PAPER 2

RESEARCH METHODS
Psychology is a science
Peoples intuitions about human behaviour also known as folk psychology after
turn out to be wrong. This is one primary reason that psychology relies on science
rather than common sense.

Researchers search for evidence and consider alternatives before accepting a

claim about human behaviour as true.

Psychology is a science because it takes the scienti c approach to understanding

one aspect of the natural world: human behaviour

The general scienti c approach has some fundamental features:

1. scientists learn about the natural world systematically, by carefully planning,

making, recording and analysing observations of it.

2. The second feature is that it is concerned with empirical questions - these are
questions about the way the world actually is

3. The third feature of science is that it creates public knowledge, that is scientists
publish their work to be reviewed by their peers.
fi
fi
UNIT 1: RESEARCH METHODS
1.1 Experiments

1.2 Self-reports

1.3 Case studies

1.4 Observations

1.5 Correlations

1.6 Longitudinal studies

1.7 The de nition, manipulation, measurement and control of variables

1.8 Sampling of participants

1.9 Data and data analysis

1.10 Ethical considerations

1.11 Evaluating research: methodological issues

fi
 Research questions:
There are three main kinds of research in psychology:

descriptive - observes and describes phenomena

(without manipulation)

correlational - examines relationships between

variables focusing on whether and how strongly
variables are related to each other

causal questions - investigates cause-and-e ect

relationships by determining how an independent
variable (cause) e ects a dependent variable
(e ect).
ff
ff
ff
 Research Research Research Research Research methods
questions study paradigm Designs

Descriptive What Observational Quantitative Observations 1.Observations

questions questions studies and • Structured (QT)
Qualitative Interviews • Unstructured (QL)
Self report
studies 2.Interviews
• Unstructured (QL)
Association How Correlational Quantitative Self reports 1.Questionnaires
and questions studies • Close ended (QT)
prediction • Open ended (QL)

2.Interviews
• Structured

Causational Why Experimental Quantitative Experiments 1.Lab

questions questions studies and 2.Field
Qualitative 3.Quasi
4.RCT
Variables
Research questions in psychology are about variables.

A variable is a quantity or a quality (category) that varies across

people or situations.

A quantitative variable is a quantity such as height, that is

typically measured by assigning a number to each individual.

Examples: con dence levels, how depressed someone is, how

many siblings they have and so on

A categorical variable is a quality such as sex that is typically

measured by assigning a category label to each individual.

Examples: gender, nationality, whether someone is married or

not, is receiving psychotherapy or not and so on.
fi
 Kinds of Data

Quantitative data (QT) Qualitative data (QL)

Objective Subjective
factual/not impacted by interpretation description/impacted by interpretation

Statistical Analysis Thematic analysis

Idiographic
Nomothetic

Provide explanations
Provide evidence

• Self reports: • Self reports:

Questionnaires: close-ended Questionnaires: open-ended
Interviews: structured, close-ended Interviews: Unstructured, open-ended
questions questions

Observations: Observations:
Structured + semi structured Unstructured
Example of data
1. Physiological measures

Pulse rate - a heightened pulse can indicate fear or anxiety

MRI scans - changes in di erent regions of the brain in terms of volume and activity

2. Psychometric tests

IQ test

BDI (Becks depression inventory)

GAD-7

3. Self reports- questionnaires and interviews

Close ended questions

FFMQ: ’I perceive my feelings and emotions without having to react to them’ (rating scale:
never or very rarely true, rarely true, sometimes true, often true, very often or always true)

Open ended questions

4. Observations

Structured observations
ff
1.1 EXPERIMENTS
PAPER 2 RESEARCH METHODS
 EXPERIMENTs

IV DV

Independent Dependent
variable variable
CAUSE EFFECT
Manipulated Measured
Changed Result

Sleep Concentration
 1.1 Experiments
Experiment:

An experimental research is used to identity causal links between variables. That is,
they test whether one variable causes a change in another variable.

Researchers manipulate one variable (IV) and observe/measure the e ect it has on
the other variable (DV) .

Independent variable (IV):

the factor under investigation in an experiment which is manipulated to create two

or more conditions (levels) and is expected to be responsible for changes in the
dependent variable.

Dependent variable (DV):

the factor in an experiment which is measured and is expected to change under the
in uence of the independent variable.

All other variables are kept constant.

fl
ff
Practise
Identify the IV and DV:
Do people who wear blue shirts (cause) get more
job interviews (e ect)?

Does ca eine e ect memory?

Scores on tests improve with time (e ect) spent

studying (cause).

Does aggression increase with the consumption of

alcohol?
ff
ff
ff
ff
 Operationalising variables:
DEFINING in a way that you can MEASURE that
variable.

Why is it important to operationalise variables?

Operationalising variables is important in psychology and research in general because it involves
de ning variables in a clear, measurable, and speci c manner, which enhances the reliability,
replicability, and validity of the research.

For example: consider the abstract concept of "aggression" in psychological research. By

operationalising this variable, researchers might de ne it as the number of aggressive behaviours
(e.g., verbal insults, physical actions) displayed by participants within a speci ed time during an
experiment.

This concrete de nition allows researchers to measure and compare aggression consistently across
di erent studies, improving the scienti c rigour of psychological research.
ff
fi
fi
fi
fi
fi
fi
Practise questions:
A hypothesis in a study states ‘Recall will be better after a short delay than after a long
delay’.

Suggest one way to operationalise ‘recall’ in this study.

..................................................................................................................................................

..............................................................................................................................................[1]
Suggest one way to operationalise ‘short delay and long delay’ in this study.
..................................................................................................................................................

.............................................................................................................................................[1]
Elea is conducting an experiment to test whether people who sleep for more than eight
hours per night are happier than people who sleep less than this.
(a) (i) Elea’s dependent variable is ‘happiness’. Suggest how this could be operationalised.
..................................................................................................................................................................
…………………………………………………………………………………………………………
[1]
(ii) Suggest one disadvantage of operationalising happiness in this way.
..................................................................................................................................................................
…………………………………………………………………………………………………………
..................................................................................................................................................................
…………………………………………………………………………………………………………
[2]
Describe why psychologists operationally define the independent variable and the dependent variable in
experiments, using any examples.
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
.................................................................................................................................................................[6]
Practise question:
Albert is investigating autism. He plans to obtain a sample of people with autism and a control group.
He will compare how well these participants judge the feelings of dolls which are either dancing,
crying, waving or smiling.
(a) Identify the independent variable in this experiment.
..................................................................................................................................................
..............................................................................................................................................[1]
(b) Suggest how the dependent variable in this experiment could be measured.
..................................................................................................................................................................
…………………………………………………………………………………………..………………
..................................................................................................................................................................
………………………………………………………………………………………………………[2]
Zho is investigating how quickly fish learn to respond to different coloured lights. She trains
fish in two tanks to swim to the end of the tank for food in response to a light. She uses a
red light with one tank and a green light with the other tank. She times how quickly the fish
swim to the end of the tank.
(a) Identify the dependent variable in this experiment.
..................................................................................................................................................
..............................................................................................................................................[1]
(b) Identify the independent variable in this experiment.
..................................................................................................................................................
..............................................................................................................................................[1]
Elea is conducting an experiment to test whether people who sleep for more than eight
hours per night are happier than people who sleep less than this.
(a) (i) Elea’s dependent variable is ‘happiness’. Suggest how this could be operationalised.
..................................................................................................................................................................
…………………………………………………………………………………………………………
[1]
(ii) Suggest one disadvantage of operationalising happiness in this way.
..................................................................................................................................................................
.................................................................................................................................................................
.................................................................................................................................................................
.................................................................................................................................................................
[2]
Olivia is conducting an experiment to investigate whether students concentrate better in
class before or after eating. Her experimental design is a repeated measures design.
(a) Identify the independent variable (IV) in this experiment.
..................................................................................................................................................

..............................................................................................................................................[1]
(b) Identify the dependent variable (DV) in this experiment.
..................................................................................................................................................

..............................................................................................................................................[1]
Establishing a causal
relationship: controlling variables
In order to establish a causal relationship the researcher must be able
to assert with some degree of con dence that it was indeed the IV that
caused the change in the DV. In order to do this the researcher must
control all the other variables.

This is a challenge because as human beings we are impacted by a

number of apparent and invisible variables.

Researchers typically try to overcome this challenge by:

1. Controlling extraneous variables i.e setting up a laboratory

experiment

2. Setting up a control group.

fi
 Kinds of variables:
Participant variables: Individual di erences between
participants (e.g. age, gender, personality, intelligence)
that could e ect the DV. These variables can hide or
exaggerate di erences between the levels of the IV i.e.
confound the results. In such a case they would be
confounding variables.

Situational variables: Situational variables refer to the

environmental or contextual factors that can in uence
participants' behaviour, emotions, and responses
during a study. These variables relate to the conditions
under which the study is conducted, including the
physical and social aspects of the environment.
ff
ff
ff
fl
Participant variables Situational variables

Gender Lighting conditions

Cognitive ability Noise

Past experiences Time of day

Instructions or instruments
Health conditions
used
 Kinds of variables:
controlled variables: A controlled variable in an experiment, is a factor or condition
that is intentionally kept consistent and unchanged throughout the experiment.

This is done to ensure that any observed e ects or di erences in the dependent
variable(s) are the result of the independent variable(s) being manipulated and
not due to variations in other factors.

To remove its e ects a control variable may be held at a constant level during
the study or managed by statistical means

uncontrolled variables: a variable which either acts

randomly (called extraneous variables), a ecting the DV in all levels of the IV or

systematically, i.e. on one level of the IV (called confounding variables) so can

obscure the e ect of the IV, making the results di cult to interpret.

Good control of extraneous variables raises validity.

The e ect of uncontrolled variables should be minimised e.g. by keeping all

factors except the IV the same (controlled) each condition (or level of the IV)
ff
ff
ff
ff
ff
ffi
ff
 Kinds of variables:
Extraneous Variables

• These are any variables other than the independent variable (IV) that might a ect the dependent
variable (DV). They can in uence the outcome of the experiment if not controlled.

• Extraneous variables can exist in any experiment but do not necessarily correlate with the IV.

• Example: In an experiment measuring the e ect of sleep on test performance, room

temperature or noise level could be extraneous variables if they in uence participants’ focus
but are unrelated to how much sleep they got.

Confounding Variables

• These are a subset of extraneous variables that vary systematically with the IV and therefore
provide an alternative explanation for the results.

• They make it unclear whether the IV or the confounding variable caused changes in the DV.

• Confounding variables threaten the internal validity of an experiment because they introduce a
hidden relationship.

• Example: In the same experiment on sleep and test performance, if participants who got less
sleep also had to wake up earlier (leading to increased fatigue), fatigue would be a confounding
variable. This is because it is tied to both the amount of sleep (IV) and test performance (DV).
fl
ff
fl
ff
 Extraneous variable:
room temperature

IV: DV:
Sleep Test performance

Confounding variable:
Fatigue

IV: DV:
Sleep Test performance
Practise question:
Explain why it is important to control extraneous variables in experimental studies.
..................................................................................................................................................................
..................................................................................................................................................................
..................................................................................................................................................................
................................................................................................................./...............................................
[2]
Why is it important to control extraneous variables?

Maintaining Internal Validity:

• One of the primary goals of experimental research is to establish a cause-

and-e ect relationship between the independent variable (the variable
being manipulated) and the dependent variable (the variable being
measured).

• Extraneous variables are any variables, aside from the independent

variable, that could potentially in uence the dependent variable.

• If these extraneous variables are not controlled, they can confound or

obscure the true relationship between the independent and dependent
variables.

• By controlling extraneous variables, researchers can enhance the internal

validity of their experiments, which means that the observed e ects are
more likely to be a result of the manipulated independent variable rather
than other factors.
ff
fl
ff
Mikko is conducting an experiment to find out whether people are more easily persuaded by advertisements
that are in colour or in black and white. He has put advertisements up in one town library that are all in colour
and the same advertisements in black and white in a library in another town.
(a) Explain the importance of one control in Mikko’s experiment.
. …………………………………………………………………………………………………………………………….
……………………………………………………………………………………………….……………………………
……………………………………………………………………………………………………………………….……
………………………………………………………………………………………..……………………………………
……………………………………………………………………………….……………………………………………
……….……………………………………………………………………………………………/…………………[3]

(b) Suggest how Mikko could operationalise the dependent variable in his experiment.
………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………......
………………………………………………………………………………………………………………………….
………………………………………………………………………………………………………………………….[2]
 Kinds of groups in an experiment (IV):
The experimental condition/ experimental group:

In the experimental condition/s the IV is present.

one or more of the situations in an experiment which represent

di erent levels of the IV and are compared (or compared to a
control condition).

The control group/control condition:

In the control condition the IV is absent

The group that is exposed to the same experimental

environment but is not exposed to the independent variable

It provides a ‘base line’ for comparison with the experimental

group.
ff
Practise question:
Describe what is meant by ‘experimental conditions’ and ‘control conditions’ in research,
using any examples.
.....................................................................................................................................................................
.....................................................................................................................................................................
.....................................................................................................................................................
.....................................................................................................................................................
.....................................................................................................................................................
.....................................................................................................................................................
.....................................................................................................................................................
.....................................................................................................................................................
.....................................................................................................................................................
…………………………………………………………………………………………………………[6]
 Practise:
What e ect does ca eine have on memory?

Experimental group: (IV)- drink co ee

Control group: (IV)- did not drink co ee

Alcohol (IV) and aggression (DV)

Experimental group: (IV)- participants will consume alcohol

Control group: (IV) - participants will not consume alcohol

Do people with pink shirts get more jobs interviews?

Violence on television and aggression in children

ff
ff
ff
ff
 Levels of the IV:
Levels of the independent variable (IV) represent di erent conditions or values set by the
researcher to examine its e ects on the dependent variable.

In general, the number of levels of an independent variable is the number of experimental

conditions.

IV: music (kinds o ) DV: change in pulse rate

Classical Experimental IV present +

IV (1) - music group manipulated
DV

Rock Experimental IV present +

IV (2) - music group manipulated
DV

No Control
IV (3) - music group
IV absent DV
ff
ff
ff
 Why do researchers
need a control group?
• To establish a baseline for comparison with the
experimental groups. In this way the control group allows
researchers to measure the e ects of the experimental
treatment by providing a reference point una ected by the
manipulated variable.

• The control group helps establish causal relationship

between the IV and DV by showing that in the absence of
the IV, there was no signi cant change in the DV, therefore
con rming that the changes seen in the DV in the
experimental group were due to the manipulation of the IV.
fi
fi
ff
ff
 LEVELS OF THE IV

Independent
GENDER
variable

Two MALE FEMALE

levels

Independent Treatment
variable status

Three Treatment Control Placebo

levels
 LEVELS OF THE IV

Independent
Sleep duration
variable

Manipulated into
0 hrs 4 hrs 7 hrs 9 hrs
four levels

recommend
no sleep short extended
ed

Dependent
Cognitive performance
variable
 LEVELS OF THE IV

Independent depression severity

variable (de ned by BDI scores)

Five
0-9 10-18 19-29 30-39 +40
levels

minimal mild moderate severe extreme

depression depression depression depression depression

reaction time
Dependent
(measured in ms using a computerised
variable task)
fi
Levels of the IV: Example1
Placebo vs. Medication

Experiment: Investigating the E ect of Medication vs. Placebo on Depression

Hypothesis: Patients who take the true medication pill will experience a greater reduction in depression
levels compared to those who take the placebo pill or those in the control condition (no pill).

