Detailed Notes

AQA Psychology A-level
Topic 7: Research Methods

Detailed Notes
This work by PMT Education is licensed under https://bit.ly/pmt-cc

https://bit.ly/pmt-edu-cc CC BY-NC-ND 4.0
https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
Experimental Method
The experimental method concerns the manipulation of an independent variable (IV) to have
an effect on the dependent variable (DV), which is measured and stated in results. These
experiments can be: field, laboratory, quasi or natural.
Aims
An aim is a general statement made by the researcher which tells us what they plan on
investigating, the purpose of their study. Aims are developed from theories and develop from
reading about other similar research.
Hypotheses
A hypothesis is a precise statement which clearly states the relationship between the
variables being investigated. The hypothesis can either be non-directional or directional. A
directional hypothesis states the direction of the relationship that will be shown between the
variables whilst a non-directional hypothesis does not.
E.g. If a researcher is carrying out a study to investigate whether sleep helps memory
performance:
● A directional hypothesis for this would be - “The more sleep a participant has the better
their memory performance.”
● A non-directional hypothesis would be - “The difference in the amount of hours of sleep a
participant has will have an effect on their memory performance, which will be shown by the
difference in the memory test scores of the participants.”
A directional hypothesis tends to be used when there has already been a range of research carried
out which relates to the aim of the researcher’s investigation. The data from this previous research
would suggest a particular outcome. However if there has been no previous research carried out
which relates to the study’s aim or the research is contradictory than a non-directional hypothesis is
appropriate.
Independent and dependent variables

The independent variable refers to the aspect of the experiment which has been manipulated
by the researcher or simply changes naturally to have an effect on the DV which is then
measured. The dependent variable is the aspect of the study which is measured by the
researcher and has been caused by a change to the IV. All other variables that could affect the
DV should be carefully controlled so that the researcher is able to confidently conclude that the
effect on the DV was caused by only the IV.
In order to properly test the effect of the IV we need different conditions: the experimental condition
and the control condition. You can have various experimental conditions which will allow you to
compare the effects of different levels of the IV.
Operationalisation of variables
Operationalisation refers to the act of a researcher clearly defining the variables in terms of
how they are being measured. This means the variables should be defined and measurable. The
hypotheses states should also show this operationalisation e.g. the aforementioned directional
hypothesis would be even better if operationalised:
“Participants that get at least four hours of sleep will show better performances on the
memory test, shown by them achieving higher scores than the participants that got less
than four hours of sleep.” It could even be further operationalised when more details of the
investigation are given, such as the number of questions in the test, hence the maximum score a
participant can achieve.
Control of Variables
Extraneous variables and confounding variables
In an experiment, the only aspect that should affect the DV is the IV. Any other variables that may
interfere with the IV or the DV should be removed from the experiment or well controlled. Such
variables can be confounding or extraneous. An extraneous variable refers to any other
variable which is not the IV that affects the DV and does not vary systematically with the IV,
they are essentially nuisance variables. Examples are the lighting in the lab or the age of
participants - these variables do not confound the results of a study but just make them harder to
detect.
A confounding variable is also described as a variable other than the IV which has an effect on
the DV. Unlike the extraneous variable, confounding variables do change systematically with the
IV. With these variables it becomes difficult for the researcher to be sure of the origin of the
impact of the DV as the confounding variable (not the IV) could have been the cause. An example
for the aforementioned sleep study would be time of day the experimental task is done - those who
complete the memory test later in the day may be more tired and therefore do worse, obscuring the
true relationship between lack of sleep and memory performance. Therefore, potential confounding
variables must be identified and controlled; in this case the participants should take the test at the
same time of day.
Demand characteristics and Investigator effects
Demand characteristics refer to any cue the researcher or the research situation may give
which makes the participant feel like they can guess the aim of the investigation . This can
cause the participant to act differently within the research situation from how they would usually
act. This is as participants from the start of the experiment are trying to figure out what's going on
in this new situation they find themselves in - this is known as participant reactivity. They may
change their behaviour to fit the situation rather than acting naturally.
They may act in a way they think the researcher wants them to which is known as the ‘Please-U
effect’ or they may intentionally underperform to sabotage the study’s results, the ‘screw-U
effect’. This unnatural behaviour then affects the validity of the results, hence demand
characteristics provides a problem for research.
Participant reactivity may also lead to investigator effects which refers to any unwanted
influence from the researcher’s behaviour, either conscious or unconscious, on the DV
measured (the research’s results). This includes a variety of factors :- the design of the study,
the selection of participants and the interaction with each participant during the research
investigation.
Randomisation and Standardisation
To minimise the effects of extraneous or confounding variables different steps can be taken by the
researcher like randomisation and standardisation. Randomisation is the use of chance to
reduce the effects of bias from investigator effects. This can be done for the design of
materials, deciding the order of conditions, the selection of participants e.t.c.
Standardisation describes using the exact same formalised procedures and instructions for
every single participant involved in the research process. This allows there to eliminate
non-standardised instructions as being possible extraneous variables.
Experimental Method: Types of Experiment
Design Description Strengths Limitations
Laboratory An experiment that High degree of control- Experimenter’s bias- this

takes place in a experimenters control all bias can affect results and
special environment variables,the IV has been participants may be
whereby different precisely replicated, influenced by these
variables can be leading to greater expectations.
carefully controlled. accuracy. Low ecological validity-
Replication - researchers high degree of control
can repeat experiments makes the situation artificial,
and check results. unlike real life.
Field An experiment Naturalistic - so more Ethical considerations-

conducted in a more natural behaviours hence invasion of privacy and
natural environment, high ecological validity. likely to have been no
not in a lab but with Controlled IV informed consent.
variables still being Loss of control- over
well controlled. extraneous variables hence
precise replication not
possible.
Quasi An experiment Controlled conditions- Cannot randomly allocate

whereby the IV has hance replicable, likely to participants- to conditions
not been determined have high internal validity. so there may be
by the researcher, confounding variables
instead it naturally presented. This makes it
exists e.g gender harder to conclude that the
difference studies. IV caused the DV.
Natural An experiment in Provides opportunities- Natural occurring events-

which the IV is not for research that would may be rare this means
brought about by the have otherwise been these experiments are not
researcher hence impossible due to likely to be replicable hence
would have happened practical or ethical hard to generalise findings.
even if the researcher reasons. Very difficult to
had not been there High external validity- as randomise- participants
e.g. if studying you are dealing with real into groups so confounding
reactions to life issues. & extraneous variables
earthquakes. become a problem.
Sampling
The researcher needs to decide how they select participants to take part in their investigation.The
population is a group of people from whom the sample is drawn.
E.g. If the sample of participants are taken from the sixth formers going to schools in
London, the findings of the study can only be applied for that certain group of people and
not all the sixth formers in the UK.
There are various methods that a researcher can use to select participants:
Sampling Method Explanation Strengths Limitations
Opportunity Participants happen to Easy method of Not representative of

