IAS Psychology

Research Methods Workbook

Exam Specification 2

Research Methods: Introduction 6

Independent and Dependent Variables 6

Aims and Hypotheses 8

Extraneous Variables and Control 9

Experimental Design 12

Evaluating Experimental Designs 13

Types of Data 15

Reliability and Validity 16

Populations and Sampling 18

Sampling Techniques 19

Ethics 20

Types of Experiment 22

Evaluating Different Types of Experiment 23

Questionnaires 24

Interviews 25

Evaluating Self Reports 26

Observations 27

Case Studies 31

Correlations 32

Glossary 35

1
 Exam Specification
Subject content What students need to learn:

Self-reporting data
1.2.1 Designing and conducting questionnaires and interviews,
considering researcher effects.
1.2.2 Primary and secondary data.
1.2.3 Unstructured, semi-structured and structured interviews, open, closed
(including ranked scale) questions.
1.2.4 Alternative hypotheses.

Sample selection and sampling techniques

1.2.5 Random, stratified, volunteer, and opportunity sampling
techniques.
Topic A
Social Psychology Quantitative data
1.2 - Methods 1.2.6 (List A) Analysis of quantitative data: calculating measures
of central tendency (mean, median, mode), data tables (frequency tables
and summary tables), graphical presentation (bar chart, histogram),
measures of dispersion (range and standard deviation), percentages, ratios,
fractions.
1.2.7 Normal and skewed distribution.
1.2.8 Analysis of qualitative data using thematic analysis.

Ethical guidelines
1.2.9 British Psychological Society (BPS) code of ethics and conduct (2009),
including risk management when carrying out research in psychology.

1.4.1 One practical research exercise to gather data relevant to

topics covered in social psychology. This practical research exercise must
adhere to ethical principles in both content and intention.

In conducting the practical research exercise, students must:

Topic A ● design and conduct a questionnaire to gather both qualitative

and quantitative data to look for a difference in the data
Social Psychology
● consider questionnaire construction, sampling decisions and
1.2 - Practical
ethical issues
Investigation
● collect and present an analysis of quantitative data using measures of
central tendency (mean, median, mode as appropriate, measures of
dispersion, (including range and standard deviation as appropriate), table
and graphical representations (summary table, frequency table, bar graph,
histogram, as appropriate)
● collect and present an analysis of qualitative data using thematic analysis
● consider strengths and weaknesses of the questionnaire
● write up the procedure, results and discussion section of a report.

2
Subject content What students need to learn:

Experiments and experimental design

2.2.1 Designing and conducting experiments, including field and lab

2.2.2 Independent and dependent variables.
2.2.3 Experimental and null hypotheses.
2.2.4 Directional (one-tailed) and non-directional (two-tailed) tests and
hypotheses.
2.2.5 Experimental and research designs: repeated measures,
independent groups, matched pairs, their issues and controls.
2.2.6 Operationalisation of variables, extraneous / confounding variables
2.2.7 Control groups, counterbalancing, randomisation, order effects.
2.2.8 Situational and participant variables.
Topic B 2.2.9 Objectivity, reliability, validity (internal, predictive, ecological)
Cognitive 2.2.10 Experimenter effects, demand characteristics, control issues.
Psychology 2.2
- Methods
Decision making and interpretation of inferential statistics

2.2.12: (List B)
• Levels of measurement.
• Wilcoxon signed ranks test of difference (also covering Spearman's rank
correlation coefficient (formula) and Spearman's rank (critical values table)
and Chi-squared distribution once Unit 2 covered).
• Probability and levels of significance (p≤.10 p≤.05 p≤.01).
• Observed and critical values, and sense checking of data.
• One- or two-tailed regarding inferential testing.
• Type I and type II errors.

2.4.1 One practical research exercise to gather data relevant to topics covered
in cognitive psychology. This practical research exercise must adhere to ethical
principles in both content and intention.

In conducting the practical research exercise, students must:

● design and conduct a laboratory experiment using a repeated

Topic B
measures design to gather quantitative data and include descriptive statistics
Cognitive as analysis and Wilcoxon non- parametric test of difference
Psychology 2.4 ● make design decisions when planning and conducting experiment, using
- Practical repeated measures design, sampling decisions, operationalisation, control,
Investigation ethical considerations, hypothesis construction, experimenter effects and
demand characteristics and order effects
● collect, present and comment on data gathered, including using
measures of central tendency (as appropriate); measures of dispersion (as
appropriate); bar graph, histogram, frequency graph as relevant; normal
distribution if appropriate and draw conclusions
● use a Wilcoxon non-parametric test of difference to test significance,
including level of significance and critical/observed values
● consider strengths and weaknesses, and possible improvements
● write up the procedure, results and discussion section of a report.

3
Subject content What students need to learn:

Correlational research

3.2.1 The use of the correlational research method in psychology, including

co-variables.
3.2.2 Types of correlation: positive, negative and including the use of scatter
diagrams.
3.2.3 Issues surrounding the use of correlations in psychology; issues with
cause and effect, other variables.
Topic C
Biological Scanning techniques
Psychology
3.2 - Methods 3.2.5 Brain-scanning techniques (CAT, PET, and fMRI). The use of
brain-scanning techniques to investigate human behaviour, including
aggression.

Twin studies

3.2.6 The use of twin studies to investigate genetic relatedness and aggression.

3.4.1 One practical research exercise to gather data relevant to topics covered
in biological psychology. This practical research exercise must adhere to ethical
principles in both content and intention.

