment: An investigation looking for a casual relationship in which an
independent variable is manipulated and is expected to be responsible for
changes in the dependent variable.
Independent variable: The factor in an experiment which is manipulated to
create two or more conditions (levels) and is expected to be responsible for
changes in the dependent variable.
Dependent variable: The factor in an experiment which is measured and is
expected to change under the influence of the independent variable
Extraneous variable: A variable which either acts randomly, affecting the DV in
all levels of the IV or systematically i.e., on one level of the IV.
Operationalise: To make the hypothesis testable and measurable.
- Ensure you have written your variables in your hypothesis in a way that
someone else can read and repeat your experiment
Hypothesis and Aims
Hypothesis: A testable statement of what the researcher predicts will be the outcome of the
study.
Aim: Usually starts with 'to investigate', the aim of a study is a statement of what the
researcher intends to investigate.
Directional hypothesis: A one-tailed directional hypothesis predicts the nature
of the effect of the independent variable on the dependent variable.
Non-directional hypothesis: A two-tailed non-directional hypothesis predicts
that the independent variable will have an effect on the dependent variable, but
the direction of the effect is not specified.
Null hypothesis: There is no relationship between the two variables being studied (chance)
(H0).
Alternative hypothesis: There is a relationship between the two variables being
studied (significant in terms of supporting the theory being investigated) (H1).
[Link] [Link]
This work by PMT Education is licensed under CC BY-NC-ND 4.0
[Link] [Link] [Link]
Falsification: The action of falsifying/ proving a theory or piece of information to
be wrong. Experiments
1. Laboratory Experiment: An experiment conducted in an artificially
controlled environment. There is usually a standardised procedure. The
researcher manipulates the IV to see what effect it has on the DV.
- There are strict controls over EVs.
- Participants are aware that they are taking part in a study thought they may
not know the true aims.
- Looking for cause-and-effect relationship.
Strengths
- High control over extraneous variables means you can make statements
about cause and effect and therefore increase internal validity.
- Replication is possible due to standardisation, which means you can
test if the findings are reliable.
Limitations
- The lab environment may be artificial which means we may not be able
to generalise the findings to real-life situations (low ecological
validity).
- The task can be artificial e.g. memory test so it may lack mundane realism.
- People know they are being tested and therefore pick up cues from the
environment and change their behaviour (demand characteristics). This
lowers internal validity.
2. Field Experiment: An experiment conducted in a more natural
environment - anywhere outside a lab where the investigated
behaviour could naturally occur. The researcher still manipulates the IV
to the what effect it has on the DV
- The IV is still deliberately manipulated by the researcher but participants are not
aware that they are participating in a research experiment
Strengths
- The environment is natural so there is normally higher ecological validity.
- The participants do not normally know they are in an experiment and
therefore there is a reduction in demand characteristics.
Limitations
- There is less control of extraneous variables so cause and effect are
more difficult to establish. This can lower the internal validity.
- There can be ethical issues if participants are unaware that they are
being studied (e.g. lack of informed consent).
[Link]
[Link] [Link] [Link] [Link]
[Link]
3. Natural Experiment: Experiments that are generally conducted in every
day (real-life) environments. The experimenter has no control over the IV as
it occurs naturally in real life. - They are more common where the researcher
cannot ethically manipulate the IV eg. natural disaster or size of a family
Strengths
- They provide opportunities for research that might not otherwise
be undertaken for practical or ethical reasons. They offer unique
insights.
- They often have high ecological validity because they study real-life events.
Limitations
- It can be difficult to establish causality (cause and effect) due to the
lack of controls. - A naturally occurring event may happen very rarely,
reducing the opportunities for research.
- Participants may not be randomly allocated to conditions. This
means you cannot be sure that the IV is affecting the DV and
reducing validity.
4. Quasi (not on specification): When the IV is naturally existing in the
person/ The researcher cannot manipulate IV.
- Cannot randomly allocate therefore it is not a ‘true’ experiment.
- The IV naturally exists in the person e.g. Down Syndrome/ autism/ neurotypical.
Strengths
- They are often carried out under controlled conditions (the same
strengths as lab experiments).
Limitations
- The participants cannot be randomly allocated to conditions. This
means that there may be participant confounding variables.