sampling be available at the time recruitment which is the whole population
which the study is being time saving and less hence lacks
carried out so are costly. generalisability.
recruited Researcher bias is
conveniently. presented as they
control who they want
to select.
Random sampling This is when all No researcher bias- Time consuming-

members of the researcher has no need to have a list of
population have the influence of who is members of the
same equal chances picked. population (sampling
of being the one that frame) and then
is selected. The contacting them takes
method used is :- each time.
member of the Volunteer bias-
population is assigned participants can
a number then either a refuse to take part so
random number table can end up with an
or a random number unrepresentative
generator or the sample.
lottery method is used
to randomly choose a
partner.
Systematic A predetermined Avoids researcher Not truly unbiased

sampling system is used bias and usually unless you use a
whereby every nth fairly representative random number
member is selected of population. generator and then
from the sampling start the systematic
frame. This numerical sample.
selection is applied
consistently.
Stratified sampling With this method the No researcher bias- Time consuming to
composition of the the selection within identify strata and
sample reflects the each stratum is done contact people from
varying proportions randomly. each.
of people in particular Produces A complete
subgroups (strata) representative data representation of the
within the wider due to the target population is
population. Firstly you proportional strata not possible as the
identify strat. Then you hence generalisation identified strata
calculate the required is possible. cannot reflect all the
proportion needed for differences between
each stratum based on the people of the
the target population. wider population.
Then select sample at
random from each
stratum using a random
selection method.
Volunteer sampling Involves self selection Quick access to Volunteer bias- they
whereby the participant willing participants study may attract a
offers to take part either which makes it easy particular profile of a
in response to an and not time person. This means
advert or when asked consuming. generalisability is then
to. As participants are affected.
willing to take part Motivations like
they are more likely money could be
to cooperate in the driving participation so
study. participants may not
take study seriously,
influencing the results.
Experimental Design
Design Description Strengths Limitations Solution
Independent The participants -There are no - No control over Random

groups only perform in order effects participant allocation solves
design one condition of presented. variables the first limitation
the independent - Participants are whereby different mentioned.This is
variable (IV). less likely to abilities of as it ensures that
guess the aims of participants in the each participant
the study various conditions has the same
(demand can cause chance of being in
characteristics changes to the one condition of
are eliminated). DV.. the IV as another.
-You need more
participants than
other designs to
gather the same
amount of data.
Repeated The same - Eliminates - Order effects Counterbalancing

measures participants take participant presented e.g. - this is when half
part in all variables. boredom may of the participants
conditions of the - Fewer mean in second do conditions in
IV. participants condition done one order and the
needed, so not participant does other half do it in
as time not do as well on an opposite order.
consuming task.
finding and using
them.
Matched Pairs of - No order - Time

pairs participants are effects. consuming and
first matched on - Demand expensive to
some variable characteristics match
that has been are less of a participants.
found to affect the problem. - A large pool of
dependent potential
variable (DV), participants is
then one member needed which
of each pair does can be hard to
one condition get.
and the other - Difficult to know
does another. which variables
are appropriate
for the
participants to be
matched.
Pilot Studies
A pilot study is a small-scale version of an investigation which is done before the real
investigation is undertaken. They are carried out to allow potential problems of the study to be
identified and the procedure to be modified to deal with these. This also allows money and time to
be saved in the long run.
Single-Blind and Double-Blind Procedures

Single-blind procedure
A research method in which the researchers do not tell the participants if they are being
given a test treatment or a control treatment. This is done in order to ensure that participants do
not bias the results by acting in ways they “think” they should act-avoids demand characteristics.
Double-blind procedure
A research procedure in which neither the participants nor the experimenter knows who is
receiving a particular treatment. This procedure is utilised to prevent bias in research results.
Double blind studies are particularly useful for preventing bias due to demand characteristics or the
placebo effect. Gives a way to reduce the investigator effects as the investigator is unable to
unconsciously give participants clues as to which condition they are in.
Control group/condition - sets a baseline whereby results from the experimental condition
can be compared to results from this one. If there is a significantly greater change in the
experimental group compared to the control than the researcher is able to conclude that the cause
of effect was the IV.
Observational Techniques
Type of observation and Strengths Limitations

description
Naturalistic- watching and - High ecological validity - Low ecological validity if

recording behaviour in the -High external validity as participants become aware
setting where it would done in a natural environment that the are being watched.
normally take place. - Replication can be difficult.
- Uncontrolled confounding
and extraneous variables are
presented.
Controlled- Watching and - Researcher is able to focus - More likely to be observing

recording behaviour in a on a particular aspect of unnatural behaviour as
structured environment e.g. behaviour. takes place in an unnatural
lab setting. - There is more control over environment.
extraneous and confounding - Low mundane realism so
variables low ecological validity.
- Easy replication. - Demand characteristics
presented.
Overt- participants are - Ethically acceptable as - More likely to be recording

watched and their behaviour is informed consent is given. unnatural behaviour as
recorded with them knowing participants know they are
they are being watched. being watched.
-Demand characteristics likely
which reduces validity of
findings.
Covert- the participants are -Natural behaviour recorded - Ethical issues presented
unaware that their hence high internal validity of as no informed consent given.
behaviour is being watched results. Also could be invading the
and recorded. -removes problem of privacy of the participants.
participant reactivity whereby
participants try to make sense
of the situation they are in,
which makes them more likely
to guess the aim of the study.
Participant- The researcher - Can be more insightful -There's always the possibility
who is observing is part of which increases the validity of that behaviour may change if
the group that is being the findings. the participants were to find
observed. out they are being watched.
- Researcher may lose
objectivity as may start to
identify too strongly with the
participants.
Non-participant- The - Researcher can be more - Open to observer bias for

researcher observes from a objective as less likely to example of stereotypes the
distance so is not part of identify with participants since observer is aware of.
the group being observed. watching from outside of the - Researchers may lose
group. some valuable insight.
Observational Designs
One problem with carrying out observations is that observer bias is easily presented. This is when
an observer’s reports are biased by what they expect to see. A solution to this problem is
checking the inter observer reliability of the observation. This is done by many researchers
conducting the observational study , their reports are then compared and a score calculated using
the formula :-
Total number of agreements / total number of observations x 100 .
The score that shows high inter observer reliability is any score above 80%.
There are different types of observational designs and each has their strengths and weaknesses :-
Design and description Strengths Limitations
Unstructured- consists of - More richness and depth of - Produces qualitative data

continuous recording where detail. which is more difficult to
the researcher writes record & analyse.
everything they see during the - Greater risk of observer
observation bias e.g. only record ‘catch
the eye’ behaviours.
Structured- Here the - Easier as is more - Not much depth of detail.