In conducting the practical research exercise, students must:

● design and conduct a correlational study to gather quantitative data and

include descriptive statistics as analysis and a non- parametric test of
relationship into aggression or body rhythms
● make design decisions when planning and gathering data for a correlational
Topic C
analysis, including co-variables, operationalisation, ethical considerations,
Biological hypothesis construction and controls
Psychology 3.4 ● collect, present and comment on data gathered, including descriptive
- Practical statistics and graphical representation (scatter graph)
Investigation ● use inferential statistical testing (Spearman’s rank test) and explain the
significance of the result and the use of levels of significance. Students must
also be able to use a correlation co-efficient (strength/direction) to explain the
relationship
● consider strengths and weaknesses of the correlation and possible design
improvements
● write up the procedure, results and discussion section of a report.

4
Subject content What students need to learn:

Observations

4.2.1 The use of the observational research method in psychology, including

the gathering of both qualitative and quantitative data (including tallying, event
and time sampling).
4.2.2 Types of observation: participant, non-participant,
structured, naturalistic, overt and covert.

Content analysis

4.2.3 Use of content analysis as a research method.

Case studies
Topic D
4.2.4 Freud’s use of the case study as a research method.
Learning
Theories & Quantitative data and qualitative data
Development
4.2 - Methods 4.2.5 List A from Topic A.
4.2.6 Analysis of qualitative data using thematic analysis.
4.2.7 List B from Topic B, focusing on the chi-squared test.
4.2.8 Animal research and ethics
• The use of animals in laboratory experiments where results can be related to
humans.
• Ethical issues regarding the use of animals in laboratory experiments,
including the Scientific Procedures Act (1986) and Home Office Regulations.

4.4.1 Two observations (one observation can be carried out if both qualitative
and quantitative data are gathered in the
same observation).

In conducting the practical research exercise, students must:

Topic D
Learning ● design and conduct two observations (or one that gathers both qualitative
and quantitative data) and include descriptive statistics as analysis, qualitative
Theories &
analysis using themes and a chi-square statistical test on an aspect of learned
Development
behaviour
4.4 - Practical ● make design decisions when planning and conducting the observation(s),
Investigation including the use of behavioural categories, coding sheets and tallying, control,
hypothesis construction, ethics and observer bias/effects
• collect and present and comment on data gathered, including descriptive
statistics; percentages, frequencies (as appropriate), graphical representation
(bar chart, histogram as appropriate)
• use a chi-square statistical test and explain significance of result
• conduct a thematic analysis on the qualitative data gathered and describe the
findings
• consider the strengths and weaknesses of the observation(s) and possible
design improvements
• write up the procedure, results and discussion section of a report.

5
 Research Methods: Introduction
Knowing how to conduct good research in Psychology is important as it allows us
to draw valid conclusions about human behaviour. There are many ways of
conducting research, for example using questionnaires, interviews, observations,
correlations, or in-depth case studies on individual people. You will learn about
all of these methods during the A Level Psychology course.
However, the method we will consider first is the use of experiments. In an experiment, the
researcher aims to manipulate a variable and measure its effect, whilst controlling all other
variables.
Task: To find out more about the research process, watch the following video:

Crash Course – Research and Experimentation

Independent and Dependent Variables

In Psychology, a variable is simply anything that can change.
Changing variables and testing their effect using an experiment is how psychologists and other
scientists work out the cause of something. Experiments use two key variables: the independent
variable (IV) which is the manipulated variable and the dependent variable (DV) which is the
measured variable.

6
Researchers must ensure that they operationalise the variables for their research and this means
putting them into a measurable form. Here is an example for an experiment that investigates the
effect of alcohol on reaction time:

IV = Alcohol (Condition 1: 4 units of wine, Condition 2: 1 unit of wine, Control: no alcohol)

DV = Reaction times (as measured by how fast a person hits the screen when they perceive a
hazard on a driving simulator)

Some things are easy to operationalise (e.g. height would be measured as the distance from the
bottom to the top in cm); whereas some are more difficult (e.g. emotions such as love and anger).

Why is operationalising necessary?

Operationalising variables allows other researchers to replicate (copy) experiments in order to
confirm or challenge the results.

Task: Identify the IV and the DV in the following studies, and explain how they could be
operationalised:
● A researcher investigated the effect of an audience on time taken to build a tower
Answer:
IV - whether there is an audience present or not (15 people / no-one)
DV - time taken to build tower (minutes/secs)

● A group of psychologists wanted to find out whether memory was affected by gender
Answer:

● A study was carried out to look at whether aggression levels differed depending on family
size
Answer:

● An experiment was run to look into whether pictures are easier to remember than words.

Answer:

7
 Aims and Hypotheses
The aim is an important starting point for psychological research because it clearly signposts the
topic being investigated and directs the overall ambition of the study. Once the focus is clear,
psychologists then make a prediction about what is likely to happen in the investigation based on
previous knowledge. These predictions are known as hypotheses.
In experiments the psychologist first puts forward an experimental hypothesis; this is the
statement they wish to test and this can be written in different ways:

• Directional (one-tailed) hypothesis: this is a

statement predicting the direction of a relationship
between variables.

• Non-directional (two-tailed) hypothesis: this is a

statement predicting that one variable will be
related to one another but it won’t state in which
direction.

• Null hypothesis: this is the default hypothesis if

there is a greater likelihood of the results occurring
by chance. It simply states that the results
obtained are due to chance and not the effect of
changes in the independent variable.

Example:
How you would
write hypotheses
for an
investigation into
the effect of
teacher dress
style on student
behaviour.

8
Task: Write an appropriate experimental and null hypothesis for the experiment described
below (you must decide whether it should be directional or non-directional)

Answer:

Extraneous Variables and Control

In order to find a valid relationship between the independent and independent variables, it is
important to control or eliminate other factors that may have an unwanted effect on the
dependent variable. These are called extraneous variables.