Variables:

• Independent Variable (IV): Type of Medication

◦ Level 1: True medication pill

◦ Level 2: Placebo pill

◦ Level 3: Control condition (no pill)

• Dependent Variable (DV): Overall change in depression (measured by a depression scale such
as the BDI score)
ff
 Levels of the IV: Example 2
For example, suppose a teacher wants to know how three di erent studying techniques
a ect exam scores. She randomly assigns 30 students each to use one of the three
studying techniques for a week, then each student takes the exact same exam.

In this example, the independent variable is Studying Technique and it has three
levels:

• Technique 1
• Technique 2
• Technique 3
In other words, there are the three experimental conditions that the students can
potentially be exposed to.

The dependent variable in this example is Exam Score, which is “dependent” on the
studying technique used by the student.
ff
ff
BANDURA ET AL
THE BOBO DOLL
EXPERIMENT

CAN AGGRESSION BE
LEARNT?
 Bandura et al: the process

Room 1 Room 2 Room 3

Subjects exposed
to either an Children are
Aggression
aggressive or observed for
arousal
nonaggressive imitative behaviour
model
 Understanding the levels of the IV:
Bandura et al: Aggression

Independent variable: Dependent variable:

exposure to aggression imitative/learnt aggression

Experimental groups /conditions Response measures

Imitation of physical Imitative verbal
Aggressive Non- Aggressive
aggression aggression

Same sex: Same sex:

Imitative non-
aggression verbal Mallet aggression
F-subject with F-model F-subject with F-model
response
M-subject with M-model M-subject with M-model

Di erent sex: Di erent sex:

Sits on Bobo doll Punches bobo doll
F-subject with M-model F-subject with M-model
M-subject with F-model M-subject with F-model
ff
ff
 Practise question:
Bandura et al. (aggression) used a control group.
(a) Identify how the control group in the study was different from an experimental group.
.....................................................................................................................................................
................................................................................................................................................ [1]
(b) Explain why control groups are important in psychological research.
........................................................................................................................................................
........................................................................................................................................................
.......................................................................................................................................................
.....................................................................................................................................................
[2]
What we have done so far:
What is psychology?

Psychology is a science and involves research

What is a variable? (quantitate and qualitative variables)

Experiments:

What is a causal relationship?

Independent and dependent variables

What are extraneous and confounding variables

How do experimenters make sure extra variables do not impact the validity of their
research?

Two kinds of groups in an experiment: the experimental group and the control
group

How do experimenter set up di erent IV conditions

Example: Bandura et al.

ff
 Experimental design:
Experimental design: the way in which participants are allocated to levels of the IV. Three kinds of designs:
 Independent measures design

Each participant does

IV 1 OR IV 2 OR IV3
one level of the IV only

Repeated measures design

Each participant does

IV 1 AND IV 2 AND IV3
each level of the IV

Matched pairs design

Each participant does
IV 1 AND IV 2 AND IV3
each level of the IV
 Independent measures design

Each participant does

Co ee OR Tea OR Water
one level of the IV only

Repeated measures design

Each participant does

Co ee AND Tea AND Water
each level of the IV

Matched pairs design

Each participant does
Co ee AND Tea AND Water
each level of the IV
ff
ff
ff
 1. Independent measures design: an experimental design in which a
di erent group of participants is used for each level of the IV (condition).
It is an experimental design in which each participant performs in only
one level of the IV.

• Strength: This design is good because the participants only encounter the
experimental design setting once. They are therefore unlikely to respond
due to demand characteristics. Also there are no order e ects.

• Weakness: More participants are needed. One problem is that there might
be individual di erences between participants (participant variables) that
might in uence the ndings. This risk can be reduced by random
allocation of participants to di erent conditions.

• random allocation: a way to reduce the e ect of confounding

variables such as individual di erences. Participants are put in each
level of the IV such that each person has an equal chance of being
in any condition.
ff
fl
ff
fi
ff
ff
ff
ff
 2. Repeated measures design: the same group pf people participate at
every level of the IV. It is an experimental design in which each participant
performs in every level of the IV.

• Strength: The main advantage is that each person acts as their own
baseline. Any di erence between participants that could in uence their
performance and therefore the DV, will a ect both levels of the IV in the
same way. Individual di erences (or participant variables) are therefore
unlikely to bias the ndings

• Weakness: As the participants perform the same or similar task

repeatedly this can lead to the order e ect. Repeated performance can
cause the participants to improve at the task: a practise e ect.
Alternatively repetition might make performance worse if the participants
are tired or bored- the fatigue e ect.

Order e ects can be xed by either: randomisation or

counterbalancing.
ff
ff
fi
fi
ff
ff
ff
ff
ff
fl
2.b Randomisation
Imagine an experiment with two conditions: learning while
listening to music (M) and learning with no music (N).

In randomisation, participants are randomly allocated to

do either condition M followed by N, or vice versa.

As some will do each order, any advantage of doing one

of the conditions rst will probably be evened out in the
results.

To be more certain that possible e ects are evened out,

counterbalancing can be used.
fi
ff
 2.b. Counterbalancing:
Suppose we used a repeated measures design in Day 1 Day 2
which all of the participants rst learned words in
‘(IV1) loud noise' and then learned it in ‘(IV12) no
noise.' We would expect the participants to show
better learning in 'no noise' simply because of
order e ects, such as practice.
Group A IV 1 IV 2
However, a researcher can control for order e ects
using counterbalancing.

The sample would split into two groups

experimental (A) and control (B). For example,
group 1 does ‘A’ then ‘B,’ group 2 does ‘B’ then ‘A’
this is to eliminate order e ects.
Group B IV 2 IV1
Although order e ects occur for each participant,
because they occur equally in both groups, they
balance each other out in the results.
ff
ff
ff
fi
ff
 INDEPENDENT MEASURES DESIGN
Experiment: IV - noise (IV1: loud nouse, IV2: no noise) DV: Attention

MEAN SCORES
p1 p2 p3 p4
IV 1: Loud noise=
3+4/2

3.5
IV 1
Test score: Test score:
Loud noise 3/10 4/10 IV 2: no noise=
8+9/2

8.5

Participants
scored more
when their
attention was
IV 2 tested in the ‘no
No noise Test score: Test score: noise’ condition
as compared to
8/10 9/10
the loud noise
condition
 INDEPENDENT VARIABLE
DEPENDENT VARIABLE
IV 1: loud noise
EXTRANEOUS VARIABLE

(loud music played in a

test score out of 10
10/10 room) PARTICIPANT VARIABLES points on an attention
Individual di erences in test
participants in Group A

INDEPENDENT VARIABLE
DEPENDENT VARIABLE
IV 2: no noise (control)

(loud music played in a

EXTRANEOUS VARIABLE test score out of 10
10/10 room) points on an attention
PARTICIPANT VARIABLES
test
Individual di erences in
participants in Group A
ff
ff
 REPEATED MEASURES DESIGN
Experiment: IV - noise (IV1: loud nouse, IV2: no noise) DV: Attention

MEAN SCORES
p1 p2 p3 p4
IV 1: Loud noise=
3+4+2+4/4
IV1 IV2 IV1 IV2
3.25
(loud noise) (no noise) (loud noise) (no noise)
DAY 1
Test score: Test score: Test score: Test score: IV 2: no noise=
7+8+8+9/4
3/10 8/10 2/10 9/10
8.0

Participants
scored more
IV2 IV1 IV2 IV1 when their
(no noise) (loud noise) (no noise) (loud noise) attention was
DAY 2 tested in the ‘no
noise’ condition
Test score: Test score: Test score: Test score:
as compared to
7/10 4/10 8/10 4/10 the loud noise
condition
 INDEPENDENT VARIABLE

IV 1: loud noise DEPENDENT VARIABLE

EXTRANEOUS VARIABLE
(loud music played in a test score out of 10
10/10 room) points on a test of
ORDER EFFECTS
attention
Order (IV1 rst then IV2)

INDEPENDENT VARIABLE
DEPENDENT VARIABLE
IV 2: no noise (control)

(loud music played in a

EXTRANEOUS VARIABLE test score out of 10
10/10 room) points on a test of
ORDER EFFECTS
attention
Order (IV1 rst then IV2)
fi
fi
 3. Matched pairs design: an experimental design in
which participants are arranged into pairs. Each pair
is similar in ways that are important to the study and
one member of each pair performs in a di erent level
of the IV.

1. By using di erent participants in each group

order e ects are avoided and

2. the matching of participants minimises the

in uence of individual di erences.
fl
ff
ff
ff
ff
3. Matched pairs:

matched
Group A with IV1
group B

matched
Group B with IV 2
group A
Strengths and weaknesses of experimental design:
Practise questions:
Bella is conducting a laboratory experiment to find out about people’s fear of animals. She
asks each participant if they have any phobias. If the participant says yes they are asked to
leave. The remaining participants are shown photographs while their pulse rate is measured.
The independent variable (IV) is whether the photographs are of animals with eight legs or
animals with four legs.
The experiment used a repeated measures design.
Explain one advantage of this design in this study.
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
.................................................................................................................................................................[2]
Ed and Jim are planning a field experiment about learning. They want to know whether
young children learn to use mobile (cell) phones from older brothers or sisters. They each
offer their phone to a younger brother or sister and also to a cousin who has no older
brothers or sisters. They time how long it takes each child to find a game to play with on the
phone.
Explain the experimental design being used in this experiment.
..................................................................................................................................................................
..................................................................................................................................................................
..................................................................................................................................................................
.................................................................................................................................................................
[2]
Richard is conducting an experiment to investigate whether daydreams are affected by the type of
background noise. He has two groups who are both given a boring task in a room. For one group
there are children laughing outside the open window. For the other group there are adults talking
loudly in the corridor. Different participants are used in each group.
(a)
(i) Identify Richard’s experimental design. Include a reason for your answer.
......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
.................................................................................................................................................................[2]

(ii) Suggest one advantage of this experimental design in this study.

......................................................................................................................................................................
......................................................................................................................................................................
......................................................................................................................................................................
……………………………………………………………………………………………………………[2]
 Reliability - in experiments

measures
• tools like pulse meters, MRI scans are
objective and therefore reliable
- how objective were
the tools used to
• self reports: close ended questionnaire +
structured interviews + structured
measure the DV or IV
observations provide objective and
quantitive data and are therefore improve the
reliability of a study

Reliability
consistency

procedure
- how standardised was
the procedure?
• if within each level of the IV the procedure is
consistent the study will have reliability

lower in eld lower in the

experiments eld because
because it is harder to
standardisation control
is di cult extraneous
fi
ffi
fi
 Validity - in experiments

the extent to which the researcher can be sure that

internal validity the changes measured in the DV were caused by the
manipulation of the IV.
addresses whether causality was
established
a ected by
high in lab lower in eld • control of extraneous variables (participant +
experiments experiments situational),
• control groups,
• demand characteristics,
because of because of
• order e ects,
Validity high degree of low degree of
• experimenter bias,
controls controls
accuracy • social desirability

ecological validity
the extent to which the results of the study can be
addresses whether the results are applied to real world contexts
true/applicable to the real world

lower in lab higher in eld

experiments experiments
improves of the study has high mundane realism i.e.
is able to measure participants natural and normal
because of low because of responses
mundane high mundane
realism realism
ff
ff
fi
fi
 LABORATORY EXPERIMENTS

INTERNAL ECOLOGICAL
VALIDITY VALIDITY

FIELD EXPERIMENTS

INTERNAL ECOLOGICAL
VALIDITY VALIDITY
 Factors that impact Internal Validity:
Participant variables: (independent measures)

• individual di erences between participants (such as age, personality and intelligence) that could a ect
their behaviour in a study. They could hide or exaggerate di erences between levels of the IV.
Participant variables such as IQ can e ect studies on memory studies.

• Solution: random allocation: a way to reduce the e ect of confounding variables such as individual
di erences. Participants are put in each level of the IV such that each person has an equal chance of
being in any conditions

• It is di cult to control participant variables in the eld, in the laboratory the researcher can be more
selective about the participants and therefore set up controls.

Situational variables:

can be things like the time of day, room temperature or how much sleep a participant has had.

Solution: Standardisation: Situational variables are easier to control and standardise in the lab, but
harder to control in the eld.

Order e ects: (Repeated measures)

the e ect of boredom, fatigue or learning (the practise e ect) when participants go through di erent IV
conditions in an order. Can e ect validity in repeated measures design.

Solution: Counterbalancing: participants are divided into two groups, one group does the IV
conditions in one order (IV1 then IV2) and the other does the IV conditions in the reverse of that order
(IV2 then IV1)
ff
ff
ffi
ff
ff
fi
ff
ff
fi
ff
ff
ff
ff
ff
 Demand characteristics (higher in repeated measures):

Demand characteristics are features of the experimental situation which give away the aims.
They can cause participants to try to change their behaviour, which reduces the validity of the
study.

Solution: a blind study- participants do not know whether they are in the experimental group or
the control group

Demand characteristics are higher in lab experiments, lower in eld experiments

Experimenter bias:

When the researchers expectations in uence the experiments results. It might be subtle clues in
the way they phrase things, or body language in the experimenter that e ect the participants
behaviour. For example: An experimenter who wants to see a drug passed

Solution: a double-blind technique- the experimenter and the participant both don’t know how
was in the control group and who was in the experimental group.

Social desirability bias (higher in the lab):

when people behave in a certain way because they want to portray themselves in a better light.
Social desirability is likely to be higher in the lab as participants know they are being studied.

Solution:

covert observation: where participants are not aware of being observed or studied

ller questions: in questionnaires help reduce social desirability bias by masking the true
purpose of the survey and diverting attention from sensitive or leading questions
fi
fl
fi
ff
 Factors that impact Ecological validity:
• Control over variables: if the variables are controlled such as
in a lab experiment - the condition becomes so arti cial (which
improves internal validity) and unlike real life that it can loose its
applicability to real life (ecological validity)

• Ecological validity is the extent to which the results of a study

can be applied to the real world. Studies that have low
ecological validity cannot be generalised to real-life situations
and therefore have low external validity. Also, ecological validity
is higher when the research has mundane realism. Mundane
realism is harder to achieve in the lab and is not problematic in
the eld.

• Mundane realism references the extent to which the

experimental situation/setting is similar to situations
people are likely to encounter outside the laboratory.
fi
fi
 Can you apply ndings to real
How real is the setting?
world situations?