researcher quantifies what systematic. - Difficult to achieve high inter
they are observing using - Quantitative data is collected observer reliability as filling
predetermined list of which is easy to analyse and the predetermined lists in is
behaviours and sampling compare with other data. subjective.
methods. - There is less risk of
observer bias.
Whilst conducting structured observations, behavioural categories can be used. This is when a
target behaviour which is being observed is broken up into more precise components
which are observable and measurable e.g. aggressive behaviour can be broken down to -
shouting, punching, swearing etc. When forming a behavioural categories list, it is important to
make sure that behaviours do not overlap with other behaviours, so very similar behaviours should
not be listed e.g. grin and smile. They should be clearly operationalised. During structured
interviews there are different types of sampling methods:
Method and description Strengths Limitations
Time sampling- this is the - It reduces the number of - The small amount of data
recording of behaviour within observations that has to that you collect within that
a timeframe that is made so it is less time time frame ends up being
pre-established before the consuming. unrepresentative of the
observational study. observation as a whole.
Event sampling- this involves - It is good for infrequent - If complex behaviour is being
the counting of the number behaviours that are likely to observed, important details
of times a particular be missed if time sampling of the behaviour may be
behaviour is carried out by was used. overlooked by the observer.
the target group or individual - If the behaviour is very
you are watching. frequent, there could be
counting errors.
- It is difficult to judge the
beginning and ending of a
behaviour.
Correlations
A correlation is a mathematical technique that is used to investigate an association between
two variables which are called co-variables. Correlations differ to experiments as :-
● The variables are simply measured, not manipulated like in experiments.
● Only an association is found, no cause-and-effect relationship found hence the terms
DV and IV are not used.
During correlational studies correlation coefficients are calculated. This value determines the
strength and the relationship between two variables. This doesn’t necessarily mean that one
variable is causing another, but that there is a relationship of some sort.
There are various relationships which can be shown between the co-variables :-
● Negative correlation - when one variable increases the other decreases. When the data is
presented on a scattergram the line of best fit has a negative gradient. It has a correlation
coefficient of less than 0.
● Positive correlation - when one variable increases the other also increases. When the
data is presented on a scattergram the line of best fit has a positive gradient.It has a
correlation coefficient of more than 0.
● Zero correlation - no relationship is found between the co-variables. When the data i s
presented on a scattergram, no line of best fit can be drawn as the points on the
scattergram are random. It has a correlation coefficient equal to 0.
Image Source
● Curvilinear relationship- as one variable increases, so does the other but only up to a
certain point after which as one variable continues to increase the other begins to
decrease. Ona graph this forms an inverted U shape. An example of such a relationship is
shown by the Yerkes-Dodson Law from the topic of Memory which shows how anxiety
affects eyewitness testimony.
Just as you have hypotheses for experiments researchers also state hypothesis for correlational
studies. A directional hypothesis states whether there will be a negative or positive
correlation between the co-variables being studies whilst a non-directional hypothesis only
states there will be a correlation but the type is unknown.
Strengths and Limitations of Correlations:
Strengths Limitations
- They can be used as starting points to - It is difficult to establish a cause and effect
assess patterns between co-variables before relationship, really only an association is
committing to conducting an experimental found.
study. -The third variable problem is presented -
- Quick and economical to carry out. this is when there is a chance that there is
- Secondary data can be used in the another variable, a third variable which the
correlational study which makes it even less researcher is unaware of that is
time consuming. responsible for the relationship between
the co-variables.
- Lastly, correlations tend to be misused or
misinterpreted especially when made public
by the media - correlation is often presented as
causation.
Data Analysis: Types of Data
Type and description Strengths Limitations
Qualitative data- data which - More richness and depth of - Difficult to analyse.
is displayed in words , is detail. - Difficult to make
non-numerical. - Allows participants to further comparisons with other data.
develop their opinions hence - Researcher bias presented
has greater external validity. as conclusions rely on the
- A more meaningful insight subjective interpretations of
into the participants’ views is the researcher (interpretative
achieved. bias).
Quantitative data- data that - Can be analysed statistically - Lack of depth in detail.
is displayed numerically, not so converted to graphs or - No meaningful insight into
in words. charts. participants' views.
- This makes it easy to make - As participants are not able
comparisons with other data. to develop their opinions the
results have low external
validity.
Primary data - this is when -Targets the exact information - Requires time and effort.
information is obtained first which the researcher needs, - Can be expensive.
hand by the researcher for so the data fits their aims and
an investigation. objectives.
Secondary data - this is when - Expensive - It may be likely that the data
information is collected by - Data is accessed so requires is outdated or incomplete.
someone else other than the minimal effort to collect. - The data may not be
researcher yet is used by the reliable- the researcher was
researcher for their not there when the study was
investigation. Also known as conducted so is likely to be
‘desk research’. unsure of the validity of the
results.
Meta-analysis - this is when a - More generalisability is - Publication bias such as the

researcher combines results possible as a larger amount of file drawer problem may be
from many different studies data is studied. presented- this is when the
and uses all the data to form - The researcher is able to researcher intentionally does
an overall view of the subject view the evidence with more not publish all the data from
they are investigating. confidence as there is a lot of the relevant studies but
it. instead chooses to leave out
the negative results. This
gives a false representation of
what the researcher was
investigating.
Data Analysis: Descriptive Statistics

Descriptive statistics are the use of tables, graphs, and summary statistics to analyse data.
Measures of central tendency

These measures refer to any measure which calculates an average value within a set of data.
Measure Calculation method Strengths Limitations
Mean- arithmetic Total of all values in a - Makes use of all - It is influenced by

average. set of data is divided values. outliers (extreme
by the number of - Good for interval scores) so it can be
values. data. unrepresentative.
Median Arrange data from - Not affected by - Not as sensitive as

lowest to highest then extreme scores. mean, does not use
find the central value. - Good for ordinal all data.
data.
Mode The most frequently - Useful for nominal - Is not useful when
occurring value in a data (data in there are several
set of data. categories). modes.
Measures of dispersion
These measures refer to any measure that calculates the variation in a set of data.
Measure Calculation method Strengths Limitations
Range Minus the lowest - Easy to calculate. - Affected by extreme

score from the highest values.
score. - Does not use all
data.
Standard Deviation The square root of the - Precise measure - Difficult to calculate.
(SD) variance calculates where all data values - Affected by extreme
SD. A low SD means are taken into values.
that more data is account.
clustered close to the
mean hence there is
less data spread
Presentation and Display of Quantitative Data