Sometimes an extraneous variable can influence the dependent variable, and make it look as
though the effect is from the independent variable; this is called a confounding variable.
Extraneous and confounding variables can be divided into three types - situational variables (an
extraneous variable found in the experimental setting), participant variables (different
characteristics of the participants themselves) and experimenter effects (the way an experimenter
may influence the outcome of the experiment by their actions or mere presence). Experimenter
effects can cause demand characteristics which is when the participants alter their behaviour to
meet the expectations of the experimenter.
They can either be controlled by ensuring they are held constant for all participants, or eliminated
by removing the possibility of them occurring in the first place.

9
Example:
The table below shows some of the extraneous variables that may affect an investigation into the
effect of teacher dress style on student behaviour.
Task: Can you identify which of the example extraneous variables are confounding?

Task: List some extraneous variables that could affect the following studies and then consider
how you would control or eliminate them:
- An investigation into sporting experience and the ability to shoot a basketball hoop from
the halfway line.

Answer:

- An investigation to see if more cars stopped at a pedestrian crossing on a busy street for
men or women.

Answer:

10
 - An investigation into essay writing skills of history and maths undergraduate students.

Answer:

- An investigation into alcohol and driving performance.

Answer:

Standardisation: A common way to control extraneous variables is to keep the experimental

experience the same for all participants: this is known as standardisation.
Task: Read the following scenario and annotate it, picking out the elements of the investigation
that you could standardise
A study was conducted at a high school to investigate the effect of noise levels on students’
performance on a task.

40 pupils from Year 12 were recruited to take part, and were split into two groups of 20 – group A and
group B. Group A were sent to Room 1 to take part in the ‘noisy’ condition and group B went to Room 2
to take part in the ‘no noise’ condition.

Teacher A created a list of anagrams for group A, whilst Teacher B created a similar list for group B.
Both groups had to complete the list of anagrams as quickly as possible and were timed with a
stopwatch.

The only difference was that Mrs Scales banged a drum in EC1 for the entire duration whilst Mr Nutt
sat quietly in BS1 and let the students get on.

The results showed that Group B on average completed the anagrams significantly faster than Group A.

It was therefore concluded that lots of loud noise reduces performance on tasks.

Single and Double Blind Procedures: To control demand characteristics, a

single blind procedure can be used, where participants are unaware of the
study aim so that it does not affect their performance. To eliminate
experimenter effects a double blind procedure can be used when neither the
experimenter nor the participant know the true aim of the study.

11
 Experimental Design
The design of an experiment just means how you allocate your participants to the different
conditions of the IV.
For example, say you are conducting a study to see the effect of music on memory. Music is the IV
and memory is the DV, so the conditions of the IV will be a) Music playing and b) No music playing.
Let’s say you have 20 participants. You have two main choices:
1) Repeated measures design. This is where the same participants take part in each condition.
Get all 20 to do the memory test without music. Then, have them all do the memory test
again but this time with music. You then would compare the results from the two
conditions.
OR
2) Independent groups design. This is where participants are different in each condition. Split
the group in half. 10 will do the memory test with music playing, and the other 10 will do
the memory test without music playing. You then would compare the results of the two
groups to see if music affects memory.
Note: You could also use a matched pairs design, which is the same as independent groups, but
when participants in each group are paired up according to key variables, such as age, gender or
IQ.
Task - Answer the questions below

12
 Evaluating Experimental Designs
Repeated Strengths No participant variables: because the groups are made up of the same
measures people, differences between individuals won’t affect the results.

Number of participants (PPs): fewer PPs are required to get the same
amount of data, therefore it is cheaper and less time consuming than the
other designs

Weaknesses Order effects can affect the validity of results as they can create
differences in the results between the two conditions of the IV which are
not the result of changes in the IV:

- PP’s can become practised at the task they are being asked to do,
so could do better in the second condition.
- On the other hand, PPs can also become bored with the task,
which could lead them to do worse in the second condition.

To overcome order effects, researchers use a technique known as

‘counterbalancing’, where you mix up the order in which p’s complete the
tasks, e.g. some will do condition 1 then condition 2; and some will do
condition 2 first then condition 1. This helps to control order effects, as
the effect would be equal across conditions.

Independent Strengths No order effects: as PPs are only tested in one condition of the IV, they
groups are only tested once, so won’t experience practice or boredom effects.

Demand characteristics less likely (than with repeated measures): as PPs

are only tested in one condition of the IV, they are less likely to work out
what is being investigated, and therefore less likely to change their
behaviour as a result.
Weaknesses Participant variables: differences between the people in each group
might affect the results (e.g. participants in group 2 may have a higher IQ
and therefore may perform better than participants in group 1)

Number of participants: need twice as many participants to get the same

amount of data, compared to using the repeated measures design – this
is more time consuming, and can be more expensive.

Matched - Strengths Same as for Independent Groups, but with an additional strength:
pairs design
Less participant variables: important differences are minimised through
matching

Weaknesses Number of participants: need twice as many participants to get the same
amount of data, compared to using the repeated measures design – this
is more time consuming, and can be more expensive

Practical difficulties: If pre-testing is required to match PPs (e.g. IQ tests),

this design can become even more time consuming, and even if PPs are
matched on important variables, there could still be other PP variations
that still affect the results.
13
Task: Read the following scenario and then answer the question.

Describe one disadvantage of the design used in this experiment (3 marks)

Answer: The design used is repeated measures and the disadvantage of this is that the
performance of the PPs may be affected by order effects, for example they may do better when
there is an audience because it is the second time they built the tower, so are better practiced.

Task: Do it yourself! Design an experiment to investigate stress in some way. You must have an
aim, experimental hypothesis, design and list at least 3 controls.