Bedroom 1:
Lab low mundane Ecological
steel bed, tube lights, realism validity is low
white paint, plain room

Bedroom 2:
mundane
Lab Ecological
realism
cozy bed, rug, lamp, side slightly higher validity is better
table

Bedroom 1:
mundane
Field Ecological
realism
participant sleeps in their high validity is high
own bed
fi
Kinds of experiments:
1. Laboratory experiment: an experiment carried out
in a controlled/arti cial environment

2. Field experiment: an experiment carried out in a

natural environment
fi
 Evaluating Lab Experiments:
Advantages Disadvantages
Experimental situation may not be applicable
or relate to the real world low mundane
realism- low ecological validity
Experimenter can control extraneous
variables- so that the change in the DV The sample may not be representative, and
can be attributed to the e ect of the IV- being the in the lab may impact participants
which increases internal validity behaviour because of the social desirability
e ect.
The procedure can be standardised
which raises reliability and allows for Participants may be responding to demand
replication. characteristics which would lower validity

The researcher can assess responses by Practical and ethical constraints- It may be
qualitative or quantitative means. unethical or impossible to assign people to
groups. For example: Unethical to deliberately
deprive children of sleep. Impossible or
expensive to send people on trips to study jet
lag
ff
ff
Describe the differences between a field experiment and a laboratory experiment, using any
examples.

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

...................................................................................................................................................

................................................................................................................................................[6]
Ed and Jim are planning a field experiment about learning. They want to know whether
young children learn to use mobile (cell) phones from older brothers or sisters. They
each offer their phone to a younger brother or sister and also to a cousin who has no
older brothers or sisters. They time how long it takes each child to find a game to play
with on the phone.
(a) Suggest why Ed and Jim chose a field experiment as their research method.
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
[2]
Gavin is testing the hypothesis that ‘People are more likely to drop rubbish on the ground when there is
already rubbish on the ground’. He is conducting a field experiment in a quiet street. His independent
variable is whether the street already has rubbish on the ground or not. Gavin’s dependent variable is
whether people drop a leaflet given to them by a stooge onto the ground or not.

(c) Explain one reason why Gavin chose to conduct a field experiment rather than a laboratory experiment.

......................................................................................................................................................................

.......................................................................................................................................................................

.......................................................................................................................................................................

......................................................................................................................................................................

…………………………………………………………………………………………………………………………[2]

(d) Explain one disadvantage of using a field experiment in Gavin’s study.

......................................................................................................................................................................

.......................................................................................................................................................................

.......................................................................................................................................................................

......................................................................................................................................................................

…………………………………………………………………………………………………………………………[2]
Hypotheses in
experimental research
a hypothesis is a testable statement based on the
aims of a research

Freuds - unconscious mind - not observable or

measurable

Bandura et al - level of aggression - observable and

measurable

a hypothesis is falsi able

hypothesis have operationalised variables

fi
 Different kinds of hypothesis
Alternative hypothesis: a testable statement predicting a di erence
between in the DV between levels of the independent variable in an
experiment

Non directional hypothesis: a statement predicting only that one

variable will be related to another, for example that there will be a
di erence in the DV between the levels of the IV in an experiment.

Directional hypothesis: a statement predicting the direction of a

relationship between variables, for example in an experiment whether
the levels of the IV will produce an increase or decrease in the DV.

Null hypothesis: an alternative hypothesis is the opposite of a null

hypothesis. A testable statement saying that any di erence in the
results is due to chance, that is, that no pattern in the results has arisen
because of the variables being studied.
ff
ff
ff
 Non-directional (two-tailed
hypothesis)
There is a di erence between the e ect of
variable X (IV1) and variable Y (IV2) on the DV.

there is a di erence in energy levels (DV) of

people who sleep 8 hours (IV1) and those who
sleep 4 hours (IV2).

There is a di erence between the

e ectiveness of mind maps (IV1) and revision
apps (IV2) in helping students to learn (DV).
ff
ff
ff
ff
ff
 Directional (one-tailed
hypothesis)

Participants exposed to (IV1) will show (increase/

decrease) in (DV) compared to participants exposed
to (IV2) who will show (increase/decrease) in (DV).

Participants who sleep 8 hours (IV1) will have higher

energy levels (DV) compared to those who sleep 4
hours (IV2).

Students who use mind maps (IV1) will learn more

e ectively (DV) compared to those who use revision
apps (IV2
ff
Null hypothesis
‘There will be no di erence in the DV between condition X (IV1)
and condition Y (IV2) ’ or

There will be no di erence between the e ectiveness of mind

maps and revision apps in helping students to learn

‘Any di erence in the DV between condition X (IV1) and condition

Y (IV2) is due to chance’.

Any di erence in e ectiveness of mind maps and revision apps in

helping students to learn is due to chance

* Make sure that you always state all/both of the levels of the IV (e.g. IV1 & IV2) and
the DV otherwise your null hypothesis will not make sense. For example, the null
hypothesis: ‘There is no di erence between mind maps and revision apps’ is meaningless=
0 marks - because the DV is not mentioned.
ff
ff
ff
ff
ff
ff
ff
 Type of
Definition Example
Hypothesis

Predicts that there will be

There will be a difference in memory recall
a difference between two
Non-Directional between participants who study in a quiet
variables but doesn't
Hypothesis room and those who study with background
specify the direction of
music.
the difference.

Predicts the direction of Participants who study in a quiet room will

Directional
the difference between have better memory recall than those who
Hypothesis
two variables. study with background music.

There will be no difference in memory recall

States that there is no
between participants who study in a quiet
Null Hypothesis relationship or difference
room and those who study with background
between variables.
music.
 Non-Directional
Example Directional Hypothesis Null Hypothesis
Hypothesis

1. Mindfulness and
Anxiety
There will be a difference in Students who practice There will be no difference in
anxiety levels (DV) between mindfulness (IV1) will have anxiety levels (DV) between
Students who practice students who practice lower anxiety levels (DV) than students who practice
mindfulness (IV1) and mindfulness (IV1) and those students who do not practice mindfulness (IV1) and those
students who do not who do not (IV2). mindfulness (IV2). who do not (IV2).
practice mindfulness (IV2)
with anxiety levels (DV).

2. Coffee and Reaction

Time
There will be a difference in People who drink coffee (IV1) There will be no difference in
reaction times (DV) between will have faster reaction times reaction times (DV) between
People who drink coffee people who drink coffee (IV1) (DV) than people who do not people who drink coffee (IV1)
(IV1) and people who do and people who do not (IV2). drink coffee (IV2). and those who do not (IV2).
not drink coffee (IV2) with
reaction time (DV).

3. Introverts/Extroverts
and Social Behavior There will be a difference in There will be no difference in
Extroverts will engage in more
social behaviour between social behaviour during group
social behavior during group
Introverts and extroverts introverts and extroverts tasks between introverts and
tasks than introverts.
with social behavior during during group tasks. extroverts.
group tasks (DV).
Penny is using cats and parrots in her experiment. Penny houses each animal alone
and only gives them their daily food every evening. Her independent variable is the
species. She thinks that parrots will share food because they are social animals
whereas cats live on their own. To test this, two animals of the same species are put
together with a small bowl of food every afternoon.
(a) Penny’s dependent variable is whether the animals share the food. Suggest how
Penny could operationalise this dependent variable.
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
………………………………………………………………………………………………………
[2]

(b) Write an operationalised non-directional (two-tailed) hypothesis for Penny’s

experiment.
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
………………………………………………………………………………………………………
[2]
(a) 1 mark for identifying way to operationalise. 1 mark for elaboration/justi cation e.g. how it would be quanti ed.

• food sharing could be operationalised as how much food each animal eats;

• e.g. by counting the number of pieces of food each animal takes;

• e.g. by weighing the animals

• see whether one animal waits for the other/see if they both get something to eat;

• e.g. by timing how long before each animal starts eating;

(b) 2 marks for a fully operationalised non-directional hypothesis, correctly using cats and parrots, and saying how food
sharing will be measured

• 1 mark for a non-directional hypothesis which is not operationalised

• Cats (IV1) and parrots (IV2) will differ in the time they spend sharing food (DV) (2)

• There will be a difference in the number of pieces of food (DV) shared between the cats (IV1) and parrots (IV2) (2)

• There will be a di erence in food sharing between the cats and parrots (1)

• There will be a di erence in helpfulness between the cats and parrots (1)

• There will be a di erence in food sharing between animal species (1)

ff
ff
ff
fi
fi
Otto’s aim is to investigate which age groups of internet users are most likely to play
internet-based games.
(a) Suggest an operationalised directional (one-tailed) hypothesis that Otto could test.
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
[2]
• 2 marks for a one-tailed hypothesis with both variables if correlational, 2 levels of IV and a DV if
experimental with at least one operationalised variable

• 1 mark for a one-tailed hypothesis with 2 levels of IV and a DV

• 0 marks for a non-directional (two-tailed) hypothesis

• 0 marks for a null hypothesis

Hypothesis may be experimental or correlational

• Adults are more likely to play internet-based games than younger people = 1 (experimental not
operationalised)
The older you are the more likely you are to play internet games = 1 mark (correlational not
operationalised)

• Older people/adults are more likely to play internet based games for more than 2 hours per day
than younger people / children. = 2 marks

• 12 year olds are more likely to play internet based games than people over 40 = 2 marks

• The older you are in years the more likely you are to play an internet game. = 2 marks

• The older you are the less likely you are to play an internet game for more than 8 hours a day. = 2
marks
An experiment is testing the aim that smiling a ects helpfulness. The procedure is to
investigate whether people leaving a shop will hold a door open for people who are
smiling.

(a) Suggest an operationalised directional (one-tailed) hypothesis that Otto could test.
……………………………………………………………………………………………………………
............................................................................................................................................................
............................................................................................................................................................
……………………………………………………………………………………………………………[2]
(b) Suggest an null hypothesis for this experiment.
………………………………………………………………………………………………………………
…………………………
……………………………………………………………………………………………………………[1]
ff
(a)

1 mark for a one-tailed hypothesis

2 marks for an operationalised hypothesis
Components: Smiling (IV1) not smiling (IV2) holding door/helping (DV) ‘more than’ (one tailed)

People help more when they are smiled at = 1 mark

People help more when they are smiled at but don’t when they are not = 2 marks.

This has smiling (IV1) and not smiling (IV2) there is the DV (help more) and there is 'helping more when smiled
at' so is one tailed.

People are more helpful so hold doors open more often when they are smiled at = 2 marks [more helpful/hold
doors (DV) when smiled at (IV1) but no IV2 so this can only be 1 mark]

People are more helpful than when they are not smiled at = 1 mark.

More helpful (in what way?) than when (with no detail) not smiled at?]

People are more helpful so hold doors open more often when they are smiled at than when they are not
smiled at = 2

[2 marks. Has IV1, IV2, DV and tail ( i.e. more)

People help when they are smiled at = 0 marks (too vague)

People leaving a shop will hold the door open for people who smile = 0 marks (too vagueWrite a null hypothesis
for this experiment. 1

(b) Has to have two levels of IV but DV does not have to be operationalised

Any di erence between how helpful people when they are smiled at or not is due to chance. = 1 mark

There is no di erence between helpful people when they are smiled at or not. = 1 mark

No marks for statements that are not hypotheses, or are experimental/ alternative hypotheses.
ff
ff
Carol is interested in theory of mind. She is comparing adults in ‘caring’ jobs such as
nursing with adults in ‘non-caring’ jobs that have less contact with the public. Carol
believes that adults in caring jobs will score higher on the Eyes Test than those in non-
caring jobs.
(a) Write a null hypothesis for Carol’s study.
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
…………………………………………………………………………………………………[1]
1 mark for a null hypothesis (whether operationalised or not)
0 marks for an experimental/ alternative hypothesis / correlational null.

e.g. There will be no difference between eyes test scores for caring job adults and non-
caring job adults = 1
e.g. Any difference between theory of mind in nurses and typists is due to chance = 1

e.g. There will be no difference between theory of mind and caring or non- caring jobs = 0
(nonsense)
e.g. Any difference between eyes test scores and adults job type is due to chance = 0
(nonsense)

e.g. There will be a difference between eyes test scores for caring job adults and non-caring job
adults = 0 (H1)
e.g. Theory of mind in nurses will be better than in typists = 0 (H1)
Gavin is testing the hypothesis that ‘People are more likely to drop rubbish on the
ground when there is already rubbish on the ground’. He is conducting a field
experiment in a quiet street. His independent variable is whether the street already has
rubbish on the ground or not. Gavin’s dependent variable is whether people drop a
leaflet given to them by a stooge onto the ground or not.
(a) State the type of hypothesis Gavin is testing. Include a reason for your answer.
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
[2]
1 mark for type of hypothesis.
1 mark for the reason (even if name is incorrect). This
may or may not be linked.

• directional hypothesis / one-tailed hypothesis = 1

mark

• because he is saying the way the IV will a ect the

DV (not just that there will be a di erence) = 1
mark

• as Gavin says rubbish will make people litter more,

not just that it will have an e ect on littering = 1 mark
ff
ff
ff
A hypothesis in a study about ways to reduce phobias states ‘Positive reinforcement will reduce
phobic reactions more than imagery exposure’.
Is this a directional (one-tailed) hypothesis or a non-directional (two-tailed) hypothesis? Include
a reason for your answer.
.......................................................................................................................................................
.....................................................................................................................................................
[1]
Write a null hypothesis for this study.
.......................................................................................................................................................
......................................................................................................................................................
[1]
 Sara wants to test whether people with a higher intelligence quotient (IQ) are
better at concentrating. She will conduct a laboratory experiment and
already has a test for measuring the IQ of the participants.

(a) Describe how Sara can conduct a laboratory experiment to test whether
people with a higher IQ are better at concentrating than people with a lower
IQ. [10 marks]

(b) Identify one practical weakness/limitation with the procedure you have
described in your answer to part (a) and suggest how your study might be
done di erently to overcome the problem.

Do not refer to ethics or sampling in your answer.

ff
1.2 SELF REPORTS
PAPER 2 RESEARCH METHODS
 QUESTIONNAIRES INTERVIEWS

Participants answer questions about Participants are asked questions by a

themselves in written form researcher face to face/in real time.

can be used to ask open or closed questions can be used to ask open or closed questions
but mostly used to ask closed questions but mostly used to ask open questions

mostly used to collect Quantitative data (QL) mostly used to collect Qualitative data (QL)

o er statistical comparisons/analysis o er in depth and detailed answers

• May su er from response bias and social • May su er from interviewer bias and
desirability bias. social desirability bias.

• Closed-ended questionnaires are nomothetic • Interviews may be idiographic because they

because they use standardised responses to allow for open-ended, personalised responses
identify general patterns across large groups, that explore an individual's unique
enabling broad comparisons. experiences.
ff
ff
ff
ff
 Evaluation of self report questionnaires:
One of the most common research methods is self reports in which people respond to questions about themselves regarding a
wide variety of issues such as personality traits, moods, thoughts, attitudes, preferences, and behaviours.

Data can be qualitative, but may also be quantitative depending on type of question

Strengths

Asking people directly means that participants are given the opportunity to express their feelings and explain their behaviour
rather than the researcher trying to work out reasons for their behaviour from other methods e.g observation

Questionnaires are easy to replicate.

Weaknesses:

Some participants may provide socially desirable responses/ not give truthful answers/respond to demand characteristics.