Experimental Design
Design type Description Evaluation Fixing problems
Repeated Each participant takes Advantage: Order effect are
measur part in BOTH - Participant extraneous big problem -
es conditions. They are variables are counterbalancin
design selected for BOTH controlled a method to re
groups. for (reduced but never the impact of
- A researcher fully eliminated) effects
recruits a group of - Less participants Example: Half d
participants needed as they appear while the other
- They would all in both conditions and vice versa.
complete condition Disadvantage: Sometimes refe
1 and 2 - Order effects: this is an as ABBA
extraneous variable effect
where the order in
which conditions of the
[Link] [Link] [Link]
experiment affects the
results eg. practice
effects or fatigue effects
- Demand
characteristics: a cue
that makes participants
unconsciously aware of
the aims of the study or
helps participants work
out what the researcher
wants to find and this
causes them to change
their behaviour
Indepen Each participant Advantage: Random alloca
de nt takes part in ONE - Order effects are used to try and
groups condition only. They reduced as participants combat the pro
design are selected for ONE only taken part in one of
group only. condition participant varia
- Researcher recruits - Demand characteristics
a group of are reduced - less The researcher
participants likely required to hav
- The participants are to guess the aim impact on whic
then divided participant goes
equally
Disadvantage: where: 1.
into two groups - Participant extraneous Manual/ lott
through random variables between method
allocation groups lower the 2. Electronic
- One does one and internal validity of the
the other does
study
the
- Less economical than
other - only taking
repeated measures
part in one condition as
you need twice as many
participants
Matched Each participant Advantage: Restrict the num
pairs takes part in ONE - Participants only take of variables to
condition (group) part in one match on make
only, but the condition - easier eg.
participants are order effects and demand gender
matched on variables characteristics are
considered relevant reduced
(e.g. age, gender, IQ) - Participant extraneous
- Researcher recruits variables are
a group of reduced
participants (not eliminated)
- Match the p on
specific variables Disadvantage:
eg. age or IQ - Although there is a
- Ex. a grade girl in decrease in the P
the morning so they variable, the
need a grade girl in participants cannot be
the afternoon - one truly matched
of the pair will go - More time consuming
into condition a and thus less economical
the other in - More difficult to do and
condition b might have to recruit
extra people
[Link] [Link] [Link]
Control of Variables
In psychology when conducting experiments we have to make a trade-off between control
and realism.
Aronson and Calsmith (1968) introduced the concept of mundane realism as a
potential threat to external validity (ecological validity).
Experimental realism refers to whether an experiment has a psychological impact
and ‘feels real’ to a participant and is a threat to internal validity
- Laboratory experiments have the highest control over extraneous variables
but have the least amount of realism.
- It is good to have high control as this means that you can establish
cause-and-effect relationships (causality).
- However, unless you conduct a study that has high realism then you
may not be able to generalise the findings to real situations and
environments.
Research should be a cyclical process whereby we see the findings in the
lab experiment and then we test them in the real world. They need to be
found in the lab first where it is highly controlled and we can establish
causality.
Assessing Validity
- The time of day becomes a situational
confounding variable if it has an effect on
the
results.
- Confounding variables: Any variable other
than
the IV, that has not been controlled and
therefore
may have had an effect on the DV, lowering
the
internal validity of the study.
Demand Characteristics: Any cue that may be
interpreted
by participants as revealing the purpose or aim of the
investigation. This may lead to a participant changing their
behaviour within the research situation.
- Impacts the internal validity (construct validity).
[Link] [Link] [Link]
Possible cues resulting in demand characteristics:
- The actual communication during the study: the instructions and any
implicit clues from non-verbal communication.
- What the participant may already have heard about the study (for
example, from other participants).
- The way the participant is approached initially and asked to volunteer.
- The type of person that the researcher is: whether, for example, he or she is
formal or relaxed and so on.
- The setting of the study.
Investigator Effects: Investigator effects are where a researcher (consciously or
unconsciously) acts in a way to support their prediction. This can be a particular
problem when observing events that can be interpreted in more than one way
- It might lead to the participant fulfilling the investigator’s expectations.
- These are where the investigator may operationalise the measurement of
variables in such a way that the desired result is more likely.
- The investigator loose procedure effect is where an investigator may not
clearly state the standardised instructions properly which leaves room for
the results to be influenced by the experimenter.
Ways to control these extraneous variables
1. Random allocation: Each participant has the same opportunity to be assigned to
any given group, so individual differences in responses or ability are far less
likely to consistently affect results. a. This helps control participant extraneous
variables
2. Standardised procedures: There will normally be an attempt at a standardised
procedure for each experiment.
a. This allows for replication to determine the reliability [consistency] of the
results b. This helps control the validity [limit in situational extraneous
variable increases the internal validity]
3. Randomisation: Presenting any stimuli in an experiment in a ‘random’ manner
to avoid it having an effect on the DV. It reduces the chance of practice
effects becoming a confounding variable. a. Example: Randomly choosing
the order of words in a memory recall experiment b. This helps control
situational extraneous variables
4. Single blind test: when the participant does not know which condition of a
study they are in a. Example, placebo drug or non placebo
b. This helps reduce demand characteristics
5. Double blind test: when neither participant nor the investigator know
which condition the participants are in
a. Example, neither investigator nor participant know about
placebo vs non placebo b. This reduces researcher bias or
investigator effects
Reliability and Validity
Why is it important?