Aim:

Hypothesis:

Design:

Controls:

14
 Types of Data
Qualitative vs quantitative: There are two main types of data that can be collected from research.

Quantitative data: data involving numbers e.g. if the

psychologist takes a measurement/score it will be
quantitative data. It can come in the form of averages,
percentages, or raw figures, for example.
Qualitative data: descriptive data. This means it is normally
in the form of words, but may also be in other formats like
images. This type of data may be more in-depth.

Task: Identify whether each example below is quantitative or qualitative data:

Example of data Type of Data
Average score on a maths test quantitative

Participants’ ratings, from 1 to 10, of how close they are quantitative

to their parents
A summary of how a teacher treats boys and girls qualitative
differently in her lesson
How long (in milliseconds) participants take to respond to quantitative
an image on the screen
The case notes on a boy raised as a girl by his parents qualitative

✓ A strength that quantitative data has over qualitative data is that it is easy to identify patterns
and trends.
✓ A strength that qualitative data has over quantitative data is that it provides richer, more
in-depth data that may make results more valid.
Primary vs Secondary Data:
Data can also be defined as primary or secondary data.
Most data in psychology is primary as it has been collected
by the researchers themselves. A strength of this is that
the psychologist can be confident the data has not been
misinterpreted by someone else.
Secondary data is when a researcher uses data collected
by someone else, such as government statistics. The
strength of this is that it saves time and money, and the
psychologist may have access to large amounts of data
that they would not otherwise be able to collect .

15
 Reliability and Validity
Validity is whether the results are an accurate measure of the effect of the IV on the DV. If a set of
results are valid, it means a researcher has measured what they intended to measure.
Types of validity:
● Internal validity: the extent to which the outcome of the study is the direct result of the
manipulated IV.
● Predictive validity: the extent to which results from a study or test can predict future
behaviour. This can be used to assess internal validity.
● Ecological validity: the extent to which the findings of a study can explain the
investigated behaviour in different situations.

Threats to validity:
● Extraneous variables
● Lack of objectivity
● Lack of realism

Task: What would you do in each scenario to increase validity?

1 - You want to carry out an experiment to investigate how different people will help a stranger in
need. However, you are concerned about the effect of demand characteristics.
Answer:

2 - You want to carry out an observation of the way that teachers deal with disruptive behaviour in
the classroom. However, you are concerned about observer bias.

Answer:

3 - You want to carry out a lab experiment to test how teenagers react to frustrating situations.
However, you are concerned about ecological validity.

Answer:

16
Reliability is whether the results are consistent. If a study was carried out lots of times and each
time found different results, it would not be very reliable. This can sometimes happen due to
researchers’ different interpretations of what is being studied. A good way to deal with this is to
have more than one researcher involved in a study. If the researchers agree on what is found, then
the results have inter-rater reliability. This means that more than one person agrees that
something has happened, so the results are more likely to be reliable. Another way to test
reliability is to use the test-retest method, which involves repeating a measure used in an
investigation, and then checking that both sets of results are consistent.

Task: Read the following research study -

Milgram (1963) was interested in investigating whether ordinary people will obey a legitimate authority
figure even when required to injure an innocent person.

40 male participants took part in this study, and there was also a ‘confederate’.

Participants were allocated a role of a teacher or learner (which was fixed) and took part in a word pair
recall task. The confederate was always given the role of the ‘learner’ in each trial and always acted
exactly the same for each participant. The participant was always allocated the ‘teacher’ role. The
teacher was told to administer an electric shock to the learner every time he got a question wrong on
the tasks (the electric shock was fake but participants didn’t know this!). The learner mainly gave
wrong answers, he would always give 3 wrong answers and then 1 right answer to each participant and
then he received his fake shocks after a wrong answer. Participants were observed through a one-way
mirror by multiple observers and the sessions were also filmed.

In total 65% of participants continued to deliver a deadly 450 volt shock, all participants went to 300
volts. Milgram conducted his research in other countries and found a similar level of obedience in those
too, for example UK 58% and Australia 68% obedience.

Explain two ways in which the study was reliable:

Answer:

He used a standardised procedure - so procedure is consistent for all participants and when
the study is replicated, the results will be reliable (consistent)
The study was repeated in other countries and similar results were found, suggesting
reliability.

17
 Populations and Sampling
Whatever type of research you are conducting, whether it be an experiment, a questionnaire or an
observation, you need a sample.
The sample is our group of participants. A sample is a smaller group drawn from a target
population (the group of people we want to look at).
For example, if we wanted to look at Crimson students then all students from all courses would be
our target population. However, we cannot have all of them in the study! So we need a sample.

Our sample needs to be representative of the target population. This means that our sample will
reflect who is in the target population – so it will represent a range of characteristics e.g. age,
personality and intelligence.
For example, if our target population was Crimson students then we would not want to take our
sample just from the IAL Psychology course. This is because that sample would not represent all
the ages across the school. Similarly, we would not want to choose only students from one subject,
as this would not represent the range of skills and ability across the school. It would also need to
be a mix of both genders.
There are different ways of choosing a sample. When we do this, we want to try and make sure
that our sample is representative, so that we can generalise the findings to others.

18
 Sampling Techniques

Random Opportunity Volunteer Stratified

Sampling Sampling Sampling Sampling

HOW? HOW? HOW? HOW?

When every member When the researcher When people When subgroups
of the target group samples whoever is volunteer to be in a within the target
has an equal chance available and willing study by responding population are
of being selected for to be studied. This to an advert by the identified, and then a
the sample. This could could be done by researcher (in a random sample is
be done by placing all asking people who newspaper for taken from each
names in a hat and walk by you in the example). The sub-group to ensure
drawing out the street researcher may then they are
required number select only those who proportionately
are suitable for the represented.
study.