Researchers have to be careful about use of leading questions; it could a ect the validity of the data collected.

For closed ended questionnaire:

Quantitative data can be scored (objective and reliable) and compared to all other people (statistically analysed)
completing the questionnaire.

Relatively large numbers of participants can be done relatively easily.

Closed/ xed choice questions may force people into choosing answers that do not re ect their true opinion and
therefore may lower the validity.

For open ended questionnaire:

Quantitative data o ers depth and detail into an individuals perspective or experiences, the responses of the di erent
participants are di cult to compare

Data take time and e ort to analyse (thematic analysis) and therefore sample sizes are likely to be kept small

Participants get the chance to answer on their own terms.

fi
ffi
ff
ff
ff
fl
ff
 Characteristic Qualitative Data (QL) Quantitative Data (QT)

Nature of Data Descriptive, non-numerical Numerical, measurable

Data Collection Unstructured Interviews, open-ended tools (e.g. pulse meters, MRI), closed-ended
Methods questionnaires questionnaires, structured observations

Focus In-depth exploration of individual Identi cation of patterns and trends in groups
experiences

Objectivity An issue with qualitative data (QL) A strength of quantitative (QT) data gathered
gathered from open ended questions and from close ended questions (and structured
unstructured observations, that require observations) that are based on quanti cation.
interpretation.

Analysis Thematic analysis Statistical analysis (e.g., mean, correlation)

Goal Understanding underlying meanings and Testing hypotheses, measuring variables

motivations

Examples Interview on coping with stress Number of hours slept correlated with stress
fi
fi
 Questionnaires

open ended questions close ended questions

open questions ask for descriptions/detail closed questions ask for answers from
speci c options

Qualitative data (QL) Qualitative data (QT)

closed questions ask for answers from

open questions ask for descriptions/details
speci c options

• Ask questions that begin with why, what, • Di erent kind of questions:
how
• Dichotomous questions (YES/NO)
• e.g: How would you describe your • multiple choice questions (MCQs)
relationship with your mother?
• Rating scales
• e.g. Why are you a vegetarian?
• e.g. What motivates you to exercise? • Likert scales
ff
fi
fi
close-ended questions

dichotomous questions (YES/NO)

Rating scale

multiple choice questions (MCQs)

Likert scale (attitude measurement)

 Questionnaires

open ended questions close ended questions

Strengths and Weaknesses Strengths and Weaknesses

Qualitative data provides in-depth and detail Quantitative data allows for statistical
data analysis and comparisons

Responses can be a ected by

• Problem: social desirability • Solution: anonymity/con dentiality

• Problem: demand characteristics • Solution: ller questions

Harder to give to a larger sample and Can be given to a larger sample and therefore
therefore not as generalisable o er generalisable results

Require thematic analysis which is time Statistical analysis - is quicker and easier to
consuming give to a large sample

Are subjective because responses require Provide objective data because the answers
interpretation - reduce validity do not require interpretation - improve validity
ff
fi
ff
fi
 Questionnaires

open ended questions close ended questions

Dif culty in interpreting answers can lead to Reliable as the same questions are answered
a lack of reliability by each participants (standardised) and can
be compared

Solution: Solution:
inter-rater reliability/intra-rater reliability can be tested internal consistency (split half test) and inter-rater
to see if the results are reliable reliability

The content of the questions are important - if not stated properly can be misunderstood and
therefore the study might lack face validity.

the data does not require the researchers

depth and detail - improves validity but
interpretation and is therefore objective and
subjectivity within the responses and
more valid, however the lack of depth and detail
interpretation process reduces validity. They
especially the fact participants have to selected
give participants freedom to express
a response from xed options that may not
themselves - so are likely to be more valid than
represent their true opinion/feeling reduces
closed questions (in this respect)
validity
fi
fi
 SELF REPORT

open ended questions close ended questions

smaller samples + responses decreases larger samples + statistical increases

are unique GENERALISABILITY analysis GENERALISABILITY

depth and detail increases decreases

VALIDITY no depth and detail
VALIDITY

interpretation - subjectivity decreases increases

VALIDITY statistics - objective
VALIDITY

question: interpreted di erently decreases increases

by participants RELIABILITY
questions: standardised RELIABILITY
answers: are unique to each
participant and cannot be responses: xed hence
compared easily standardised
fi
ff
 Problems with questionnaires
Validity
• operationalisation
• dif culty with lying/social desirability
• dif culty with demand characteristics
• dif culty with response biases
• dif culty with face validity
• dif culty with construct validity
• dif culty with concurrent validity

Reliability
close ended questions
• internal consistency (across items- measured using the split-half test)
• test-retest reliability (across time- measured using test-retest)
open ended questions
• inter-rater consistency (agreement between different raters)
• intra-rater consistency (consistency within the same rater over time)
fi
fi
fi
fi
fi
fi
Reliability in self reports

close ended questions quantitative data

kinds of reliability testing reliability

e.g Bandura et al.
checked that the
split-half scoring of aggressive
internal consistency - splitting the test into ratings
- across items two halves,
was similar when
(questions) administering to the
same person- the two
done by di erent
Reliability researchers
consistency

test-retest
test-retest reliability e.g if Baron-Cohen et
- giving the same test
al. used the AQ on the
- over time twice, to see if the
scores correlate well same group twice,
they should get the
same AQ each time

How to assess reliability

ff
 Reliability in self reports

open ended questions

refers to the degree
of agreement or
qualitative data consistency between
di erent raters or
if two teachers are judges when
grading the same essay, assessing the same
inter rater reliability subject, task, or
high inter-rater
- across di erent consistency means they phenomenon.
researchers give similar scores

Reliability
consistency
refers to the
if one teacher grades consistency of a
intra-rater reliability the same essay twice single rater when
- same researcher at high intra-rater they assess the
di erent times consistency means they same subject, task,
give a similar score both or phenomenon at
di erent points in
time.

How to assess reliability

ff
ff
ff
ff
Validity - in self reports
Face validity is improved by ensuring questions are clear, directly
related to the topic, and easily understood. It worsens with vague or
face validity irrelevant phrasing

The extent to which a

The BDI has good face validity because its questions are
questionnaire appears to
straightforward and directly relate to symptoms of depression, such as
measure what it claims to
sadness, loss of interest, and changes in sleep. If respondents nd the
questions relevant and understandable, it enhances face validity.

construct validity
Construct validity is improved by thoroughly testing the questionnaire
The extent to which a against established theories and ensuring it accurately measures the
Validity questionnaire accurately intended construct. It worsens when questions fail to capture the full
measures the theoretical scope of the concept or measure unrelated factors.
accuracy concept or construct it is
intended to assess.
The BDI improves construct validity by being grounded in cognitive
theories of depression i.e Beck’s work on cognitive distortions and
negative bias.
concurrent validity

Concurrent validity refers Concurrent validity is improved by using well-established, reliable

to the extent to which a tests as benchmarks for comparison, ensuring consistent
test's results align with administration conditions, and increasing the sample size to enhance
those from an the accuracy of the comparison measures depression severity.
established measure of
the same construct,
assessed at the same The BDI improves concurrent validity by comparing its results with
time. established tools like the HDRS (Hamilton Depression Rating Scale).
Strong alignment con rms it accuracy
fi
fi
 SELF REPORT

QUESTIONNAIRES INTERVIEWS

Close ended Structured semi structured Unstructured

schedule + new
Open ended schedule no schedule
questions
likely consist of close close + open ended consist mostly of open
ended questions questions ended questions

Quantitative
Quantitative + Qualitative less Qualitative
Standardised free- owing
data data Standardised data

Subjective
Objective thematic Valid
statistical Reliable Objective less reliable analysis have depth
analysis requires and detail
interpretation

the interviewer has a

the interview format/ schedule of questions but no xed questions
order/ questions are xed can improvise and add the questions depend on
or predetermined more questions if the respondents answers
necessary
fi
fl
fi
 Structured Interviews

STRENGTHS

1.Standardised - the questions, their order, interviewers clothes etc are xed for
every participant hence the measure is reliable and the results of the
interview comparable between participants
2.The questions can be close ended or open ended, therefore both QT and QL
data can be collected.
3.The interviewer can clarify the meaning of questions where needed.

WEAKNESSES

1.Lack of exibility means there is little rapport

2.The presence of the researcher leads to the social desirability effect - where
the participants' behaviour is altered because they want to make a favourable
impression on the interviewer.
3.Can't guarantee honesty of participants.
fl
fi
 Semi-structured Interviews

STRENGTHS

1. Some questions are xed which makes sure that there is some similar information from
every participant. This allows for comparisons and averages to be calculated
2. It is also possible to ask questions speci c to the participant. This allows the researcher
to develop ideas and explore issues that are particular to that person.
3. They offer more detail and possibly more validity than structured interviews but are not
as in-depth as unstructured interviews and so the extent of further knowledge gained is
not as much as with unstructured interviews.
4. Encourages two-way communication. Provides an opportunity so that interviewers can
learn answers to questions and the reasons behind the answers.
5. Allows respondents time to open up about sensitive issues.
6. Provides qualitative data to compare to previous and future data

WEAKNESSES

1. Open-ended questions are dif cult to analyse.

2. Can't guarantee honesty of participants- social desirability
3. Dif cult to compare answers as each interview will be unique.
fi
fi
fi
fi
 Unstructured Interviews

STRENGTHS

1.Produce very valid results because the researcher gets time to build a rapport with the
participant and gain their trust. This can lead to more honest and natural responses
2.Topics that require a detailed, sensitive and in depth exploration of the subjective reality of
respondents and a description of their experiences are very well covered by unstructured
interviews.
3.As there are no restrictions on the interviewer frequently the study yields knowledge that is
novel and insightful.
4.Although the social desirability effect is there because of the mere presence of the
researcher - the researcher has the opportunity to mediate it by building a rapport

WEAKNESSES

1. The study will lack objectivity because the results are not measurable or quantitative and
really on the interpretation of the researcher.
2. Furthermore, the researcher and subject are both deeply involved in the study which can lower
objectivity.
3. The data is dif cult to organise and interpret.
4. Usually two interpreters will be required to analyse/interpret the data/results.
fi
1.3 CASE STUDIES
PAPER 2 RESEARCH METHODS
Case studies
a case study is a detailed investigation of a single instance, usually just
one person although it could be a single family or institution.

The data collected is detailed and in-depth and obtained using a

variety of di erent techniques e.g. a participant maybe interviewed,
observed, given a questionnaire.

Case studies mainly collect qualitative data (QL)

useful for

looking at rare cases (Saavedra and Silverman ‘button phobia’)

studying the impact of therapy

following developmental changes

ff
Comparison with other
methods:
• In all other methods a group of participants is used
but in case studies there is only one instance
(participant)

• Experiments/correlations/self-reports using closed

questions collect quantitative data whereas case
studies collect (mainly) qualitative data

• Experiments focus on making generalisations

(nomothetic) whereas case studies look speci cally at
the way an individual is special (idiographic)
fi
 Evaluating case studies:
Strengths

ndings from case studies are highly valid - as individuals are studied in great depth and
detail

validity is also improved because of triangulation (use of multiple methods)

Weaknesses

As the sample is small and speci c the ndings can usually not be generalised.

The validity of case studies risk being compromised because of the close relationship that can
form between the researcher and participants which may reduce the objectivity of there
researcher.

Overtime there is a chance of researchers bias to occur

The close bond between the researcher and participant may also reduce reliability, as a
di erent researcher might interpret data di erently.

The level of detail can also be an ethical threat as the researcher may compromise the
participants privacy and the details might make it hard to maintain con dentiality (disguise
identity)
fi
ff
fi
fi
ff
fi
Suggest one strength of using a variety
of of techniques in a case study
Can collect qualitative and quantitative data; (advantage)

So can compare (quantitative) and look at reasons/detail

(qualitative)

• So increases validity; (detail)

• More valid as can compare di erent sources of data to
see if they demonstrate the same nding / triangulation;
(adv)

• e.g. Saavedra and Silverman could see if the feelings

thermometer and the boy’s comments were similar; (detail)
ff
fi
Strengths
In-depth and detail knowledge: Case study method is responsible for
intensive study of a unit- very detailed and in-depth study of a person or
event. It is the investigation and exploration of an event/person thoroughly
and deeply.

Valid results: - in-depth data (about the history / symptoms / therapy) can
be collected about a subject's experience.

Triangulation: Allows for the use of multiple methods - triangulation that

bring validity and reliability to the study

Application: Case studies are one of the best ways to stimulate new
research. A case study can be completed, and if the ndings are valuable,
they can lead to new and advanced research in the eld. There has been
a great deal of research done that wouldn't have been possible without
case studies. Oftentimes there are theories that may be questioned with
case studies.
fi
fi
Weaknesses
Generalisability: the sample is very small therefore generalisability is limited. If
one person is studied and they may be rare/unique/unusual, so researchers are
less likely to be able to generalise to the wider population / less representative

However, this isn't always a problem, especially if generalisation is not one of

the study's goals.The idiographic approach is concerned with
understanding unique aspects of a case in its speci c context, often using
qualitative data. Case studies are idiographic because they focus on in-
depth, detailed exploration of a single, unique case rather than generalising
to a larger population.

Reliability and subjectivity: Reliability may be a problem because the

researcher may be subjective and vary their interpretation. When conducting a
case study, it is very possible for the author to form a bias. This bias can be for
the subject; the form of data collection, or the way the data is interpreted. This is
very common, since it is normal for humans to be subjective. The researcher
can become close to a study participant, or may learn to identify with the
subject. When this happens the researcher loses their perspective as an
outsider.
fi
Practical concerns: Case studies can be very time
consuming. The data collection process can be very intensive
and long, and this is something new researchers are not
familiar with. It takes a long period of time to develop a case
study, and develop a detailed analysis.

Standardisation and replicability: The study cannot be

standardised. Many studies cannot be replicated, and
therefore, cannot be corroborated. Because the studies cannot
be replicated, it means the data and results are only valid for
that one person.