There are various ways of representing data:
Summarising data in a table
One of these ways is summarising data in a table. This is usually not in the form of raw scores but
the data has been converted into descriptive statistics for example of the form below :-
Table showing the mean and mode of scores of a memory test
Condition A Condition B
Mean 35 67
Mode 30 34
Below the table there is usually a description of what the table’s data means.
Bar Charts
This way of representing data allows for differences in data to be seen more clearly. They are used
for discrete data, which describes data that has been divided into categories. The bars do not
touch each other which shows that we are dealing with separate conditions.The amount of
frequency for each category is plotted on the y-axis (vertical axis) whilst the categories (below
these are condition A and B) are plotted on the x-axis (horizontal axis).
Histograms
In this form, the bars touch each other unlike in bar charts and this represents that we are
dealing with continuous data rather than discrete. Therefore the x-axis has equal sized
intervals of one category (e.g. scores of an english test in intervals 0-10, 11-21, 22-32, etc.) whilst
the y-axis represents the frequency (the number of people that score each mark).
Line graphs
This form also represents continuous data , whereby points are connected by lines to show the
change of values. As per usual, the IV is plotted on the x-axis while the DV is plotted on the y-axis
Scattergrams
These are used to show associations between co-variables rather than differences hence we
came across them in the correlations topic. Either of the co-variables can occupy the x-axis or
the y-axis, and each point displayed on the graph coincides with the x and y position of the
co-variables.
Distributions
Normal distribution is a symmetrical pattern of frequency data that forms a bell-shaped
pattern.
A skewed distribution is a spread of frequency data that is not symmetrical, instead the data
all clusters to one end. There are two types of these :-
● Positive skew whereby most of the distribution of data is concentrated on the right.
● Negative skew whereby most of the data distribution is concentrated on the left.
Peer Review
AO1
Peer review is the assessment of scientific work by experts in the same field, it is done to
make sure that all research intended to eventually be published is of high quality.
Main purposes of peer review:
● To know which research is worthwhile hence funding can be allocated to it.

● To validate the relevance and quality of research. This is important to prevent fraudulent
research from being released to the public.
● To suggest possible improvements or amendments to the research study.
AO3
● Anonymity is a problem; reviewers sometimes use it to settle old scores or bury rivals,
especially if they're competing for funds. This means that anonymity affects the
objectivity of reviewers. Due to this, some journals have started doing open reviewing to
avoid this problem.
● There is publication bias involved in peer review. Editors tend to prefer to publish
‘headline grabbing’ findings and positive results. This brings about the file drawer
problem whereby negative results are intentionally not published. All this causes there to be
a misconception of the current state of psychology.
● It can be difficult to find an expert. Smith (1999) argues that because of this a lot of poor
research is passed as the reviewer didn't really understand the work.
● In peer review, any research that opposes mainstream theories tends to be

suppressed. This means that established scientists’ work is more likely to be published
and the new and challenging ideas are usually rejected. This means that the rate of
change in scientific fields is slowed down.
● Fraudulent research can be long-lasting which is a large problem. An example of this

is when the MMR vaccine link to autism was found by Andrew Wakefield (1998). This
finding had implications as caused the number of measles cases to increase. It was later
found that the research was fraudulent but even now there are still many people aware of
these said risks and still anti-vaccine for MMR.
The Implications of Psychological Research for the Economy
The implications that research has refers to how what we learn from psychological research
influences our country’s economic prosperity. The economy is the state of the region's
activities of producing or consuming goods & services. Absence from work costs the
economy an estimate of 15 billion pounds a year and this absence is mainly due to mental
illness e.g. stress, anxiety. For such problems, psychology research has been able to present
solutions to them and this expresses why psychology research is important for the economy.
From the various AS and A2 topics we have learnt , research in these topics has had implications
for the economy:
Topic Links to specific areas Economy
Psychopathology Treatments - Cognitive - Workers able to return to

Behavioural Therapy and work.
Rational Emotive Behavioural
Therapy for depression, drug
therapy for OCD.
Attachment Role of the father - Tiffany - Mothers can return to work.

Field (1978) found that fathers - More flexible working
can take on the role of being a arrangements within families.
primary caregiver. - Can maximise their income
and effectively contribute to
the economy.
Social Influence Social influence leading to - Health campaigns.

social change - Minority - Unions strike- make working
influence, appealing to NSI, conditions better.
disobedient models. - Environmental campaigns-
like getting companies to
reduce their waste and use of
non-renewable energy.
Memory Eyewitness testimony - How - Led to police using the

leading questions or post cognitive interview which
event discussion can affect reduces wrongful convictions
eyewitness testimony. hence reduces waste of
money and space in jail.
Case Studies
A case study is a detailed study into the life of a person which covers great detail into their
background. It looks at the past and present behaviour of an individual to build up a case history
hence provides qualitative data.
Examples from psychology topics we have learnt :-
● The case study of HM from the memory topic. HM knew how to tie shoelaces but
couldn't remember stroking a dog. This showed us his procedural memory was intact but
his episodic memory wasn’t. This showed us that there are different types of long term
memory and these are stored in different parts of the brain.
● Little Hans case study from the approaches topic, in particular the psychodynamic
approach. Freud used his case study as evidence for the Oedipus complex.
- Detailed so able to gain in depth insight. - Not generalisable to wider populations as

- Forms basis for future research. data is only gathered from one person.
- From studying unusual cases you are able to - Various interviewer biases are presented like
infer things about normal usual behaviour of social desirability bias (from the unique
humans. person’s side) and interpretative bias ( from
- Permits investigation of situations that would the researcher’s side).
be otherwise unethical or impractical. - Retrospective studies may rely on memory
which can be inaccurate.
- They are time consuming and difficult to
replicate.
Content Analysis
Conducting a content analysis involves studying human behaviour indirectly by studying
things that we produce e.g. TV adverts, newspapers. This allows us to have insight into the
structured values, beliefs and prejudices of our society.
How to conduct a content analysis :
● Identify hypothesis that you will investigate.