Everyone has an This is a quick Least effort More

equal chance of and practical required by representative
being selected, so the way of getting a researchers to gather than random sampling
sample is likely to be sample – used often their sample, and the as it ensures that small
representative by researchers in most ethical as people subgroups are not
universities choose to take part. missed out.

Does not Sample may be Sample likely to The most

guarantee a biased as not all be biased as time-consuming
representative members of a target volunteers have method, and still does
sample – small group may be around specific personality not guarantee a
subgroups may not be at the time of the traits such as being representative sample
selected and some study. This means highly motivated. This as some selected
selected participants that the sample is means that the sample participants may
may chose not to take unlikely to be is unlikely to be chose not to take part
part in the research. representative of the representative of the in the research.
target population. target population.

19
 Ethics
All research should meet the ethical standards set out in the British Psychological Society (BPS)
code of ethics and conduct (2009) in order to protect the welfare of participants in psychological
research. This code is based on four ethical principles, and the specific standards are grouped
under one of these four principles as follows:
Principle 1 - Respect
Psychologists should:
A. Respect individual, cultural and role differences and avoid unfair or prejudiced practices
B. Keep appropriate records and safeguard the confidentiality of information relating to
participants. Respect participants' right to privacy, so no audio, video or photographic
recording without explicit permission.
C. Obtain informed consent of all participants, ensuring that they are given ample opportunity
to understand the nature, purpose, and anticipated consequences of any research
participation. Withhold information from clients only in exceptional circumstances when
necessary to preserve the integrity of research. Avoid intentional deception of clients
unless it is necessary in exceptional circumstances to preserve the integrity of research and
the deception is disclosed to clients at the earliest possible opportunity.
D. Ensure from the first contact that participants are aware of their right to withdraw at any
time from research participation.
Principle 2 - Competence
A. Develop and maintain a comprehensive awareness of professional ethics, and accept
responsibility to attempt to resolve ethical dilemmas that will inevitably arise.
B. Engage in Continued Professional Development and practice within the boundaries of
competence, seeking consultation when needed.
C. Monitor personal and professional lifestyle in order to remain alert for signs of impairment
and refrain from practice if professional competence is seriously impaired.
Principle 3 - Responsibility
A. Avoid harming participants and avoid any personal and professional misconduct that might
bring the reputation of the profession into disrepute
B. Aim to eliminate any potential risks for participants to psychological well-being, physical
health, personal values, or dignity
C. Debrief research participants at the conclusion of their participation, in order to inform
them of the outcomes and nature of the research, to identify any unforeseen harm,
discomfort, or misconceptions, and in order to arrange for assistance as needed.
Principle 4 - Integrity
A. Be honest, accurate and fair in all interactions with all persons.
B. Seek to promote integrity in all facets of scientific and professional endeavours.

20
Task: Read the very unethical (and fictional!) study below and annotate it to identify all the ethical
standards that have been broken.

A Very Unethical Study

A psychologist carried out an experiment to investigate the effect of hunger on
memory. His hypothesis was that hunger will decrease memory. He collected 50
participants through an advert in the local newspaper. The advert is shown below:

Participants arrived at the lab at 10am on the first day and were given breakfast.
They then had to take part in the memory tests. Participants were then told they
would not be given lunch or dinner, and that they were not to eat anything until 10am
the next day. Any food they had brought was confiscated.
At 10am the next day participants completed further memory tasks, before being
given breakfast again. This process continued across the five days.
At one point, a young woman complained of feeling ill and shaky due to not having
eaten. She asked if she could finish the experiment early. Researchers told her that
it would ruin the whole experiment if she were to leave, and that she was now
committed to it.
Some of the results of the study are shown below:
Participant name Average memory Average memory
score immediately score when hungry
after eating
Kate Bennett 15/20 12/20
Jack Thompson 19/20 11/20

21
 Types of Experiment
There are two types of experiments that you need to know about:
● Lab experiments are carried out in a controlled environment. The researcher directly
manipulates the IV.
● Field experiments are carried out in a natural environment (e.g. a school, a shopping centre, a
psychiatric ward). The researcher directly manipulates the IV.

Task: which is best?

Imagine you are a Psychologist...
You want to investigate whether people are more likely to obey a person dressed as a policeman
than a person dressed in normal clothes when asked to pick up a piece of litter from the street. You
really want to find out people’s genuine reaction…why might a lab experiment not be
appropriate?
Answer:

You now want to investigate whether age has an effect on how quickly someone can complete a
puzzle. Why might it not be a good idea to carry this out as a field experiment in a Primary school
classroom?
Answer:

22
 Evaluating Different Types of Experiment
Lab + Well controlled, extraneous variables minimised. Good internal validity.
IV manipulated to + Easier to replicate, and to check for reliability
observe effect on DV,
under controlled
conditions - Artificial setting, so lacks ecological validity
- Investigator, participant effects and demand characteristics more likely
which affects internal validity
Field + Less artificial so ecological validity higher
IV manipulated to + Avoids some participant effects, as participants not aware of study, so
observe effect on DV,
in a real life setting higher internal validity

- Less control of extraneous variables, which could affect internal validity

- Participants may guess what they are being studied for, which could
cause demand characteristics

However…
Experiments are not the only way to collect data. Non-experimental methods are those which do
not manipulate an IV and do not attempt to control extraneous variables. The non-experimental
methods you will learn about are:
● Self-reports: Interviews & Questionnaires
● Observations
● Case studies
● Correlations

Task: Think - what do each of the research methods above involve?

23
 Questionnaires
Sometimes rather than conducting an experiment, researchers want participants to report on their
own feelings and thoughts. This is called a self-report method. A questionnaire uses this type of
method as participants have to answer pre-set questions based on their own experiences, opinions
or feelings. Self reports use either open or closed questions, and can produce either quantitative
(numerical) or qualitative (descriptive) data.
Open question: A question that produces qualitative data. The pp is encouraged to answer in full
sentences and can expand on their answer.
Closed question: A question that produces quantitative data. The pp can only select from a set
range of options, and there is no opportunity to expand on answers
Task: Read the following questionnaire and identify which questions are open or closed.