Ethics: With small studies, there is always the question of

ethics. At what point does a study become unethical? At what
point does the pursuit of scienti c knowledge outweigh the
right to a life free from research? Also, because the researchers
became so invested in the study, people questioned whether a
researcher would report unethical behaviour if they witnessed it.
fi
Laken is planning a case study of a child with a phobia of chairs. She is worried about how she
will collect data from the child as he will not sit on a chair, so she is planning to sit on the floor
with him
(a) Explain the ethical guideline Laken is following by choosing to sit on the floor.
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
...................................................................................................................................................[2]
(b) Suggest one advantage of using a case study.
.......................................................................................................................................................
.......................................................................................................................................................
……………………………………………………………………………………………………………[2]
(c) Suggest one disadvantage of using a case study.
………………………………………………………………………………………………………………
………………………………….'.......................................................................................................
...................................................................................................................................................[2]
(a) Outline what is meant by a ‘case study’.
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
...................................................................................................................................................[2]
(b) Suggest one way that Laken could collect data from the child’s parents, other than by
interview.
.......................................................................................................................................................
.......................................................................................................................................................
……………………………………………………………………………………………………………[2]
(c) Suggest why Laken chose to use the case study method in this research.
………………………………………………………………………………………………………………
.......................................................................................................................................................
...................................................................................................................................................[2]
Designing a case study
Characteristics e.g ASD / male, age,
details about
the
How was the individual found e.g
participant /
unit
Other info e.g. diagnosis, parents, teachers, participants background)

content of
e.g. depth / detail: e.g. types of doodles, frequency, e.g. associate with emotions /
information
being with friends/being outdoors, examples of rating scales etc etc.
collected

two or more
techniques for observation, interviews, questionnaires (rating scales)
data collection

qualitative and/or quantitative;

How was data
used
triangulation – making sure the sources agree to be con dent in the conclusion. (e.g.
analysis/ depth / detail: e.g. types of doodles, frequency, e.g. associate with emotions / being
interpretation/ with friends/being outdoors, examples of rating scales etc etc.)
triangulation:
Note: this is not about reliability
fi
Jason is a university lecturer and is interested in doodling behaviour. One day, Jason sees one of his
students, Karen, doodling. Jason is especially interested because Karen has a visual impairment that
means she cannot see. Jason decides to conduct a case study on her.

Describe how Jason could conduct a case study on Karen to nd out about her doodling behaviour.

To conduct a simple and replicable case study, Jason could use a mixed-method approach (triangulation), combining
one quantitative and one qualitative method to collect data.
Quantitative Method: Observation with Behavioral Categories

Jason could use structured observations as the quantitative method to objectively record Karen's doodling behavior.
He can create a set of behavioral categories to note how frequently and under what conditions Karen doodles during
lectures. Some example categories could include:
• Doodling Frequency: Number of times Karen engages in doodling per class.
• Duration of Doodling: Length of time Karen doodles during a specific period.
• Situational Context: Whether Karen is doodling during challenging topics, during lecture breaks, or during more
relaxed discussions.
Jason would observe Karen over several lectures (e.g., five sessions) and record the frequency and duration of her
doodling using a simple tally chart or a time log. These behavioral categories can later be subjected to statistical
analysis, such as calculating averages or frequencies, to quantify her doodling habits.
Qualitative Method: Semi-Structured Interview

In addition to observation, Jason could conduct a semi-structured interview with Karen to gain deeper insights into
her thoughts and motivations behind doodling. This would serve as the qualitative method, allowing Karen to explain
why she engages in doodling. Sample interview questions could include:
1. “Can you describe what you experience while you are doodling?”
2. “What emotions or thoughts do you have when doodling during class?”
3. “Does doodling help you concentrate, or is it more of a relaxing activity?”
These open-ended questions allow Karen to express her subjective experience in her own words, providing rich
qualitative data.
fi
Jason is a university lecturer and is interested in doodling behaviour. One day, Jason sees one of his
students, Karen, doodling. Jason is especially interested because Karen has a visual impairment that
means she cannot see. Jason decides to conduct a case study on her.

Describe how Jason could conduct a case study on Karen to nd out about her doodling behaviour.

Data Analysis
• Quantitative Data: The data from the observations (e.g., doodling
frequency) can be analyzed using basic statistical analysis. For
instance, Jason could calculate the average number of doodles per
class or determine if doodling frequency increases during specific
lecture contexts.
• Qualitative Data: The interview responses can be analyzed using
thematic analysis to identify recurring themes in Karen's motivations
or feelings about doodling. Jason would transcribe the interview, then
code it for key themes, such as "doodling as a focus tool" or "sensory
experience."

Ethical Considerations: Jason should ensure informed consent and

confidentiality. Given Karen’s visual impairment, he may need to provide
consent materials in an accessible format.
fi
Identify one practical weakness/limitation with the procedure
you have described in your answer to part (a) and suggest
how your study might be done di erently to overcome the
problem. Do not refer to ethics or sampling in your answer.

One practical weakness in Jason's study is the potential for

observer bias during structured observations. As Jason is
personally involved, his expectations may in uence how he
records Karen's doodling behavior, leading to inconsistent data.

To overcome this issue, Jason could implement inter-rater

reliability by having a second, independent observer trained in
the same behavioural categories observe Karen's doodling during
the same sessions. By comparing their ndings, Jason can
ensure consistency in data collection, enhancing the reliability and
objectivity of the quantitative data. This would provide a more
accurate representation of Karen's doodling habits
ff
fi
fl
1.4 OBSERVATIONS
PAPER 2 RESEARCH METHODS
observations involve watching and recording human or animal participants.

Observations can be used either as a research method or as a means to collect

data in other research methods such as case studies, experiments or
correlations. You should be able to distinguish between these two situations.

• Observation is being used as a research method in itself when the study

consists solely of a means to collect data by watching participants and
recording their behaviour directly to provide data.

• Observations are used as a technique to collect data about variables in

other research methods

◦ when they are used to measure the dependent variable in an experiment

or

◦ one or both variables in a correlation.

◦ In a case study, observations can be used alongside other techniques to

explore a single instance in detail.
Observations
Data is recorded that is observable (not inferred)

Quantitative data (objectivity, factual)

FREQUENCY (number of times something happens) and DURATION (how long it

happened for)

operationalise happiness = no of smiles

number of times someone smiles in 30 minutes (metric: no of smiles)

number of passengers in a train carriage during a 7 minute period (Piliavin et al) (metric
number of people)

the time it took for the rst helper to respond (metric: seconds)
typically gathered from structured observations

Qualitative data

verbal comments made by participants on a train (Piliavin et al)

descriptive account of the behaviour of participants in a shop

typically gathered from unstructured observations
fi
 Structured observation: researchers recorded the time taken for the rst helper to help (latency),
the number of of people who can to help.
Controlled observation: the environment was manipulated i.e a drunk/ill victim collapsed in the
train carriage
Piliavin et al Covert observation: the passengers did not know their behaviour was being recorded or that they
were part of a study
Participant observation: the confederates as the ill/drunk victim and the model interacted with the
participants through each condition

Controlled observation: took place in a laboratory where all the variables and measurements were
controlled, while the behaviour of participants was observed and recorded.
Participants level of obedience was measured through observation. This was operationalised as the
maximum voltage given in response to the orders. Observers also noted the participants body
language and any verbal comments or protests made throughout the procedure.

Milgram

fi
 This was a eld experiment, where the independent variable was gender (male or female),
observations were used as a technique to measure the dependent variable of activities with the
toys.
Hasset et al For each social group, seven trails, each lasting 25 minutes, were observed using two video
cameras. Each trial began with all the moneys in the group indoors while one plush toy and one
wheeled toy were placed 10 meters apart in the outdoor enclosure. The video tape was analysed
by two observers working together to achieve consensus.

They identi ed each animal interacting with a toy and coded speci c behaviours directed towards
the toys using a behavioural checklist. The exact time at which activity occurred was also recorded
so in addition to the frequency of behaviour, duration was also recorded.
fi
fi
fi
BANDURA ET AL (BOBO DOLL)
• Bandura et al conducted an experiment where the dependent variable (DV) was the learning the child
displayed. This was measured through a controlled observation of the children and measures of
aggressive behaviour were recorded.

• Prior to the experimental part of the study, the children were observed in their nursery school by the
experimenter and a teacher who knew them well. They were rated on four ve-point scales
measuring physical aggression, verbal aggression, aggression to inanimate objects and aggression
inhibition (anxiety). They were then assigned to three groups, ensuring that the aggression levels of
the children in each group were matched. Of the 51 children rated by both observers (the rest were
rated by only one observer), similar ratings were generally produced. Their ratings were compared as
a measure of ‘inter-rater reliability’, which showed a high correlation between the observers, of r =
0.89.

• A test of the child’s aggression then followed in which the child was observed for 20 minutes using a
one-way mirror (covert). For the aggressive model group, this was a test of delayed imitation. This
experimental room contained a three foot (92 cm) Bobo doll, a mallet and some other objects. It also
contained some non- aggressive toys, including a tea set, crayons, etc. These toys were always
presented in the same order (controlled)

• The children’s behaviours were observed in ve second intervals (240 response units per child).
There were three ‘response measures’ of the children’s imitation, with a range of possible activities in
each. Partially imitative aggression was scored if the child imitated these behaviours incompletely.
fi
fi
 • striking the Bobo doll with the mallet, sitting on the doll and punching it in the nose,
Imitation of physical
kicking the doll, and tossing it in the air.
aggression:

Imitative verbal • repetition of the phrases, ‘Sock him’, ‘Hit him down’, ‘Kick him’, ‘Throw him in the air’
aggression: or ‘Pow’.

Imitative non- • repetition of ‘He keeps coming back for more’ or ‘He sure is a tough fella’.
aggressive verbal
responses:

The two behaviours here were:

mallet aggression: • striking objects other than the Bobo doll aggressively with the mallet

sits on Bobo doll: • laying the Bobo doll on its side and sitting on it, without attacking it.

two other categories:

aggressive gun play: • shooting darts or aiming a gun and ring imaginary shots at objects in the room.

non-imitative • physically aggressive acts directed toward objects other than the Bobo doll and any
physical and verbal hostile remarks except for those in the verbal imitation category (e.g. ‘Shoot the
aggression: Bobo’, ‘Cut him’, ‘Stupid ball’, ‘Horses ghting, biting’ ‘Knock over people
fi
fi
OBSERVATIONs can be conducted in many di erent ways, for example

Structured/Unstructured

• structured (observing known categories) or unstructured

(recording any events)

Controlled/Naturalistic

• naturalistic (observing whatever is happening) or controlled

(constructing events to observe)

Overt/Covert

• The role of the observer may be obvious to the participants (overt)

or hidden (covert)

Participant/Non-participant

• the observer themselves may be part of the social situations

(participant) or not (non-participant)
ff
Bandura et al
1. Structured + controlled + non-participant + covert observation:

1. structured - there were response categories, each behaviour that fell

into one of the categories e.g. imitative verbal aggression (sock him) was
counted.

2. controlled - it was conducted in a lab where extraneous variables were

controlled and the rooms were manipulated according to the researchers
purpose, e.g aggressive and non-aggressice toys were placed

3. non-participant - the researchers did not interact with the children in

Room 3 when they were being observed (although they were exposed to
a model earlier)

4. covert - the researchers observed through a one-way mirror and the

participants did not know they were being observed
 Piliavin et al
Structured + controlled + participant + covert observation:

1. structured - the observers were measuring latency i.e. the time it took for
the rst helper to help and the number, sex and race of passengers in the
critical area etc.

2. controlled - although it was conducted in the participants everyday setting

i.e a train carriage, it was a controlled observation because the setting was
manipulated i.e. the participants were exposed to an ill or drunk victim

3. participant - the observers were part of the social setting; they pretended
to be passengers and made conversation with other passengers while
recording their observations

4. covert - the role of the two female observers was not made obvious to the
passengers
fi
 Strengths of naturalistic observation:
1. High Ecological Validity: Observing behaviors in a natural setting provides a
Naturalistic more realistic view of how participants behave in everyday contexts.
2. Reduced Demand Characteristics: Participants are often unaware they’re
Conducted in the participants being observed, so they’re more likely to act naturally, enhancing the validity of
normal envirnment without the findings.
interference from researchers in
either the social or physical Weaknesses of naturalistic observation:
environment
1. Low Reliability: Lack of control over variables can make it difficult to replicate
the study or achieve consistent findings, reducing reliability.
2. Potential Ethical Issues: Observing people in natural settings without their
knowledge may raise ethical concerns, particularly around informed consent
and privacy.

Strengths of controlled observation:

Controlled
1. High Reliability: Controlled settings allow for precise control over variables,
Conducted in a situation that making the study easier to replicate and results more consistent.
has been manipulated by the 2. Internal validity: By controlling external variables, researchers can focus on
researchers - either social (using specific behaviours or factors, making it easier to isolate cause-effect
relationships.
a model) or physical
environment (choosing toys).
Weaknesses of controlled observation::
Controlled observations can be
done in either the participants 1. Low Ecological Validity: The artificial setting may not reflect real-life
normal environment or an conditions, so behaviours observed may not accurately represent how
arti cial environment participants act in natural environments.
2. Demand Characteristics: Participants may be aware they’re being observed,
which could alter their behaviour, impacting the validity of the findings.
fi
 Strengths of unstructured observations:

Unstructured 1. High Validity: Researchers record all behaviours as they occur, capturing the
complexity and natural flow of interactions for richer data.
the observer records the whole 2. Flexible and Adaptive: Allows researchers to explore new behaviours or
range of possible behaviours, patterns that emerge, providing a more holistic view of participants.
which is usually con ned to a
pilot stage at the beginning of a Weaknesses of unstructured observations:
study to re ne the behavioural
categories to be used 1. Low Reliability: Without a structured system, observations can vary widely
between researchers, making replication challenging.
2. Data Overload: The abundance of unfiltered data can complicate analysis, and
important insights may be lost in the process of managing extensive
information.