● Create a coding system depending on what you are investigating e.g. 1= male, 2= female.
● Gather resources.
● Conduct content analysis and record data in a table.
● Analyse data which is descriptive and qualitative e.g. using ‘thematic analysis’- allows
themes, patterns and trends to emerge in data.
● Write up a report in the format of a scientific report.
- Strong external validity as the data is already - Observer bias is presented but it can be
in the real world so it has high mundane eliminated by achieving inter-observer
realism. reliability.
- Produces large data set of both quantitative - Content of choice to analyse can be biased
and qualitative data that is easy to analyse. by researcher.
- Easy replication. - Interpretative bias - the researcher may
- Ethical issues like ‘right of privacy, ignore some things but pay extra attention to
confidentiality, informed consent’ are avoided others.
as data is already in the public domain.
Levels of Measurement
Quantitative data can be divided into different levels of measurement, either - nominal, ordinal or
interval.
Nominal data refers to a type of data that is in the form of categories. It is discrete- one item
can only appear in one category. It does not enable sensitive analysis as it does not yield a
numerical result for each participant.
Ordinal data refers to data which is represented in a ranking form e.g. 1= hates maths,
10= loves maths. There are no equal intervals between each unit. A weakness of it is that it lacks
precision as is based on the subjective opinion of people.
Interval data refers to the type of data that is based on numerical scales which include equal
units of precisely defined size. This is the most sophisticated form of data as it is based on
objective measures. It is needed for the use of a parametric test.
Appropriate measures for each level of data:
Level of data Measures of central Measures of dispersion

tendency
Nominal Mode n/a
Ordinal Median Range
Interval Mean Standard Deviation
Reporting Psychological Investigations: Scientific Report
Psychologists use a particular format to write up their research for publication which is known as a
scientific report. A scientific report consists of various sections:
● Abstract - this part includes a summary of all the key details of the research report.
These key details include the aim, hypothesis, method, results and conclusion. It is usually
about 150-200 words long and is the part that is supposed to be read to know whether the
research study is worth examining any further.
● Introduction - This includes information of past research on a similar topic whereby
relevant theories, studies and concepts are mentioned. At the beginning it tends to be
broad but as it continues towards the end the information becomes more specific until the
aims and hypotheses of the study are presented.
● Method - This part includes a description of what the researchers exactly did when
they undertook the study. This includes the design, sample collected (specific details e.g.
target population, sampling method, demographic data of participants), materials used ,
procedure (specific e.g. standardised instructions for each participant), ethics etc. There
should be sufficient detail included so that any other person is able to read this part
of the report and replicate the investigation precisely.
● Results - This includes all the findings from the study, presented even with inferential
and descriptive statistics. If qualitative data is collected then this section may include a
thematic analysis.
● Discussion - This is where the researcher considers what the findings exactly mean for
us and for psychological theories. Usually the findings are summarised here then they
are discussed in context to the introduction. Limitations thus ways of improving the study
and the wider implications it may have for society may also be discussed here.
● Referencing - This is the last part of the scientific report which is basically a list of all the
sources that were quoted or referred to in the report. These can vary - journal articles,
websites, books- and full details are given so that a reader is able to find the exact source
the researcher was referring to.
Books are referenced differently from journals:

● Books are referenced in this order: author(s), date, title of book (in italics or
underlined), place of publication, publisher.
E.g. Flanagan, C. and Berry, D.(2016). A level Psychology Year 2. Chettenham:
Illuminate Publishing.
● Journals are referenced in this order: author(s), date, title of article, journal
name, journal volume, issue number (all appear in italics or underlined apart
from author, date and title of journal article), page range.
E.g. Gupta,S.(1991). Effects of time of day and personality on intelligence test
score. Personality and individual differences, 12 (11). 1227-1231.
Introduction to Statistical Testing: Use of the Sign Test
Statistical testing provides a way of determining whether hypotheses should be rejected or

accepted. It can tell us whether differences or relationships between variables that have been
found during experiments are statistically significant or if they have only occurred due to
chance.
An example of a statistical test is the sign test. A sign test can only be used for a study that :-
● Looked for a difference not an association.
● Used a related experimental design- repeated measures design.
● Collected nominal data.
How to conduct a sign test:
● Step 1 - State the hypotheses- this includes both the alternative and the null hypothesis.
● Step 2 - Record data and work out the sign. For example, the sign will be negative (-) if the
value has decreased in the second condition but positive (+) if it has increased. If the value
has stayed the same , this value will be ignored and the N adjusted to exclude it.
● Step 3- Find the calculated value for the sign test, S, which is the number of times the less
frequent sign occurs.
● Step 4 - Find the critical value of S - use the calculated N value (which is the total number
of values with the ignored values excluded) and p≤ 0.05 which means there's a less than
5% probability that the results occurred by chance.
If S ≤ critical value- reject the null hypothesis, there is a significant difference.
If S ≥ critical value - accept the null hypothesis, there is no significant difference.
● Step 5 - State conclusion whereby you refer back to the hypothesis mentioning the IV and
Dv and support your conclusion with the exact values of -the critical value, S, N and what p
value you used.
Choosing an Inferential Statistical Test
When choosing an inferential statistical test you have to think about three factors:
● The design of the study - Did it involve an unrelated design which is of the independent
groups design? Did it involve a related design which could be either using the repeated
measures or matched pair experimental design?
● The level of data collected during the study - either ordinal, nominal or interval.
● Whether a difference or correlation is being measured
These factors have been summarised into the table below which shows which statistical tests to
use in different situations:
Type of data Test of difference Test of association/

correlation
Unrelated Related
Nominal Chi-square Sign test Chi-square
Ordinal Mann-whitney Wilcoxon Spearman’s Rho
Interval Unrelated t-test Related t-test Pearson’s R
Remembering the table above will help you answer exam questions which can either ask you why
a student may decide to use a particular test after providing you with a stem of a research study or
may ask you which statistical test is appropriate to use for that study.
Use of statistical tests
Statistical tests are used to determine whether a significant difference or correlation exists.
This is discovered using the calculated value (the result obtained from the statistical test) and the
critical value (the numerical boundary that stands between accepting or rejecting the null
hypothesis when a hypothesis is being tested). The critical value is worked out from a table of
probability values and depends on various factors: whether it was a one or two tailed test, the
P value and either the N value or the degrees of freedom value.
Rule of R - If there is an R in the name of the statistical test the calculated value has to be
gReater or equal to the critical value for the result to be significant. If this is the case then the
null hypothesis can be rejected and the alternative hypothesis is supported. If there is no R in the
test’s name then the calculated value has to be less than or equal to the critical value for it to be
significant.
Probability and Significance
Significance is a statistical term which lets us know how sure we are about a correlation or
difference existing. If significant, we reject the null hypothesis and accept the alternative
hypothesis. The difference between these two types of hypotheses is the null (H0) one states
‘there is no difference or correlation between the conditions’ whilst the alternative (H1) one
states ‘there is a difference between the conditions’.
Probability is a calculation of how likely it is for an event to happen - 0= statistical