24
 Likert Scale: A Likert scale is a fixed choice rating scale used in psychological research to measure
the level of agreement/disagreement with statement on a continuum. Respondents are only given
a certain number of options to respond and cannot respond in any manner they choose. Likert
scales are typically 5 or 7 responses on a spectrum ranging from agreement with the statement to
disagreement.
Ranked Scale question: A rating scale question where items are ranked against other items in a list,
often in order of preference.

Interviews
Interviews are another self-report method. They are different from questionnaires in the sense
that they are always carried out face to face. The psychologist asks questions which are generally
open rather than closed so that more in-depth information can be given. Whereas questionnaires
are always structured, interviews can be structured, unstructured or semi-structured:

• Structured: these have pre-determined questions (much like a questionnaire) and can be
delivered face-to-face, over the phone or via video call and there is no deviation from the
questions.
• Unstructured: a conversation with a researcher about a specific topic. Questions can be added
in real time during the interview to explore particular points in more depth.
• Semi-structured: these have questions in a fixed order, but the researcher is also allowed to add
additional questions throughout the interview to allow the pp to expand on areas of interest

Task: Read the following two examples of interviews, and answer the question below.

Interview 1 Interview 2
Researcher: What is your job? Researcher: What is your job?
Participant: I’m a store manager Participant: I’m a store manager
Researcher: What is the most stressful thing Researcher: That must be pretty busy. What
about your job? kind of shop is it?
Participant: Probably dealing with Participant: Yes, it can be. It is an electrical
customers, there’s certain things they can shop
be really demanding about. Researcher: Does it get particularly stressful
Researcher: Do you have trouble sleeping around Christmas then?
after a busy day? Participant: Yes absolutely, that is definitely
Participant: No not generally, I tend to the most stressful time for us. The rest of
forget about work as soon as I’m home the time it is relatively easy going.

Name the two types of interviews being used above, and explain how they are different from
each other.
Answer:

25
26
 Evaluating Self Reports
Task: For the following statements, use two different colours to identify the strengths and
weaknesses of questionnaires and interviews. Check that you understand all the key terms in
bold text, and check definitions in the glossary.

Questionnaire Interview

If questions are not constructed carefully, In all interviews, researcher characteristics

responses can lack validity. For example, (e.g. sex, age, personality) may influence the
confusing questions can produce incorrect answers that the respondent gives, so the
answers, likert scales with “neutral” options validity of the data could be affected by
can lead to a lack of opinion being expressed investigator effects.
and likert scales that all follow the same
response pattern can lead to response bias if Structured interviews are standardised, so can
respondents fall into a pattern of always be easily repeated (like questionnaires) and
“agreeing” or “disagreeing”. are therefore easier to administer and test for
reliability than other forms of interview.
People may lie on questionnaires or try to give However data may be superficial and lack
a socially desirable answer. This affects the depth as the respondent cannot express
validity of the research. To help avoid this, a themselves fully.
researcher may use lie detector questions, and
if too many of these are answered, the Unstructured interviews produce more
respondent is excluded from further analysis. qualitative data that can provide more insights
However, questionnaires can be done into the way that someone is thinking and
anonymously, reducing the chance of social feeling. However, the interviewer needs to be
desirability bias. highly skilled to use non-judgmental language
otherwise their questions can become
Questionnaires can be tested for reliability and subjective and be able to analyse answers on
validity. The test-retest method can be used to the spot. They must also be sensitive when
test for external reliability and the split half asking for personal information so that they do
method for internal reliability. Face validity not breach the respondent’s right to privacy.
can be checked by reading each question and
deciding whether it makes sense in terms of Semi-structured interviews use more open
the construct being tested, and predictive questions than structured interviews, so can
validity can be tested by checking whether the gather more qualitative data allowing for
construct being measured can predict the deeper analysis, however they do not have a
same construct in the future. standardised format to follow, so are harder to
replicate than structured interview.

27
 Observations
We have considered some of the problems of self-report, such as
people not telling the truth or not being able to explain themselves.
To overcome this problem, researchers can observe people instead.
Although observations can take place as part of an experiment,
many observations take place in people’s natural environment
without the manipulation of an IV. This means that psychologists
can watch their participants going about their normal behaviour.
There are many different types of observation that may be used in research:

Type of Observation Definition

Structured The observer records behaviour within a controlled
environment, such as a lab.

Naturalistic The observer records behaviour within the participants’ own

environment, so the situation has not been created by the
researcher.

Overt Participants are aware that they are being observed.

Covert Participants are not aware that they are being observed.

Participant The researcher takes an active role in the situation being

observed.

Non-participant The researcher does not become involved in the situation

being observed, e.g. by watching through a one-way mirror
or using a remote video recording device.

Sometimes an environment can be too busy to try and record everything that is happening. So we
decide how we are going to ‘sample’ the behaviour:

Time Sampling: behaviour, as specified on a Event Sampling: every occurrence of

predetermined checklist, is observed and behaviour, as specified on a predetermined
recorded at specific time intervals (e.g. every checklist, is observed and recorded within a
10 minutes for a period of 15 seconds) specified period of time

+ More representative over time + Less likely to miss predefined events

+ Easier, more reliable observations + Increases validity and reliability
- Can miss important events - Can be hard if lots of behaviour occurs
at once
- Can miss events not coded for

28
Task: State whether the following observations are naturalistic or structured; covert or overt;
participant or non-participant.
Example: A researcher secretly joins a religious cult to see if people are being brainwashed. He
records his observations in a daily diary. Covert participant naturalistic observation
1. A researcher watches infants and their carers through a one-way mirror in a playroom in
her lab to investigate infant co-operation.
Answer:

2. A researcher secretly observes student behaviour by enrolling on an A Level course and

pretending to be a mature student.