Strengths of structured observations:

Structured
1. High Reliability: Using a standardised coding system or checklist ensures
a study in which the observer behaviours are consistently recorded, making the study more replicable.
records only a limited range of Behavioural categories improve inter-observer reliability (the consistence
behaviours that are de ned in between two researchers watching the same event i.e. whether they will
behavioural categories (the produce the same records)
2. Easier Data Analysis: Predetermined categories streamline data collection and
activities recorded in an
analysis, enabling quantitative analysis for trends or patterns.
observation - they should be
operationalised and break the Weaknesses of structured observations :
continuous steam of anti city
into discrete recordable events- 1. Reduced Validity: Focusing on specific behaviours may overlook important,
must be observable actions spontaneous actions, potentially limiting the depth of the findings.
2. Limited Flexibility: Structured observations don’t allow researchers to adapt
rather than inferred states)
categories during the study, which might restrict insights into unexpected
behaviours.
fi
fi
fi
 Strengths of Participant Observations:

1. Increased Validity: Being a participant allows researchers to gain a deeper,

first-hand understanding of group dynamics, enhancing the internal validity of
participant observer the data collected.
2. Insight into Hidden Behaviours: As a participant, the researcher can access
an observer who watches from behaviours and interactions that might otherwise be hidden or altered in the
the perspective of being part of presence of an observer, providing richer data.
the social setting e.g may
engage in conversation with Weaknesses of Participant Observations:
adults, or play with children
1. Low Reliability: Each researcher’s experience as a participant may differ due to
subjective interpretation or situational differences, making the study hard to
replicate reliably.
2. Ethical Issues (in Covert Studies): Deception may be required to maintain the
researcher’s cover as a participant, raising ethical concerns

non-participant Strengths of Non-Participant Observations:

1. Higher Reliability: Researchers remain uninvolved, allowing them to objectively
a researcher who does not record behaviours, which can lead to more consistent findings and make replication
become involved in the situation easier.
being studied. 2. Fewer Ethical Concerns: Observing from a distance (especially in public settings)
reduces the need for deception, addressing ethical concerns like informed consent
This can be achieved by and right to withdraw (depends on whether the researcher is overt or covert)
watching through a one way
glass, or by keeping apart from Weaknesses of Non-Participant Observations:
the social group of the 1. Lower Validity: Observing from a distance may mean missing subtle or contextual
participants for example by details, potentially impacting the depth and accuracy of data collected.
sitting in a separate place such 2. Limited Insight into Participants' Perspectives: Without direct involvement,
as a bench in the park researchers may find it challenging to fully understand participants' experiences or
motivations, reducing the richness of the data.
 Strengths of overt:
1. Ethically Sound: Participants know they’re being observed, so informed
consent is obtained, making the study ethically robust.
2. Practical: Since participants are aware of the observation, it is easier to record
data and researchers can openly discuss the study with them afterward, which
overt observer can provide additional insights and verify findings.

the role of the observer is

obvious to the participants Weaknesses of overt observation:
1. Demand Characteristics: Awareness of being observed may lead participants
to alter their behaviour, potentially impacting the validity of the findings.
2. Limited Authenticity: Participants may consciously or unconsciously behave
in a socially desirable way, which could limit the observation of genuine,
spontaneous behaviours.

covert observation Strengths of covert observations:

participants are unaware of the 1. Validity: as participants are not aware of being observed their behaviour is
true nature of the observation/ unlikely to be affected by being observed, so demand characteristics and social
the role of the observer is not desirability are reduced.
obvious
Weaknesses of covert observations:
e.g because they are hidden or 1. Ethical issue: partipcants are being watched without their permission which
disguised or they are are breaks the ethical guidelines of consent, they are also denied the right to
unaware of the role of the withdraw, also if their work out the role of a participant covert observer it may
observer (have not given cause them distress, it is also difficult and sometimes impossible to debrief
informed consent- are being participants.
deceived) 2. Practical issue: the observer must be either hidden, far away or disguise their
role, they may make data collection more difficult, potentially reducing the
validity and reliability of the study.
 Evaluation
refers to the degree of agreement or consistency
between di erent raters or judges when assessing
the same subject, task, or phenomenon.
inter observer reliability
- across di erent
researchers
Reliability
consistency
behavioural categories ensure that researchers
interpretation is not required while recording data
behavioural categories therefore making the data collection process
- prede ned/ pre objective and the results reliable i.e. if someone
determined categories else counted the same instances they would get
the same results

Naturalistic observations record participants

behaviour as it comes naturally and include
everything that occurs rather than recording only
what was ‘expected’ the results can therefore have
depth and detail and therefore validity.
Validity naturalistic behaviour
accuracy

In covert observations the participants behaviour is

likely to be more natural and therefore the results of
the study a more valid depiction of their actual
behaviour.
fi
ff
ff
of controlled observation:
Designing an observational study:
Where pick a reasonable location e.g train station, school, hospital, library, mall - depending on the question

sampling technique (unless the question states it is not required): opportunity/volunteer/random

Who sample size (n=?)

indicate some kind of time frame

When School: 30 minutes every other day (Mon, Wed, Fri) during break time
Train station: 30 minutes every morning and evening, thrice a week, for two weeks

Participant/Non-participant: is the researcher interacting with, or part of the social group of the
How participants?
Covert/Overt: are the participants aware of being observed?

Structured/Unstructured: are there pre-de ned categories (behavioural categories?

e.g the researcher will tally the number of times a participant was verbally aggressive
(frequency), record the length of time time in seconds each child played with a toy, etc
Data time intervals: Bandura et al. The 20 minute session was divided into 5-second intervals
Collection by means of an electric interval timer, thus yielding a total number of 240 response units for
each subject.
Controlled/Naturalistic: did the researcher manipulate/change the envirnment (whether in the lab
or the eld) in any way?

Data Quantitative Data: Statistical analysis

Analysis Qualitative Data: thematic analysis

Ethical considerations: if required by the question i.e consent, no harm, debrie ng etc.
Evaluation Validity: improves if participants natural behaviour is recorded
Reliability: inter-observer reliability (using two observers)
fi
fi
fi
Jim is planning an observational study about the way people respond to facial expressions. He intends
to go to a train station and either smile or frown at people. He wants to see how people respond to him.

(a) Describe how Jim could conduct an observational study at a train station to nd out how people
respond to smiling and frowning.
To conduct an observational study on responses to facial expressions, Jim could implement a covert,
structured, participant observation in a controlled setting at a train station.
SAMPLING: Using opportunity sampling, Jim would record responses from individuals he encounters
during each interval.
WHEN: Jim will go to a train station every Monday, Wednesday, Saturday, morning (9am-10am) and evening
(5pm-6pm) to collect data on a diverse sample of people.
HOW: He would engage with individuals without disclosing his role as a researcher (covert). While people
would see him smiling or frowning, they would not know they are part of a study, minimising demand
characteristics and promoting authentic reactions. By smiling or frowning, Jim actively manipulates the social
situation to observe immediate and unaltered responses to his facial expressions (controlled).
DATA COLLECTION: Jim would use a structured observation approach by creating a pre-defined
checklist/behavioural categories of potential responses, such as ‘smiling back’, ‘frowning’, ‘looking away’,
or verbal responses. This structured design enhances reliability, ensuring consistent data collection across
all participants and making the study easier to replicate.
The procedure would involve defining specific time intervals (e.g., every 15 minutes) to alternate between
smiling and frowning. He would approach participants from a set distance (e.g., five feet away), maintaining
uniformity in interactions by making brief eye contact and displaying a distinct smile or frown for
approximately three seconds. He will set another control by wearing the same clothes for each observation.
DATA ANALYSIS: in a note book he will record how the participant responds e.g ‘smiling back’ and then
tally/count and make a table of the frequency of each response. He will then make a bar chart, where on the
x axis will be the different behavioural categories, and on the y axis the no of times each category was
observed. He will then be able to statistically compare the participants reaction to smiling and frowning.
fi
Dr Birkin is planning an observational study of children’s
imaginary play behaviour (for example, a child might use a
leaf as a plate, or a big cardboard box as a boat or a house).

(a) Describe how Dr Birkin could conduct an observational

study of children to investigate their imaginary play
behaviour.

(b) Identify one practical weakness/limitation with the

procedure you have described in your answer to part (a) and
suggest how your study might be done di erently to
overcome the problem.

Do not refer to ethics or sampling in your answer.

ff
 smiling back frowning looking away verbal response

smiling 15 5 25 10

frowning 5 15 30 5

30

23

15

0
smiling back frowning looking away verbal response
1.5 CORRELATIONS
PAPER 2 RESEARCH METHODS
 De nition: Correlational analysis is a statistical technique used to examine the relationship between
two measured variables (co-variables).

• A correlation is a relationship / link between two measured variables;

• A positive correlation – as one variable increases, the other also increases (has a coe cient
above 0 up to +1)

• A negative correlation – as one variable increases, the other decreases (has a coe cient below
0 down to -1)

Purpose: It identi es patterns and trends between variables but does not establish cause and
e ect.

Use Case: Especially useful when variables cannot be manipulated (as in an experiment) due to
ethical or practical reasons.

Example 1: Measuring children's exposure to violent television and correlating it with their
aggression levels in school, since experimentally increasing exposure would be unethical.

Example 2: Studying the link between sleep duration and academic performance, as it might be
impractical to control students' sleep patterns directly.

Key Insight: Helps researchers understand potential links between variables in real-world settings.
ff
fi
fi
ffi
ffi
 Measuring variables in correlational
studies:
NUMERICAL DATA - Quantitative Data

each variable must exist over a range, sometimes called ‘continuous data’, and it must be possible measure them
numerically.

E.g age = continuous data

E.g Gender, kinds of religious membership, = you cannot - separate categories

Variables can be:

duration of time (seconds, minutes, hours, days, years)

totals of tallies (acts of verbal aggression, smiles, sneezes, jumps)

numbers from a rating scale (how happy are you? 0-5, 0= not at all, 5= extremely happy)

test scores (5/10, 9/10)

Di erent techniques can be used to collect data for correlations

self reports (questionnaires and interviews: close ended questions)

observations (structured - where there is QT data)

di erent kinds of tests and tasks (IQ test)

measurement tool (pulse meter)

ff
ff
positive correlation: A positive correlation is a relationship between two variables that
move in the same direction. A positive correlation exists when one variable decreases
as the other variable decreases, or one variable increases while the other increases.

negative correlation: A negative correlation is a relationship between two variables that

move in opposite directions. In other words, when variable A increases, variable B
decreases. A negative correlation is also known as an inverse correlation.
Examples
e.g. Piliavin et al. found that as group size increased the
likelihood of helping also increased (positive);

e.g. Dement and Kleitman found dream duration positively

correlated with number of words in dream narrative (positive);

e.g. Baron-Cohen et al. found that as AQ score increased,

eyes test scores decreased (negative / inverse);

e.g. if a researcher tested helping and age, they might nd

that the older you are, the more helpful you are (positive);

e.g. a correlation of obedience and education, might nd that

as education increased, obedience decreased (negative);
fi
fi
correlations can guide ideas for (future)
experimental work

can study variables which cannot practically be

manipulated (such as activity levels in dreams)

can study variables which cannot ethically be

manipulated (such as phobia strength)

correlations cannot determine causal

relationships between the variables being tested;
i.e. whether one causes the other to changed; or
whether they are both dependent on a third factor
What is the difference between
causation and correlation?
in a correlation a change in one variable is related to the
change in the other variable, although these cannot be
assumed to be causal.

In a causal relationship, a change in one variable is

responsible for (i.e causes) the change in the other variable.

Experiments can be used to establish causal relationships

between the IV and the DV because changes in the IV are
shown to be the cause of the changes in the DV.

If however, two variables do not correlate we can be certain

there is no causal relationship between them.
For example, researchers have found a correlation between the
amount of violent television children watch and how aggressive
they are.

1. One explanation of this correlation is that watching TV

violence causes kids to become more violent themselves.

2. It is equally probable, however, that the reverse is true: that

kids who are violent to begin with are more likely to watch
violent TV.

3. Or there might be no causal relationship between these two

variables; instead, both TV watching and violent behaviour
could be caused by a third variable, such as having
neglectful parents who do not pay much attention to their
kids.
A laboratory experiment and correlational study are di erent in the following ways:

• Variables:
◦ In correlational research we do not (or at least try not to) in uence any variables but only measure them and
look for relations (correlations) between some set of variables, such as blood pressure and cholesterol
level.

◦ In experimental research, we manipulate some variables and then measure the e ects of this manipulation
on other variables; for example, a researcher might arti cially increase blood pressure and then record
cholesterol level.

• Relationship between Variables:

◦ A laboratory experiment seeks to study a cause-e ect relationship. Hence, one variable - the independent
variable - is clearly de ned as the cause, and the other - the dependent variable - as the e ect.

◦ A correlational study, however, just investigates whether two variables are interdependent i.e. whether they
are related to each other. There is, however, no expectation of causation in such studies.

◦ If a researcher nds that there is a correlation between two variables, it means that there is a possible
causal relationship etween these variables.

• Data Analysis:
◦ Data analysis in experimental research also comes down to calculating “correlations” between variables,
speci cally, those manipulated and those a ected by the manipulation.

◦ However, only experimental data can conclusively demonstrate causal relations between variables. For
example, if we found that whenever we change variable A then variable B changes, then we can conclude
that “A in uences B.”

◦ Data from correlational research can only be “interpreted” in causal terms based on some theories that we
have, but correlational data cannot conclusively prove causality.
fi
fl
fi
b​​
fi
ff
ff
fi
ff
fl
ff
ff
 Correlational analysis Experiments
one variable is manipulated (IV)
two measured variables (e.g,
and the other variable is
variable 1 and variable 2)
measured (DV)

establish that two variables co- establish a causal relationship

vary between the IV and DV

RELATIONSHIP CAUSAL RELATIONSHIP

explanations
describe
(state it’s cause)
Hypotheses in correlational
studies
A hypothesis in a correlational study needs to
operationalise the co-variables as it predicts what
there will be be a relationship between them.

For example:

There will be a correlation between the amount of

sleep and emotional reactivity

There will be a correlation between the number of

hours a person sleeps and their emotional reactivity
as indicated by how loudly they cry during a sad lm.
fi
 Type of
Definition Example
Hypothesis

"There will be a correlation between

Predicts that there will be a correlation
Non-Directional the amount of time spent on social
between two variables, but does not
Hypothesis media and levels of anxiety in
specify the direction of the relationship.
teenagers."

"Increased time spent on social media

will be positively correlated with higher
Predicts not only that there will be a
levels of anxiety in teenagers.”
Directional correlation between two variables but also
Hypothesis specifies the direction of the relationship
"Increased time spent on social media
(positive or negative).
will be negatively correlated with higher
levels of anxiety in teenagers."

"There will be no correlation between

States that there will be no correlation
Null Hypothesis time spent on social media and levels of
between the two variables being studied.
anxiety in teenagers."
 Non-Directional
Study Example Directional Hypothesis Null Hypothesis
Hypothesis

"There will be a
"Increased social media use "There will be no correlation
Social Media correlation between
will be positively correlated with between social media use
and Anxiety social media use and
higher anxiety levels." and anxiety levels."
anxiety levels."

"There will be a
"Higher exercise frequency will "There will be no correlation
Exercise and correlation between
be negatively correlated with between exercise frequency
Stress exercise frequency and
stress levels." and stress levels."
stress levels."

"More hours of sleep will be

positively correlated with
Sleep and "There will be a higher academic performance.”
"There will be no correlation
correlation between
Academic sleep duration and “There will be a positive
between sleep duration and
Performance academic performance."
academic performance." correlation between (V1)
academic performance and
(V2) hours of sleep”
"There will be a "Higher income levels will be "There will be no correlation
Income and Job correlation between
positively correlated with between income level and
Satisfaction income level and job greater job satisfaction." job satisfaction."
satisfaction."
 Non-Directional
Study Example Directional Hypothesis Null Hypothesis
Hypothesis

"There will be a correlation

"Increased hours per day spent on "There will be no correlation
between the hours per day
Social Media social media will be positively between the hours per day spent
spent on social media and
and Anxiety anxiety levels as measured by
correlated with higher self-reported on social media and self-reported
anxiety scores." anxiety scores."
self-reported anxiety scores."

"There will be a correlation "Higher exercise frequency per week "There will be no correlation
Exercise and between exercise frequency per will be negatively correlated with between exercise frequency per
Stress week and stress levels as stress levels as measured by a week and stress levels as
measured by a stress scale." stress scale." measured by a stress scale."

Sleep and "There will be a correlation "There will be no correlation

"More hours of sleep per night will
between hours of sleep per night between hours of sleep per night
Academic and academic performance as
be positively correlated with higher
and academic performance as
Performance GPA."
indicated by GPA." indicated by GPA."

Income and "There will be a correlation "Higher annual income will be "There will be no correlation
between annual income and job positively correlated with higher job between annual income and job
Job satisfaction as indicated by a satisfaction as indicated by a satisfaction as indicated by a
Satisfaction satisfaction survey score." satisfaction survey score." satisfaction survey score."
Evaluating correlations
Validity - (de nition): a correlational study can only
be valid if the measures of both variables test real
phenomena in e ective ways - to achieve this the
variables must be clearly de ned and relate directly
to the relationship being investigated.