impossibility and 1= statistical certainty. The usual level of significant in psychology is 0.05.
Therefore the p value is usually equal to or less than 0.05 (5%) which means that the
probability of the difference in the study’s findings being due to chance is 5% or less so
researchers have a 95% confidence level in their results. If there is any risk attached to the
research like a ‘human cost’ e.g. with clinical drug trials then the p values is set at 0.01 (1%)
instead.
Type I and Type II errors
When researchers conduct inferential statistical tests they can make either of two types of errors
when forming a conclusion from the test:-
● Type I (optimistic) error is the incorrect rejection of a null hypothesis which is

actually true. Researchers claim to have found a significant difference when there actually
isn't any (a false positive).
● Type II (pessimistic) error is the failure to reject the null hypothesis that is false.
Researchers claim that there is no significant difference when there actually is one (a false
negative).
Features of a Science
Paradigms & paradigm shifts
A paradigm is a set of shared ideas and assumptions within a scientific discipline. A

paradigm shift is a significant change in these central assumptions within a scientific
discipline, resulting from a scientific revolution. Kuhn (1962) suggests that paradigms are what
separate scientific disciplines from non-scientific disciplines. Kuhn also believes that paradigm
shifts show progress within a science. In respect to this feature, psychology has too much
disagreement and conflicting approaches (e.g.is behaviour biological from genes or from
conditioning experiences?) so isn't able to qualify as a science. It is referred to as a pre-science.
Theory construction & hypothesis testing
A theory refers to a set of general principles and laws which can be used to explain specific
events or behaviours. Theory construction takes place through gathering evidence from
direct observation during investigations. You should be able to make different hypotheses from
a theory, then when the hypothesis is supported the theory is strengthened. If it is not then the
theory may need to be revised. Deduction refers to the process of deriving new hypotheses
from an already existing theory e.g. Baddeley and Hitch modified the Working Memory Model in
2000 as they added the episodic buffer to the model.
Falsifiability
Falsifiability is the principle that states a theory cannot be considered scientific unless it
allows itself to be proven untrue. Popper (1934) argues that this is a key criterion for a scientific
theory. Popper proposed the theory of falsification which states that successful theories that
have been constantly tested and supported simply haven't been proven false yet. Sciences
that can't be proven wrong are known as ‘pseudosciences’- a good example is Freud’s concepts
from the psychodynamic approach like the Oedipus’s Complex.
Theories which survive more falsify attempts are seen as the strongest. This theory of falsification
explains why when stating hypotheses for an investigation the alternative hypothesis is always
accompanied by the null hypothesis. This also explains why we never use the word ‘proves’ in
investigations even if the results support the researcher’s hypothesis. The hypothesis-deductive
method refers to the process of formulating hypotheses that can either be proved or
disproved by experimentation.
Replicability
Replicability refers to the extent to which scientific methods and their results can be
repeated by other researchers across other contexts and circumstances. It is used to assess
validity and reliability of results from a research study.
Objectivity & the empirical method
Objectivity is when all possible biases from the researcher are minimised so that they don't
influence or distort the research process. The empirical method is when evidence is collected
through making direct observations and through direct experiences. A theory is not able to be
scientific unless it can be empirically tested and verified using either the empirical experimental or
observational method.
Psychology as a science:
Supporting arguments Against
- Produces intuitive results which are against - Experiment interpretations can be subjective.
common sense. - Not all research is generalisable e.g. from
- Scientific methods are used in many research case studies.
studies giving them scientific credibility. - Psychologists do often make inferences of
- Findings from studies do positively impact behaviour rather than directly measuring it , for
society & individuals e.g. Cognitive behavioural example this is usual for cognitive
Therapy to treat depression. psychologists that infer about cognitive
processes from brain scans (Memory topic link
here).
Ethical Issues & Ways of Dealing with Them
Issue Explanation Solution
Informed consent Participants must be told the There are various methods of
purpose of the investigation dealing with informed consent:
(their aims) and about any - Prior general consent-
potential risks they may be participants give permission to
subject to when taking part in take part in many studies
it. This allows them to make whereby one of them involves
an informed decision on deception so effectively they
whether they want to are consenting to getting
participate in the research deceived,
study. - Presumptive consent-
when a researcher gathers
Researchers don’t always opinions from a group like the
wish to disclose this participants in the study but
information as it could lead to does not inform the actual
demand characteristics being participants. Allows demand
presented hence result bias. characteristics to be
eliminated.
- Retrospective- this is when
the participants are asked for
consent after they have
participated in the study.
Deception This is the act of deliberately - Debriefing- all participants

withholding information from would be debriefed after the
participants or misleading study, it can be a written or
them during the research verbal debrief. During the
study. This is only seen as debrief the true nature of the
acceptable when the study must be said and the
participants knowing the true participants should be told
nature could guess the aims what their data will be used
of the investigation or when for. After the debrief
the deception will not cause participants have the right to
distress. choose to withhold or
withdraw their data.
Protection of harm Participants must be protected - If the participants have been
from physical and subject to any stress or
psychological harm. It is the psychological harm, the
job of the researcher to make researcher should provide
sure of this. counselling if it is required.
- A Cost-Benefit Analysis
All through the investigation, should be done before a study
participants are also reminded is carried out. This is done by
that they do have the right to the ethics committee whereby
withdraw, especially if the the pros and cons of the study
study is causing them harm. are weighed up to determine
whether the study will be
ethical. This can be difficult
and an example of where this
was done but went wrong is
for Zimabardo’s Stanford
Prison Experiment in 1973
(Social Influence topic).
Privacy and confidentiality Right of privacy refers to the - Anonymity can be

right that participants have to maintained. This is achieved
controlling information about by the researchers not
themselves- how much is recording any personal details
released and how it is used. It of their participants so that
can be difficult to avoid none of the results data can
invading a participant's be traced back to them.
privacy for example if it is a Instead the researchers can
field study- these are done in refer to the participants using
natural environments. The numbers or initials when
right of privacy can extend to writing up the investigation
the location of the study e.g. HM case study.
whereby the institution is not -The participant should be
named. reminded during both the
Confidentiality refers to the briefing and debriefing of the
right participants have which investigation that their data
concerns any personal data of will be protected
theirs being protected.
Self-Report Techniques & Design
Self-report techniques refer to any sort of method where a person is asked to give their opinions,
feelings, experiences and behaviours in relation to a particular topic. There are two types of these
non-experimental investigations:
● Questionnaires - These assess a person’s thoughts or experiences through a number
of different written questions.
● Interviews - This involves a live encounter where a set of questions is asked by an
interviewer to an interviewee to assess their thoughts or experiences.
Questionnaires
There are two types of questionnaires:
Open Question - This is - Rich in depth and detail. - Difficult to convert to