Answer:

3. A Principal observes a geography lesson. He sits with the students and takes part in the
activities that they do, whilst writing down notes on his clipboard.

Answer:

4. A researcher wants to observe how many people come to a full and complete stop at stop
signs. The researcher uses video footage from CCTV cameras mounted on the stop sign.

Answer:

29
Task: What are the strengths of the different observations? To help you, read through each
weakness, and see how you could turn it into a strength for a different observational technique.
Type of observation Strength Weakness

Overt Participants can behave unnaturally

because they know they are being
watched, which limits the validity of
the data collected.

Covert Participants do not know they are

being observed, so cannot give
consent, creating an ethical issue.

Participant When an observer is involved in the

group they may influence the
participants’ behaviour and become
too involved so they stop being
objective.

Non-participant The researcher may miss details of

an activity or lack insight as they are
not involved.

Naturalistic There is a lack of control over

extraneous variables, reducing
internal validity and as they are not
standardised they are harder to
replicate reliably.

Controlled Situations can be artificial, so people

do not behave naturally, meaning
results lack ecological validity.

A general weakness of all types of observations is that they can be prone to observer bias.
There is always a risk that the researcher will only see what they want to see, particularly as many
behaviours are open to subjective interpretation. To control for this, it is important to use well
defined behaviour categories on the behavioural checklist and for thorough researcher training.

30
Task - Exam practice:

Evaluate the use of the observational method in this study (5 marks)

Answer:

31
 Case Studies
Task: read the following case study and highlight/annotate parts that show the key features,
strengths and weaknesses of case studies:

32
 Correlations
Correlations measure two co-variables to see if there is a relationship between them. Like case
studies, correlations use other methods to collect data – mainly questionnaires and observations.
However, unlike case studies, they use quantitative data. For example:

● Is there a correlation between the number of crimes a father and son commit?
● Is there a correlation between stress and illness?
● Is there a correlation between days absent from school and exam performance?

The researcher will look to see if there is a positive, negative or no correlation.

● A positive correlation means that when one variable increases, so does the other.
● A negative correlation means that when one variable increases, the other decreases.
● A zero correlation means that there is no relationship between the two variables.
Task: What type of correlation do you think the following co-variables would show?
1. Temperature and amount of work completed in lesson
Answer:

2. Temperature and ice cream sales

Answer:

3. Amount of revision done and exam scores

Answer:

4. No. of pets and reading ability

Answer:

5. Amount of exercise and hours slept

Answer:

Correlations are represented on a

scatter diagram. Scatter diagrams
allow researchers to see the
strength and direction of the
correlation – is it positive or
negative? Is it strong or weak?

33
The strength and direction of a correlation is shown by calculating a correlation coefficient (more
on this when we do inferential stats!). A correlation coefficient of +1 represents a perfect positive
correlation, -1 would show a perfect negative correlation, and 0 would indicate no correlation at
all. Sometimes correlation coefficients will be somewhere between, for example +0.85 shows a
strong positive correlation, and -0.35 shows a weak negative correlation.
Task: A psychologist investigated how many daily hassles participants reported in a four week
period, and correlated this with a stress score. The higher the stress score, the more daily hassles a
participant was experiencing.
Plot the data (and label) from the study using a scatter diagram and decide what type of
correlation is shown:
Participant 1 2 3 4 5 6 7 8 9 10
No. of daily 15 12 8 4 23 15 9 10 18 7
hassles

Stress score 12 10 5 7 19 14 3 8 20 5

*Plot one variable on the x axis and the other on the y axis – it doesn’t matter which way round. Each cross
you plot will represent one participant.

34
Evaluating Correlations
Task: Read the following evaluation points of correlations. Label them as strengths or weaknesses.

Correlations allow psychologists to carry out investigation on things that

cannot be experimented on. For example, it would be unethical to
manipulate people’s hormone levels in an experiment to see if it gives them
depression. However, it is possible to simply measure people’s depression
levels and then correlate these with their existing hormone levels.

Although correlations can inform us about the relationship between

variables, it does not tell us which variable causes the other. Only
experiments can do this reliably because they are able to control all other
variables. For example, if a psychologist finds a positive correlation between
depression levels and hormone levels, they cannot be sure whether
depression affects hormone levels or whether hormone production affects
depression.

Correlations need to use quantitative data, which can lead to a lack of

construct validity. For example, although it might be reliable to measure
hormone levels, some psychologists would argue that it is not valid to
measure something complex like depression in terms of numbers.

Correlations have high levels of ecological validity compared with

experiments as nothing is set up or manipulated.

Task: A psychologist investigated how many daily hassles participants reported in a four week
period, and correlated this with a stress score. The higher the stress score, the more daily hassles a
participant was experiencing.
Explain one strength and one weakness of using a correlation to investigate the relationship
between daily hassles and stress.

Answer:

35
 Glossary
1. Aim: A description of what the researcher is trying to find out from conducting an investigation

2. Bar chart: A graph used for discrete categories, so there are gaps between the bars.

3. Case Study: A research method that involves the study of one person or event that is usually unusual in
some way. It involves the collection of quantitative and qualitative data using a range of methods.

4. Closed question: A question that produces quantitative data. The pp can only select from a set range of
options, and there is not opportunity to expand on answers

5. Confounding variable: extraneous variable that interferes with the action of the independent variable on
the dependent variable

6. Correlation: A research method that involves plotting two variables on a scatter graph to see if they relate
to each other.