- Good psychology has to be both reliable and valid in order for people to
trust the results - Particularly important for psychology:
- Constructs that are studies may be abstract and hard to define
- We use operational definitions
[Link] [Link] [Link]
- People disagree on the best way to measure certain constructs
- Applies to the methodology used and the effects observed:
- Methodology: design, measures and procedures
- Effects: the pattern of results
Reliability: The extent to which something is consistent.
Reliable methodologies will produce the same (or very similar) results each time
they are used with a particular sample of individuals e.g. diagnostic test for
depression.
Reliable effects are replicated across a number of different studies and
individuals e.g. the Stroop effect.
Measures of reliability:
1. Test-retest reliability: This measures test consistency,i.e. the reliability of a
test measured over time. Example: if a person completes the same test
twice at different times, are the scores the same?
- Two tests correlation coefficient of 0.8 or above -> we can assume the test
is reliable 2. Inter-rater/ observer reliability: inter-rater reliability is the degree
of agreement among raters - 0.8+ high correlation between the observers/
raters -> measure is reliable
Validity: Refers to the extent to which something is measuring what it is claiming to measure.
Valid methodologies: Measure the construct that they are intended to
measure e.g. does you aggression questionnaire really measure
aggression?
Valid effects: Can be attributed to the independent variable i.e.. they are not
produced by confounding variables.
Types of Validity:
1. Internal validity: Refers to the extent to which a study establishes a cause-and-
effect relationship between the IV and the DV. How much are extraneous variables
being controlled for? 2. Construct validity: Refers to the degree to which a test
measures what it claims, or purports, to be measuring, e.g. how effectively does a
mood self-assessment for depression measure the construct of ‘depression’.
- Affected by demand characteristics
3. Concurrent validity: This asks whether a measure is in agreement with a pre-
existing measure that is validated to test for the same (or a very similar)
concept. This is gauged by correlation measures against each other.
- E.g. does a new test measuring intelligence, agree with an
existing measure of intelligence?
4. Predictive validity: The degree to which a test accurately predicts a criterion
that will occur in the future.
5. Face validity/ logical validity: A simple form of validity where you apply a
superficial and subjective assessment of whether or not your study or
test measures what it is intended to measure.
6. External validity: Generalisability, refers to the extent the results can be
generalised to other settings.
a. Ecological (setting)
b. Population (people)
[Link] [Link] [Link]
Population validity
- Cultural bias
- Individualistic: people from individualistic cultures place importance on
themselves rather than the group
- Generally people from western cultures live in a individualistic
society
- Collectivistic: importance placed on the group rather than their own
individual needs
- Generally eastern cultures
Yuki = people from individualistic and collectivistic groups judge emotions differently //
collectivists focus more on the mouth while individualistic focus on the eyes
c. Temporal (time)
Making generalisations:
Generalisation focuses on the similarities between humans and take a nomothetic
approach (one that focuses on the common features shared by human beings) to
psychology rather than an idiographic approach (focusing on what makes each of
us unique)
A benefit is that we can predict how people are likely to behave in a particular situation.
Sampling Methods
The target population is the group of people the researcher wants
to
study. They cannot study everyone thus they have to select a
sample.
- A sample is a small group of people who represent the target
population and who are studied
- It is important the sample is representative of the target
population
How do we obtain our sample?
Random sampling: A sampling technique in which every member of the
target population has an equal chance of being chosen
How to do it:
1. You need a sampling frame which is a complete list of all members of the
target population is obtained
2. All of the names on the list are assigned a number
3. The sample is selected randomly for example, a computer-based randomiser
or picking names from a hat
Advantages Disadvantages
- There is no bias with this - Impractical: It takes more time and effort than o
method. Every person sampling methods because you need to obtain a list
in
the members of the target population, identify the sa
the target population has
and then contact the people identified and ask if the
an equal chance of being
[Link] [Link] [Link]
selected. Therefore the take part and not all members may wish to take par
sample is more likely to Not completely representative: Unbiased selection d
be representative. not guarantee an unbiased sample, for example, the ra
selection may only end up generating an all-female sam
making the sample unrepresentative and therefore not
generalisable.
Opportunity sampling: A technique that involves recruiting anyone who happens to
be available at the time of your study.
How to do it:
1. The researcher will go somewhere they are likely to find their target
population and ask them to take part
Advantages Disadvantages
- Simple, quick, easy and cheap as you - Unrepresentative: The sample is lik
are just using the first participants that be biased by excluding certain type
you find. participants which means that we
- Useful for naturalistic experiments as cannot confidently generalise th
the researcher has no control over who findings e.g. an opportunity sampl
is being studied. collected in town during the day on
week day would not include those
are at work or college
Volunteer sampling: When people actively volunteer to be in a study by responding
to a request which has been advertised by the researcher (they are self selecting).