when the questions are - Useful for sensitive topics as statistical data hence more
phrased in a way that the participants can elaborate on difficult to analyse.
participant is free to answer their answers.
however they like, there are
no restrictions. This type
collects qualitative data.
Closed Question - In -Easy to analyse data and - Lack of depth and detail.
contrast, this type of compare with data from - Can be limiting which can be
questionnaire consists of elsewhere. frustrating for participants.
questions which restrict you to
a fixed number of responses.
This type collects quantitative
data.
Examples:
- Likert scale- the respondent
indicates agreement with a
statement, ranges from agree
to strongly agree.
- Rating scales- a rating
scale works in a similar way
but gets respondents to
identify a value that
represents their strength of
feeling about a particular
topic.
- Fixed choice scales- the
question includes a list of
possible options and
respondents are required to
indicate those that apply to
them.
There are various strengths and limitations of questionnaires:
- Cost-effective. - Difficult to know whether the target population

- Gathers large amounts of data quickly. it was intended for answered it e.g. if it is
- The researcher does not need to be present. online.
- They are easy to analyse. - They take a long time to design.
- As responses can be anonymous this usually - It is difficult to assess the validity as biases
means participants are more open. such as social desirability bias (when the
participant wants to present themselves in a
positive light so is not truthful) are presented.
- Participant bias presented from factors such
as time, age, gender.
- Response bias presented e.g. acquiescence
bias whereby participants simply agree with all
the questions, instead of putting effort into
considering an answer for each question.
Construction of questionnaires:
There are various factors that need to be thought about when designing questionnaires :-
● Clarity - the questions should be phrased in such a way that it is clear for the respondent
on what answer is needed from them.
● Avoid overuse of jargon, emotive language, double-barrelled questions, double
negatives and leading questions. All these can cause biases which affects the validity of
the results.
● Sequencing questions - easy ones can be first then followed by the harder ones. This
allows a build up of confidence in each participant.
● Filler questions - these are questions which have nothing to do with the aims of the
investigation and are put in to distract the participant from guessing the real aim of the
study. Therefore these eliminate demand characteristics.
● Pilot study - can be carried out to ensure that the questionnaire is suitable and if not
amendments and improvements can be made.
Interviews
There are two main types of interviews:
Structured - Involves a set of - Standardisation is - Interviewer bias which can be

predetermined questions possible. presented through aspects such
being asked during the - Easily replicable. as body language, listening skills,
interview. The interviewer - Can make comparisons when to ask a question and
asks the questions and for between participants interpretative bias (how answers
each waits for a suitable easily which is a strong are recorded).
response. benefit for job interviews. - Social desirability bias.
- Not being able to elaborate can
be frustrating for participants.
Unstructured - There are no - Lots of data is collected - Skilled interviewers needed.

predetermined questions , with more depth and - Interviewer bias also presented.
instead questions develop detail. - Social desirability bias.
as the interview goes on. - As can be tailored to - Difficult to make comparisons
This allows for questions be individuals they can between participants.
tailored to individuals and is provide more insight. - The analysis of data is difficult
more free flowing. as may have to sift through a lot
of irrelevant data.
There can also be semi-structured interviews whereby most of the questions are already set up but
the interviewer is free to ask any follow up questions on certain answers.
Construction and design of interviews:
● Recording information - this can be done in various ways e.g. writing down answers,
using a video recorder, using an audio recorder.
● Ethical issues - Informed consent is needed from the participant for the researcher to
obtain and keep the data. The participant should be reminded that their answers will be
kept confidential.
● Location - A quiet room away from other people is the most appropriate as this location is
likely to get the participant to feel comfortable and open up.
● Neutral questions - These are usually started with to make the participant feel relaxed and
help establish a rapport.
Reliability Across all Methods of Investigation
Reliability is a measure of how consistent the findings from an investigation are. Why is it
important?
● To ensure the DV is being measured accurately.
● To ensure that over periods of time, the outcome is still the same.
● To ensure that all the conclusions made are accurate and valid as if not can have
implications for theory development.
There are various types and ways of assessing each:
Type and description Ways of assessing
Internal reliability- describes how consistent - Split half method - Randomly select half of the
something is within itself. questions and put them in one form then do the
same for others. Theses two forms of the same
test are then done separately and should yield the
same score,have a correlation coefficient of ≥ 0.80
External reliability- this is when consistent -Test-retest method - the researcher administers
results are produced regardless of when the the same test on the same person on different
investigation is used or who administers it. occasions. The results should yield a correlation
coefficient of ≥ 0.80. Sufficient time should be left
between the test and retest so participants cannot
recall their answers, and this time in between
should not be too long as the person’s attitudes
may change.
- Inter-observer reliability - refers to the extent to

which there is agreement between two or more
observers involved in observing behaviour. This
eliminates subjectivity bias and may either be
carried out in a pilot study or reported at the end of
the study.
It is calculated by the formula:
Total number of agreements ≥ 0.80 = High