Reliability - (measurement): of a correlational study

depends on the measures of both variables being
consistent - which depends on whether the data
collected through for example self reports or
observations was done in an objective way.
fi
ff
fi
Evaluating correlations
Validity - (de nition)

pulse rate -

self report - (0-5) scale - MORE VALID

Reliability - (measurement): objective

pulse rate : MORE RELIABLE - does not depend on

any persons interpretation therefore objective

self report: depends on the respondents interpretation

of their own anger can be a little subjective
fi
Scatter graphs
The results of a correlational study are displayed on a scatter
graph.

To construct a scatter graph, a dot is marked at the point where an

individuals scores on each variable cross.

Sometimes, a ‘line of best t’ is drawn so it comes close to as

many points as possible.

It is important to remember that you cannot draw a causal

conclusion from a correlational study. Therefore, scatter graphs
such as the ones in only tell you that there is a relationship between
the variables but not which (if either) of them, is the cause of this
link. An experiment could help to nd this out.
fi
fi
Strong and Weak
correlations
In a strong correlation all the data points lie close
to the line, but in a weak correlation they are more
spread out.

The strength of a correlation is described as a

number - an r value from +1 to -1.

Values closer to +1 are strong positive correlation

and values closer to -1 are strong negative
correlation.

Lower or smaller values (closer to 0) are weaker

correlation- this is when the points do not form a a
clear line - and there is no signi cant correlation.

It is important to remember that you cannot draw a

causal conclusion from a correlational study.
Therefore a scatter graph can only tell you that
there is a relationship between co-variables but not
which (if either) of them is the cause of this link. An
experiment can be conducted to help nd this out.
fi
fi
 Design question
Three majors are:

• What:
• variable 1 (correct operationalisation and quanti cation of rst correlational variable e.g. a repetitive
behaviour)

• variable 2 (correct operationalisation and quanti cation of second correlational variable e.g. understanding
of lesson)

• How: –technique for producing/collecting data i.e. procedure (e.g. tests, observations, questionnaires).
The minors are:

• where: location of participants when completing the questionnaire / how it is distributed

• who: participants, sampling technique
Also:

• a statement about whether a positive or negative correlation is expected

• sampling technique/ sample size

• description of how data will analysed, e.g. use of scattergram

• ethical issues
fi
fi
fi
1.6 LONGITUDINAL STUDIES
PAPER 2 RESEARCH METHODS
 Longitudinal studies
a research method that follows a group of participants
(cohort) over time, week to decades, looking at changes
in variables to explore development or changes due to
experiences, such as interventions, drugs or therapies

Cohort: a group of participants selected at the same age

or stage

Is di erent from a cross sectional study: compares

people at di erent stages or ages by comparing di erent
groups of participants at one point in time.
ff
ff
ff
Differences between a longitudinal
study and cross sectional study:
• A cross-sectional study in psychology is a research design that involves the collection of
data from a group of participants at a single point in time.

• Cross-sectional studies are valuable for assessing the state of a population or phenomena
at a given moment but cannot establish causal relationships or investigate changes over
time, as they do not involve follow-up measurements. This design is often used to examine
the prevalence of certain traits, conditions, or behaviours within a population and to identify
potential associations or di erences between various variables at that particular time.

• In longitudinal studies, researchers can observe changes in behaviour or development over

time and examine how di erent variables may in uence these changes. In contrast, cross-
sectional studies can provide a snapshot of how di erent groups of people di er in terms of
a particular behaviour or trait, but they cannot show how individuals change over time.

• Longitudinal studies tend to be more time-consuming and costly than cross-sectional

studies, but they can provide more robust data on how di erent variables interact over time.
Cross-sectional studies can be quicker and easier to conduct, but they have limitations in
terms of causal inference and generalisability.
ff
ff
fl
ff
ff
ff
Longitudinal studies & Cross
sectional studies
Longitudinal Studies:

• Design: Similar to a repeated measures design, longitudinal studies follow the same participants over a period of time,
collecting data at multiple points.

• Strengths: This approach allows for examining changes in individuals over time, providing insights into development, progression,
and causal relationships. There are fewer participant variables as the same participants are studied throughout the process.

• Weaknesses: They can be time-consuming, expensive, and subject to participant dropouts (sample attrition), which may bias
results. There can be order e ects, if for example participants are lling the same questionnaire repeatedly over time.

• Example: Tracking a group of children’s cognitive development from early childhood through adolescence.

Cross-Sectional Studies:

• Design: Resembling an independent measures design, cross-sectional studies compare di erent groups of participants at one
point in time, rather than following the same individuals.

• Strengths: These studies are quicker and more cost-e ective than longitudinal studies, as they gather data in a single instance,
and are useful for identifying correlations between variables across age groups or populations.

• Weaknesses: They do not track changes over time and may be a ected by cohort e ects, where di erences between groups are
due to generational or cultural factors rather than age or development i.e. the results may be a ected by participant variables.

• Example: Comparing cognitive abilities in 5-year-olds, 10-year-olds, and 15-year-olds in a single study.
ff
ff
ff
fi
ff
ff
ff
ff
 EXAMPLES

6 weeks
baseline/pre intervention Intervention Post-intervention

Measures: Re-take Measures:

Testing a drug:
-Psychometric test -Psychometric test
anti-depressant
BDI BDI
(medication)
- Interviews - Interviews

0 1 2 3 4 5 6 7 8 9 10 11 12
month month month month month month month month month month month month month

LONGITUDINAL STUDIES: CROSS SECTIONAL STUDIES:

Studying the same women (n=10) Studying different women at each

over a year at different time points (2 at 1 month, 2 at
time points (e.g every month giving them 2 months 2 at 3 months, 2 at 4 months)
a questionnaire)
 Strengths:

Enable researchers to track changes. Longitudinal studies are the most valid test of developmental changes because
in cross sectional studies there may be confounding variables.

Allow researchers to determine/ provide evidence for the e cacy of interventions by comparing data pre and post an
intervention

As the group of participants are the same (cohort), participant variables do not confound the results improving the
validity of the study.

Longitudinal studies like repeated measure retest the same individuals. this means the researcher can be con dent that
any di erences found (over time) are not because of participant variables.

Weaknesses:

It can be di cult to standardise the measures used at di erent time points (e.g. interviewers might change if a
researcher leaves, new and improved measures may be introduced, etc) which impacts the reliability of the study.

Over time the researcher might form a relationship with the participants which may bias the results.

To improve reliability the researcher might test participants repeatedly over time using the same measures (e.g. BDI)
this may lead to demand characteristics being revealed or order e ects (practise e ects or fatigue e ects)

Participants can drop out of the study over time for many di erent reasons this sample attrition reduces the
representativeness of the sample there by reducing the generalisability of the study. Longitudinal studies may not be
generalizable to other populations or contexts, as they are typically focused on a speci c group of participants over
time.

They may also be in uenced by historical or cultural factors that change over time, making it di cult to interpret
changes in behaviour or development as due to the variables of interest.

If children were initially recruited then consent was likely taken from parents or guardians, as the individual grows they
might withdraw their consent.

It can be di cult to maintain con dentiality as researchers need to store personal information in order to be able to re-
contact participants.
ff
ffi
ffi
fl
fi
ff
ffi
ff
ff
ff
fi
ffi
ff
fi
Jack is conducting a longitudinal study on the effectiveness of CBT to treat Generalised Anxiety
Disorder. He conducts an experiment where he compares a group of subjects who were given
CBT for 12 weeks with another group that were given befriending sessions. He collects data
about them at baseline, then immediately after treatment, and then 9-months post treatment.
(a) Describe what is meant by a ‘longitudinal study’, using an example from Jack’s study. .
.......................................................................................................................................................
………………………………………………………………………………………………………………
...................................................................................................................................................[2]
(b) Explain the advantages of using the longitudinal method in Jack’s study.
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
.......................................................................................................................................................
.......................................................................................................................................................
……………………………………………………………………………………………………………[4]
(c) Suggest one practical difficulty Jack might face in conducting his study.
.......................................................................................................................................................
...................................................................................................................................................[2]
(d) Define "baseline measurement" and explain its importance in Jack's study.
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
...................................................................................................................................................[2]
(e) Identify and describe one potential extraneous variable that could affect Jack’s study.
.......................................................................................................................................................
.......................................................................................................................................................
……………………………………………………………………………………………………………[2]
(f) Jack collects data at three different time points: baseline, post-treatment, and 9 months
after treatment. Explain why it is important for Jack to include the 9-month follow-up in his
study design.
………………………………………………………………………………………………………………
.......................................................................................................................................................
...................................................................................................................................................[2]
(g) If Jack finds that the CBT group shows more improvement at the 9-month follow-up
compared to the befriending group, what conclusions might he draw about the long-term
effectiveness of CBT?.
………………………………………………………………………………………………………………
………………………………………………………………………………………………………………
...................................................................................................................................................[2]
(h) Evaluate the use of self-report questionnaires as Jack’s method of data collection for
measuring anxiety levels.
.......................................................................................................................................................
.......................................................................................................................................................
……………………………………………………………………………………………………………[2]
(i) Critically assess one ethical issue that Jack must consider in his study and explain how he
could address it.
………………………………………………………………………………………………………………
.......................................................................................................................................................
...................................................................................................................................................[2]
(j) In Jack's study, the group receiving befriending sessions serves as the control group. Explain the role of a
control group and why it is important in this experiment

.………………………………………………………………………………………………………………………..……………

.……………………………………………………………………………………………………………………………………...

.…………………………………………………………………………………………………………………………………[2]

(k) Discuss one way in which Jack can improve the reliability of his measurements.

.………………………………………………………………………………………………………………………..……………

.……………………………………………………………………………………………………………………………………...

.…………………………………………………………………………………………………………………………………[2]

(l) If Jack's results show that both the CBT and befriending groups improved significantly from baseline to the post-
treatment measurement, what type of statistical analysis could he use to determine if the improvement in the CBT
group is significantly greater than in the befriending group?

.………………………………………………………………………………………………………………………..……………

.……………………………………………………………………………………………………………………………………...

.…………………………………………………………………………………………………………………………………[2]

(l) Jack's data show that some participants dropped out before the 9-month follow-up. How could attrition impact
the validity of his results?

.………………………………………………………………………………………………………………………..……………

.……………………………………………………………………………………………………………………………………...

.…………………………………………………………………………………………………………………………………[2]
Dr Felix believes that adults between the ages of 55 and 65 cope less well with their job regardless of whether they stay in the same job or
change jobs. Dr Felix plans to use a sample of adults aged 55 at the beginning of the study. She will re-contact the participants by telephone
as she has a record of each individual’s telephone number and will be following appropriate ethical guidelines.

(a) Describe how Dr Felix could conduct a longitudinal study to investigate how well adults between the ages of 55 and 65 cope with their jobs.
Do not describe how Dr Felix would re-contact her participants, the sample/sampling technique or ethical issues/guidelines in your answer.

Dr. Felix will conduct a longitudinal study that spans ten years, collecting data at multiple points to observe changes in coping mechanisms of adults
between the ages of 55 and 65 cope with their jobs over time.
Tests/Tasks: Measures Used
At the beginning of the study (cohort: age 55), Dr. Felix would administer both self-report: questionnaires and interviews. A suitable tool for
measuring coping could be the close ended questionnaire, which quantifies stress levels, coping strategies, job satisfaction, and emotional well-
being. The questionnaire would include questions like:
"On a scale of 1-10, how well do you think you handle stress in your current role?"
"How frequently do you experience job-related anxiety?" (never, rarely, sometimes, often, always)
Additionally, Dr. Felix could conduct semi-structured interviews to gather qualitative data on how participants feel about their work and whether
they perceive any age-related challenges, providing richer insight into their personal experiences.
Scoring: How Data Will Be Scored or Analysed
For the quantitative data collected from the questionnaire, Dr. Felix could total the scores for each participant. This would enable her to track changes
in the scores over time. She could calculate averages for each time point to determine general trends in coping among the sample. For the qualitative
data gathered from interviews, Dr. Felix would use thematic analysis to identify recurring themes, such as "increased job stress with age" or
"improved work-life balance."
Frequency/Interval: Pattern of Testing Over Time
Dr. Felix would re-test participants at regular intervals—every two years—over a ten-year period (at ages 57, 59, 61, 63, and 65). By using the same
tests and interview questions during each data collection point, she can ensure consistency in the data collection process and accurately measure
how participants' coping mechanisms evolve. The two-year intervals will allow sufficient time for any changes in job circumstances (e.g., job change or
job retention) to influence coping strategies.
Controls/Standardisation:
To ensure consistency, Dr. Felix should standardize the testing conditions across all time points. This could include using the same questions and
scoring system at each data collection point to maintain reliability. Additionally, interviews should be conducted in similar settings (e.g., over the phone
or in a private office) to reduce environmental variations that could influence participant responses. Dr. Felix would also provide the same instructions
to each participant at every stage to avoid introducing variability in how the tasks are approached.
 Describe one practical/methodological strength of the procedure you have
described in your answer to part (a). Do not refer to re-contacting the
participants, sampling or ethics in your answer.

One practical weakness of Dr. Felix's study is the potential for social desirability bias
in self-report questionnaires, where participants may give responses they think are
more socially acceptable rather than their true feelings. This can result in inaccurate
data regarding coping mechanisms and job satisfaction which would reduce validity.

To overcome this issue, Dr. Felix could ensure that the questionnaire allows
participants to be anonymous which may encourage participants to provide more
honest responses thus enhancing the validity of the data collected.

Explain why the feature of the procedure you have identi ed in (i) is a strength.
Do not refer to re-contacting the participants, sampling or ethics in your answer.
(2 marks)

The use of anonymous data collection methods in Dr. Felix's study is a strength
because it encourages participants to provide more honest and accurate responses
about their coping mechanisms and job satisfaction. By minimising the fear of
judgment, participants are likely to share their true feelings, leading to more reliable
data.

This, in turn, enhances the validity of the results, as the collected data will better
re ect the participants' genuine experiences over time.
fl
fi
1.7 THE DEFINITION,
MANIPULATION, MEASUREMENT
AND CONTROL OF VARIABLES
PAPER 2 RESEARCH METHODS
 Aim of a study
• The intention of the study, the idea being tested or the purpose of
the research, such as to investigate a question or solve a problem.

• The aim tells you the purpose of the investigation. It is generally

expressed in terms of what the study intends to show.

• The aim of an experiment is the purpose of the study, which is typically

to investigate causal questions within a controlled setting.

• In self report, the aim is to explore and understand the subjective

experiences, attitudes, or behaviours of individuals based on their own
perceptions or accounts, without external observation or inference.

• In a correlation study, the aim is to examine and identify the potential

association or connection between two measured variables, elucidating
any potential relationship without implying causation, such as between
the number of computer games a student plays and their nal A Level
grade.
fi
Variables
variables are factors that change or can be changed

in experiments

independent variable

dependent variable

controlled/uncontrolled variables

in correlations

two co-variables
Operationalising variables
Variables must be operationalised

operationalisation: the de nition of variables so that they can be accurately manipulated,

measured or quanti ed and replicated. This includes the IV and DV in experiments and the two
measured variables in correlations.