The researcher may then select only those who are suitable for the study.
How to do it:
1. Participants self select by responding to the advert
Advantages DIsadvantages
- Most convenient and economical way to - Sample bias: Particular people (with
gather a wide range of people with higher levels of motivation and wh
particular requirements for a study have more time) are more likely to
compared to a random sample as they volunteer so may be harder to
have already agreed to participate. generalise. What is it
- Can reach a wide audience, especially that has made them decide to take
online. part? This may lead to a bias as th
may all
display similar characteristics.
Ethics
Ethics concerns questions of right and wrong. Issues arise in research if there
are conflicting values between the researcher and the participants. A cost vs
benefit analysis is taken.
Ethical Guidelines
- Informed consent
- Revealing the true aims of the study
- Participants must be aware of what they are needed to do as part of the
study in order to give valid consent
[Link] [Link] [Link]
- If the study involves children, parental consent must be obtained
- Deception
- Deceptions means deliberately misleading or withholding information
- Deceiving participants must be kept to a minimum
- Withholding details in order to avoid influencing behaviour is acceptable
but deliberately providing false information cannot happen
- If telling the truth will not have an effect on results, participant
must be informed - Confidentiality
- Communication of personal information from one person to another and
this trust must be protected
- Psychologists need to be sure that the information they publish
will not allow their participants to be identified
- Anonymity cannot be compromised
- Debriefing
- If consent cannot be obtained (e.g. field experiment) participants must
be fully debriefed afterwards
- Involves telling them about the experiment and then giving them the
option of withdrawing their information
- Right to withdraw
- Participants must be made aware of this when they sign the consent form
- Privacy/ anonymity
- A person’s right to control the flow of information about themselves
- We have a right to privacy; if this is broken then confidentiality
must be maintained - Protection from harm
- Participants should be no worse off when they leave the experiment
compared to when they arrived
- Risk is considered acceptable if it is no greater than what would
be experienced in everyday life
- Psychological and physical harm is included
Types of Data
Qualitative research: Gathers information that is not in numerical form, often
describing thoughts, feelings and opinions.
- Rich in detail and might include a reason as to why the behaviour occurred
- Subjective
- Interpreting something
- Rich insight as to why the behaviour occurred - trying to see the world
through someone else's eyes (their reality)
- The aim is not to establish a cause-and-effect relationship
Strengths:
1. Rich insight and understanding of the value/ issue
2. Helps explain the why of a phenomena
3. Less reductionist than quantitative data
Limitations:
1. More open to researcher bias (as it is subjective thus more open to
interpretation)
Quantitative research: Numerical data that can be statistically analysed
- Does not include a reason or explanation (the why) for the numerical answer given - this is
the what.
[Link] [Link] [Link]
- Objective and scientific
Strengths:
1. Allows for easier comparison and analysis of data (e.g. means)
2. Objective and scientific
3. Less chance of researcher bias
Limitations:
1. Can't tell us the why something is happening
2. View as reductionist as complex ideas are reduced to numbers
Qualitative and quantitative data are not always mutually exclusive and
whether the experiment generates qualitative data, quantitative data or both
depends upon the research question.
Data Handling/ Analysis
1. Descriptive statistics: The use of graphs, tables and summary statistics to
identify trends and analyse sets of data (comes first)
- Suggests something about the hypothesis
2. Inferential statistics: This is where a statistical test is calculated in order to
draw conclusions about hypotheses (note that inferential statistics are
not on the syllabus).
- Eg. 0.05 is the alpha value that is statistically significant
- Concludes hypothesis, statistical significance or not?
Averages: Measures of Central Tendency
- Mean:
Mathematical
average of a
set of scores
- Median: Middle
score in a
data set
- Mode: The value
that
occurs most
frequently in a
data set
- Only used with
categorical data
Measures of Dispersion
- The Range tell us whether
a set
of data are close
together or
spread out
- Wider range: Large
dispersion could
be due to
individual
differences
- Standard deviation: Standard
deviation is a single value that
tells us how far scores deviate
(move away) from the mean. The
larger the SD, the greater the
dispersion/spread within a set of
data.
- A low SD = good, therefore more reliable as a measure
- A large SD = not all participants affected by the IV - individual differences
[Link] [Link] [Link]
Graphs and Charts
Discrete data is information that can only take certain values. This type of data is often
represented using tally charts, bar charts or pie charts.
Continuous data is data that can take any value. Height, weight, temperature and length are
all examples of continuous data.
Ordinal data: Data can be put in order // rankings
Interval data: Temperature, length, time, etc // agreed value
Distribution Curves - Inferential Statistics
Normal distribution - bell-shaped curve
[Link]
edu
[Link] [Link]
[Link]
edu
[Link] [Link]