Total number of observations inter observer
reliability
Ways of improving reliability in:
Questionnaires
A questionnaire that produces low test-retest reliability may require some of the items to be
‘deselected’ or rewritten. One solution might be to replace some of the open questions where
there may be more room for misinterpretation, with closed, fixed choice alternatives which may be
less ambiguous.
Interviews
For interviews, probably the best way of ensuring reliability is to use the same interviewer each
time. If this is not practical or possible, all interviewers must be properly trained, so for example,
the interviewer should not ask leading or ambiguous questions.
They should all be able to structure their interviews in a ‘certain manner’ which can be followed by
all to ensure that everything is similar. This is more easily avoided in a structured interview where
the interviewer’s behaviour is more controlled by the fixed questions. Interviews that are
unstructured and more ‘free-flowing’ are less likely to be reliable.
Experiments
Lab experiments are often described as being ‘reliable’ because the researcher can exert strict
control over many aspects of the procedure, such as the instructions that participants
receive and the conditions within which they are tested. This control allows for the experiment
to be designed to be replicable, so that if someone else were to repeat it, they would get similar
findings (ideally).
Such control is often more achievable in a lab than in the field, since in the field variables are
much more difficult to control and the researcher doesn’t have access to it throughout the
procedure, so they just have to ‘make-do’ with the situation in hand.
This is more about precise replication of a particular method rather than demonstrating the
reliability of a finding. One thing that might affect the reliability of a finding is, if participants were
tested under slightly different conditions each time they were tested.
Observations
The reliability of observations can be improved by making sure that behavioural categories have
been properly operationalised, and that they are measurable and self-evident.
For instance, the category ‘pushing’ is much less open to interpretation than ‘aggression’.
Categories should not overlap for example, ‘hugging’ and ‘cuddling’ and all possible behaviours
should be covered on the checklist.
If the categories are not operationalised well, are overlapping, or absent different observers have
to make their own judgements of what to record where, and may well end up with differing
inconsistent records.
Validity
Validity refers to the extent to which results of a research study are legitimate. There are various
types of validity and ways of assessing them:
Types and description Ways of assessing
Internal validity - this is whether the outcomes - Face Validity - this is when a measure is
observed in an experiment are due to the scrutinised to determine whether it
manipulation of the IV and not any other factor. appears to measure what it is supposed
It is influenced by :- to. This can be done either through simply
● Confounding and extraneous variables. looking at it or passing it to an expert to
● Participant variables and demand check.
characteristics.
● Investigator bias. - Concurrent validity- this refers to the
extent to which a psychological measure
External validity - This relates to factors outside compares to a similar existing measure.
the investigation - is it generalisable to other The results obtained should either match or
settings, populations & eras. There are different be closely similar to the results of the well
forms of external validity: established and recognised test.
● Ecological validity- This is the extent to - Predictive validity- this refers to how well
which findings can be generalised to other a test can predict future events or
situations and settings. behaviours
● Temporal validity - Generalisability to other E.g. how childhood attachment measured
historical times and eras using the strange situation are able to predict
● Population validity - Generalisability to how the child will grow up to behave in
different populations of various ages, adulthood (from Attachment topic).
genders and cultures
Ways of improving validity in:
Experimental research
Using a control group in experimental research means that the researcher is better able to
assess whether changes in the dependent variable were due to the effect of the independent
variable.
For example, in a study looking at the effectiveness of a therapy, a control group who did
not receive therapy means that the researcher can have greater confidence that
improvements were due to the effects of the therapy rather than, say, the passage of time.
- Experimenters may also standardise procedures to minimise the impact of participant

reactivity and investigator effects on the validity of the outcome
- The use of single-blind and double-blind procedures are designed to achieve the same
aim
- In a single-blind procedure participants are not made aware of the aims of the study until
they have taken part (to reduce the effect of demand characteristics on their behaviour)
- In a double-blind study, a third party conducts the investigation without knowing its main
purpose either (which reduces both demand characteristics and investigator effects and
thus improves validity)
Questionnaires
Many questionnaires and psychological tests incorporate a lie scale within the questions in order to
assess the consistency of a respondent’s response and to control for the effects of social
desirability bias.
Validity may be further enhanced by assuring respondents, that all data submitted will remain
anonymous.
Observations
Observational research may produce findings that have high ecological validity as there may be
minimal intervention by the researcher. This is especially the case if the observer remains
undetected, as in covert observations, meaning that the behaviour of those observed is likely to
be natural and authentic.
In addition, behavioural categories that are too broad, overlapping or ambiguous may have a
negative impact on the validity of the data collected.
Qualitative methods
Qualitative methods of research are usually thought of as having higher ecological validity than
more quantitative, less interpretative methods of research. This is because the depth and detail
associated with case studies and interviews, for instance, is better able to reflect the participant’s
reality.
However, the researcher may still have to demonstrate the interpretative validity of their
conclusions – this is the extent to which the researcher’s interpretation of events matches
those of their participants.
This can be demonstrated through such things as the coherence of the researcher’s reporting and
the inclusion of direct quotes from participants within the report.
Validity is further enhanced through triangulation – the use of a number of different sources as
evidence for example, data compiled through interviews with friends and family, personal diaries,
observations etc.

Detailed Notes

Uploaded by

Copyright:

Available Formats

Detailed Notes

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Detailed Notes

Uploaded by

Copyright:

Available Formats

AQA Psychology A-level

Topic 7: Research Methods

This work by PMT Education is licensed under https://bit.ly/pmt-cc

Independent and dependent variables

Demand characteristics and Investigator effects

Randomisation and Standardisation

Experimental Method: Types of Experiment

Design Description Strengths Limitations

Laboratory An experiment that High degree of control- Experimenter’s bias- this

Field An experiment Naturalistic - so more Ethical considerations-

Quasi An experiment Controlled conditions- Cannot randomly allocate

Natural An experiment in Provides opportunities- Natural occurring events-

Sampling Method Explanation Strengths Limitations

Opportunity Participants happen to Easy method of Not representative of

Random sampling This is when all No researcher bias- Time consuming-

Systematic A predetermined Avoids researcher Not truly unbiased

Design Description Strengths Limitations Solution

Independent The participants -There are no - No control over Random

Repeated The same - Eliminates - Order effects Counterbalancing

Matched Pairs of - No order - Time

Single-Blind and Double-Blind Procedures

Type of observation and Strengths Limitations

Naturalistic- watching and - High ecological validity - Low ecological validity if

Controlled- Watching and - Researcher is able to focus - More likely to be observing

Overt- participants are - Ethically acceptable as - More likely to be recording

Non-participant- The - Researcher can be more - Open to observer bias for

Total number of agreements / total number of observations x 100 .

Design and description Strengths Limitations

Unstructured- consists of - More richness and depth of - Produces qualitative data

Structured- Here the - Easier as is more - Not much depth of detail.

Method and description Strengths Limitations

● The variables are simply measured, not manipulated like in experiments.

Data Analysis: Types of Data

Type and description Strengths Limitations

Meta-analysis - this is when a - More generalisability is - Publication bias such as the

Data Analysis: Descriptive Statistics

Measures of central tendency

Measure Calculation method Strengths Limitations

Mean- arithmetic Total of all values in a - Makes use of all - It is influenced by

Median Arrange data from - Not affected by - Not as sensitive as

Measure Calculation method Strengths Limitations

Range Minus the lowest - Easy to calculate. - Affected by extreme

Presentation and Display of Quantitative Data

Summarising data in a table

Main purposes of peer review:

● To know which research is worthwhile hence funding can be allocated to it.

● In peer review, any research that opposes mainstream theories tends to be

● Fraudulent research can be long-lasting which is a large problem. An example of this

Topic Links to specific areas Economy

Psychopathology Treatments - Cognitive - Workers able to return to

Attachment Role of the father - Tiffany - Mothers can return to work.

Social Influence Social influence leading to - Health campaigns.

Memory Eyewitness testimony - How - Led to police using the

Examples from psychology topics we have learnt :-

- Detailed so able to gain in depth insight. - Not generalisable to wider populations as

How to conduct a content analysis :

● Identify hypothesis that you will investigate.

Appropriate measures for each level of data:

Level of data Measures of central Measures of dispersion

Nominal Mode n/a