7. Counterbalancing: A technique used to overcome order effects in a repeated measures design when half
the pps do condition 1 of the IV first, and the other half do condition 2 of the IV first.

8. Covert Observation: When participants do not know that a researcher is in the group observing them.

9. Demand characteristics: Cues about the aim of the experiment that participants pick up from the
experimental situation, which can affect behaviour.

10. Dependent variable: The variable that you measure in the experiment that is expected to be affected by
changes in the independent variable.

11. Directional (one-tailed) hypothesis: A statement predicting the direction of a relationship between
variables.

12. Ecological validity: This is about whether a research study has a realistic environment and task so that
the behaviour produced is similar to real life

13. Ethical Guidelines: Guidelines for how to treat participants in the right way in research

14. Experimental Design: The way in which participants are grouped and tested between the conditions of
the IV

15. Experimenter bias: When an experimenter knows their hypothesis and their measurements may
conform to what they expect to see.

16. Extraneous variable: A variable that could affect the DV

17. Field experiment: A research method where an IV is manipulated in a natural environment

18. Generalisability: This is about whether a study has a varied and large enough sample to be
representative of a larger population

19. Histogram: A graph illustrating frequency to show the distribution of continuous data

20. Hypothesis: A predictive statement about what effect you expect the IV to have on the DV

21. Independent Groups: A design where different people are used in each condition of the IV, without
being matched

36
22. Independent variable: The variable that you manipulate to create two or more conditions that is
expected to have an effect on the dependent variable.

23. Inter-rater/observer reliability: The extent to which 2 researchers interpreting qualitative responses in a
questionnaire (or interview) will produce the same records.

24. Laboratory experiment: A research method where the experimenter manipulates an independent
variable (IV) and measures a dependent variable (DV) in a controlled environment.

25. Matched Pairs: A design where different, but similar, people are used in each condition of the IV

26. Mean: The average of a set of numbers

27. Median: The middle score in a set of numbers

28. Mode: The most common score in a set of numbers

29. Naturalistic observation: An observation that involves observing pps in their normal environment

30. Non-directional (two-tailed) hypothesis: A statement predicting only that there will be a difference in
the measures of the DV for the two conditions of the IV

31. Non-participant observation: An observation where the researcher is not part of the social setting being
observed. E.g. they observe behind a mirror or from a distance.

32. Normal distribution: When mean, median and mode are the same

33. Null hypothesis: A testable statement saying that any difference or correlation is due to chance

34. Objectivity: An unbiased external viewpoint that is not affected by an individual's feelings or
experiences, so should be consistent between researchers.

35. Observation: A research method that involves watching people or animals and recording their
behaviours.

36. Open question: A question that produces qualitative data. The pp is encouraged to answer in full
sentences and can expand on their answer.

37. Operationalisation: When you clearly define variables so that they can be accurately measured or
manipulated.

38. Opportunity sample: When a sample is gathered by asking anyone immediately available to take part in
the research study

39. Order effects: Effects caused by using a repeated measures designs because participants repeat a test
more than once

40. Overt Observation: When participants know who the researcher is and that they're being observed.

41. Participant observation: An observation where the researcher is part of the social setting being
observed.

42. Participant variable: Individual differences between participants (e.g. age, intelligence, personality)

43. Qualitative data: Non-numerical data - usually in the form of descriptions

44. Quantitative data: Numerical data

45. Questionnaire: A research method that uses written questions

37
46. Random allocation: A technique used to control participant variables between groups when the pps
have an equal chance of being allocated to either condition of the IV.

47. Random sample: When a sample is gathered by using a system to select participants from a target
sample list, where each person on the list has an equal chance of being selected.

48. Range: A measure of dispersion where the difference between the highest and lowest score is calculated

49. Reliability: This is about whether data is consistent.

50. Repeated Measures: A design where the same people are used in each condition of the IV

51. Representative: This is about whether a sample has similar kinds of people to the target population.

52. Research Method: The main way of collecting data in a research study

53. Sampling Method: The way in which an experimenter gathers the sample for the research study

54. Scattergraph: A way to display data from a correlation to show the relationship between two variables

55. Self report: A research method where a participant reports on their own thoughts and behaviours by
answering a series of written (questionnaire) or verbal (interview) questions in a set order

56. Semi-structured interview: An interview with questions in a fixed order, but the researcher is also
allowed to add additional questions throughout the interview to allow the pp to expand on areas of interest

57. Situational variable: An extraneous variable caused by an aspect of the environment (e.g. noise,
temperature, light levels)

58. Skewed distribution: When median and/or mode differ from the mean

59. Standard deviation: A measure of dispersion that calculates the average difference between each score
and the mean. The bigger the SD, the greater the variation in scores

60. Standardisation: When the procedure for each participant is kept exactly the same to ensure that any
difference in pps are due to the changes in the IV only

61. Stratified sample: a sample drawn in such a way that known subgroups within a population are
represented in proportion to their numbers in the target population

62. Structured Interview: An interview with questions in a fixed order, which may be scripted.

63. Structured observation: The observer records behaviour within a controlled environment, such as a lab
using a specific set of pre-defined behaviours.

64. Subjectivity: A personal viewpoint, which may be biased by one's feelings or experiences, so may differ
between researchers.

65. Target population: The group of people that we want to generalise the results of a study to.

66. Test-retest reliability: A way to measure the consistency of a test by repeating it and checking whether
the two sets of data correlate well

67. Unstructured Interview: A research method where the participant discusses something with a
psychologist

68. Validity: This is about whether the experimenter has an accurate measurement of what they think that
they are measuring

69. Volunteer (self-selecting) sampling: When a sample is gathered by placing an advert and getting
participants to volunteer themselves

38