When writing hypotheses, you should ideally operationalise the variables you are referring to.

In an experiment for example testing the e ect of age on susceptibility to false memories.

The IV would be age, with, for example, ‘young’, ‘middle-aged’ and ‘old’ groups. It is
important to know how old the people in the groups are; this is operationalisation. You
might operationalise

‘young’ as under 20 years old,

‘middle aged’ as 40–50 years old and

‘old’ as over 70.

The DV must also be operationalised, so it can be measured e ectively. We could

operationalise the DV by counting the number of details ‘remembered’ about the false
memory or how convinced the participants were that it was true.
fi
fi
ff
ff
 Controls
• a way to keep a potential extraneous variable constant, e.g. between levels
of the IV, to ensure measured di erences in the DV are likely to be due to the
IV, raising validity.
• A controlled variable in an experiment, is a factor or condition that is intentionally kept consistent
and unchanged throughout the experiment.
Controlled variable
• This is done to ensure that any observed e ects or di erences in the dependent variable(s) are the
result of the independent variable(s) being manipulated and not due to variations in other factors.
• To remove its e ects a control variable may be held at a constant level during the study or managed
by statistical means

• a variable which either acts randomly (extraneous), a ecting the DV in all levels of the IV or
systematically, i.e. on one level of the IV (called a confounding variable) so can obscure the e ect
Uncontrolled of the IV, making the results di cult to interpret.
variable • Good control of extraneous variables raises validity
• Extraneous variables are those uncontrolled variables that have a consistent e ect on the DV, and
therefore confound/confuse the results.

• Confounding variables act on the DV selectively in one level of the IV so can interfere with the
results one of two ways.
Confounding • Confounding variables can work against the e ect of the IV, counteracting its e ect on the
DV, thus preventing the real e ect from being identi ed.
variables
• Alternatively, they can increase the apparent e ect of the IV and so suggest that there is an
e ect that doesn't really exist.
• These variables are the most important to control.
ff
ff
ff
ffi
ff
ff
ff
ff
fi
ff
ff
ff
ff
ff
 • an experimental procedure where neither the participant nor the researcher
is aware of which condition the participant is in (1 mark)
double-blind:
• so a participant could be in the nalmefene (or naltrexone) group or the
control (placebo) group and not know which group they are in (2 marks)

• a control group where no test medication is received but where the

medication given has no actual e ect, but the participant may think they are
placebo-
receiving the test medication (1 mark) so a participant could be in the
controlled:
nalmefene (or naltrexone) group or the control (placebo) group and not know
which group they are in (2 marks).

• Other extraneous variables, which have a random e ect on the DV across all
levels of the IV, are not so problematic.
• The di culty is to identify which variables it will be important to control
Pilot study
before the experiment starts.
• This is one function of a pilot study, a preliminary test of the procedures of a
study.
ffi
ff
ff
 Situational variable
Experiments
- it is important to control extraneous -a factor of the environment e.g the lighting in
variables a class room
- - extraneous variables that have a
consistent e ect are called
Controlling confounding variables Participant variable
variables - -Other extraneous variables, which
have a random e ect on the DV -features of the individual e.g a participant
ability to do math
across all levels of the IV, are not so
problematic.
-

In order for a study to be valid, it is important that every participant is treated in

the same way, this is the process of standardisation.
Standardisation
of procedure • Standardised instructions: the written or verbal information given to
participants at the beginning and sometimes during a study that ensures the
experience of all participants, regardless of level of the IV, is as similar as
possible.
• Standardised procedure: involves having equipment or tests that are
consistent i.e. that measure the same variable every time and always do so in
the same way.
ff
ff
Describe ‘participant variables’ and ‘situational variables’ in research, using any examples. (6 marks)

Participant variable – confounding variable due to individual di erences;

can be temporary e.g. hunger in a study on motivation to eat;

or permanent e.g. culture in a study on attitudes to di erent animals;

e.g. personality in a study on risk taking;

e.g. participant variable in Andrade could have been how much the individual normally
doodled;

these matter because di erent participants in di erent levels of the IV / used

independent measures design;

also matter in repeated measures designs, overcome by counterbalancing;

e.g. participant variable in Milgram was experience with electricity (made them disobey
earlier)

Situational variables – confounding variable due to the in uence to the environment;

e.g. whether you had been helped recently in a study on helping;

e.g. situational variable in Piliavin et al. was the behaviour / comments of passengers;

Situational variables are reduced through controls on the environment where possible
ff
ff
ff
fl
ff
1.8 SAMPLING OF
PARTICIPANTS
PAPER 2 RESEARCH METHODS
 The group, sharing one or more characteristics from which a sample is
drawn.

Is a group of people (or animals) with one or more characteristics in

common, it could be the population of a country or a group of people
A population who share a participant interest such as ‘all football fans’.

• Piliavin subway passengers /people on the trains / people on routes at

the time being used for the study;
• Milgram readers of the (New Haven / local) newspapers;
• Perry et al. undergraduates / people from Haifa university / from Israel.

Is the group of people selected to represent the population in a study.

A sample • Is the group of people who participate in a study that are taken from
the population and should ideally be representative of that group so
that the ndings of the study are generalisable.

Sampling The method used to obtain the (sample= participants for a study from a
technique population)
fi
 Generalisability How widely ndings apply, e.g. to other settings and populations

The more representative the sample the more generalisable the

results of a study.

the larger the sample size the more representative the sample, the
Representativeness more generalisable the results

the more diverse the sample the more representative the sample,
the more generalisable the results
fi
Opportunity (convenience) sampling
Description:

• participants are chosen because they are available, for example university students are selected because
they are present at the university where the research is taking place.

Strengths:

• Commonly used method because for many investigations

• Recruitment e orts are minimal

• Quicker and easier than other methods as the participants are available, therefore a larger sample can be
readily obtained.

Weaknesses:

• Likely to be non-represented as the variety of people available is likely to be limited, so they will tend to
be similar because they have been recruited from the same context (time and place), and the sample
could therefore be biassed.

• For example, many studies are conducted using university students as they are convenient for the
researchers. However this means that the sample will be predominantly young, often Westernised,
Educated, from Industrialised, Rich Democracies (WEIRD) and this may restrict the applicability of a study
worldwide.
ff
Volunteer (self select) sampling
Description:

• participants are invited to participate, for example through advertisements via email or notices. Those who reply
become the sample.

Strengths:

• Relatively easy because the participants come to the researcher.

• Participants are likely to be committed e.g. willing to return for repeated testing

• ethical because participants approach the researchers themselves (no pressure which might be present in opportunity
sampling) which is especially important for drug trails

• It is a useful technique when looking for participants who are unusual/unique in some way for example Baron-Cohen
et al's study where people with autism spectrum disorder were needed.

Weaknesses:

• Likely to be non-representative as people who respond to requests may be similar (volunteers) e.g have free time, are
motivated, interested to participate- bias.

• The nature of the invitation for volunteer sampling might exclude some groups of people.

• Imagine a researcher is looking for a sample of students at a school and puts an advert for volunteers on the library
notice board. Students who never go to the library cannot be included so the sample might be biassed towards
those who work the hardest.
Random sampling
Description:

• every member of the population has an equal chance of being selected

• sampling all members of the population (i.e. possible participants) are allocated numbers and a xed amount
of these are selected in an unbiased way, for example by taking numbers from a hat, or use a random number
generator to choose the participants.

Strengths:

• Likely to be representative as all types of people in the population are equally likely to be chosen.

• unbiased

Weaknesses:

• recruitment can be di cult as the researchers need access to names and details of all members in your
possible population

• sometimes random generation can also exclude certain categories of people e.g certain minority ethnicities
within the population

• In reality everyone may not be equally likely to be chosen e.g. if they cannot be accessed (if the original list is
incomplete) or if mainly one type of participant e.g. girls, happen to be selected. This is particularly important if
the sample is small.
ffi
fi
 Description Description Generalisability Practical concerns Why use it? Excludes

High ease: Minimal

Participants are Low: Participants are e ort and quick to When researchers
chosen because recruit participants. Participants who
likely to be similar need a quick and
they are available, were not readily
due to recruitment easy sample,
Opportunity e.g., university available e.g the
from the same Low di culty: especially for
sampling students selected people the
context, leading to Participants are preliminary or
from the university research does
bias and limiting readily available, exploratory
where the research not know
generalisability. making it practical for studies.
is taking place.
large samples.

Moderate: Attracts When researchers Participants who

Moderate ease:
motivated and require motivated were not
Participants are Recruitment is
engaged participants straightforward as participants or motivated to
invited to
but excludes certain participants come to need a speci c/ volunteer e.g
participate (e.g., via
groups, limiting the researcher. unique population introverts,
Volunteer ads or email).
generalisability. (e.g., clinical people who dont
sampling have the time; or
studies).
Those who respond Moderate di culty: who did not
Bias may occur due
become the Limited by the receive/see the
to self-selection (e.g., Lowers sample
sample. willingness of people invitation e.g the
volunteers with free to volunteer. attrition. ad
time or interest)
High: Best method Low ease: Time-
for consuming and
Every member of
representativeness. e ortful due to the
the population has Can also exclude
Ensures a diverse need for access to When researchers
an equal chance minorities, or
sample and minimizes the full population. need an unbiased,
of being selected, others if the
Random bias, leading to representative
e.g., by using a initial list of the
sampling results that generalize sample for
random number High di culty: May members of the
well to the population. generalisable
generator or require resources like population is
Some bias may occur results.
drawing names a population incomplete
if the sample is small
from a hat. database or random
or the population list
is incomplete. generators.
ff
ff
ffi
ffi
ffi
fi
 Advantage of
random sampling
over volunteer
sampling

Advantage of Random sampling is more likely to produce a representative sample of the target
random sampling population compared to opportunity sampling. This is because every member of the
over opportunity population has an equal chance of being selected, reducing sampling bias and increasing
sampling the generalizability of the results.

Opportunity sampling is typically quicker and easier to conduct than volunteer sampling because
participants are readily available and do not need to self-select or respond to an advertisement. This saves
Advantage of time and e ort for the researcher.
opportunity
Opportunity sampling can sometimes be more representative than volunteer sampling because it includes
sampling over participants who are available at the time, rather than only those who are motivated to volunteer. Volunteer
volunteer sampling sampling often attracts a speci c type of person (e.g., highly motivated or interested individuals), which can
lead to bias. Opportunity sampling, while still not fully representative, may capture a wider range of
individuals.
ff
fi
Possible questions:

describe each sampling technique

knowledge of strengths and weakness of each + in

comparison to each other

knowing which technique is going to work best

test of your knowledge of the concept of

representativeness and generalisability
larger size = more diversity = more representatives = more generalisability
1.9 DATA ANALYSIS
PAPER 2 RESEARCH METHODS
Data Analysis

Psychologists, like all scientists, often produce

numerical results from their investigations. These
results are called the ‘raw data’. As it is di cult to
interpret large amounts of gures, the results are
often simpli ed mathematically and represented
visually on graphs.
fi
fi
ffi
Summary table
a summary table may include totals from

a tally chart in an observation

percentages from a questionnaire

summary information about averages and the spread of data

A table should have

a title

rows (going across)

columns (going down)

there should be headings to indicate what the gures are about

fi
Table 3: A summary data table for a study about student time use and stress

Student group

A level Degree level

Average time spent

working per week 6 14
(hours)

Average time spent

socialising per week 4 20
(hours)

Average stress level 10 11

 Types of Data
When psychologists collect data they can collect either numerical results, called quantitative
data, or qualitative data, which is detailed and descriptive.

Quantitative data indicates the quantity of a psychological measure, such as the amount or
strength of a response and tends to be measured on scales, such as time, or as numerical
scores on tests such as IQ or personality. Quantitative data is associated with experiments
and correlations which use numerical scales but it is also possible to obtain quantitative data
from observations, questionnaires or interviews. For example, a record of the number of times
a behaviour is seen or the total of responses to a closed question in an interview would be
quantitative data. The sources of quantitative data are typically highly objective, as the scales
or questions used need little if any interpretation, making them high in validity. In addition, the
measures used are generally highly reliable, as the measures are xed quantities.

Qualitative data indicates the quality of a psychological characteristic. Such data is more in-
depth than quantitative data and includes detailed observer accounts and responses to open
questions in questionnaires, interviews or case studies. Although there is a risk of subjectivity
in the interpretation of such data by the researcher, qualitative data may be more
representative as the participant can express themselves fully, so in some senses qualitative
data can also be valid.
fi
 Quantitative Qualitative

Typically uses objective measures.

Data is often valid as participants can express
themselves exactly rather than being limited by
Scales and measures are often very reliable.
xed choices
Strength
Data can be analysed using measures of central
Important but unusual responses are likely to
tendency and spread making it easy to compare
be ignored because of averaging

The data is often relatively subjective so nings

may be invalid as data recording or
interpretation may be biased by the
Data collection method often limits responses
researchers opinions or feelings.
Weaknesses so the data are less valid, e.g. if the participant
wants to give a response that is not available
If detailed data can only be fathered from one
or few individuals it may not generalise to the
majority.
fi
fi
uses words

uses numbers
 Qualitative Data
• Qualitative data (QL) where words or descriptors are taken and converted to quantitative data (QT).
• Labels that are given numerical value

Nominal Data Ordinal Data

• Nominal values represent discrete units that have • Ordinal data is the same nominal data, except they
no order. Just think of them as “labels”. represent discrete and ordered units.
• No inherent order or ranking • Data that can be ordered and ranked

• Note that nominal data that has no order. • Note that for ordinal data order matters.
 Nominal Data: Ordinal Data:

• Gender: men or women • Age: younger, older

• Ethnicity: caucasian, asian, • Economic status: lower class,

hispanic middle class, upper class

• Eye colour: black or blue • Likert Scales: Likert scales are

commonly used in psychology to
• Diagnosis categories: has measure attitudes or opinions
schizophrenia or doesn't
• Measuring attitudes/agreement: e.g.,
strongly agree, agree, neutral,
disagree, strongly disagree.

• Measuring mood or emotional state:

e.g., very happy, happy, neutral,
unhappy and very unhappy
 Quantitative data

have numerical value

Discrete data Continuous data

Counted Measured

Cannot be divided into smaller parts Can be subdivided into smaller pieces

For example, eye colour or gender can be An example of a continuous variable is weight or height - a
considered a discrete variable because individuals person doesn't have to be either 150 pounds or 151
are either part of a category or they aren't - there is pounds. They could be 150.6 or 150.99999 pounds
no range of answers in between.

Bar chart Histogram

Examples: Examples:

1. Number of students in a class (cannot have 4. Amount of time required to complete a task (1.5
‘half’ a person) hours)

2. Number of names recalled (cannot have half a 5. Height of children (5 ft 6 inches)

name)
6. Duration of time spent studying (5.5 minutes)
3. Number of times a behaviour is repeated e.g.
acts of aggression 7. Scale of anxiety levels (0-5)
Self-reports

12

Frequency of
participants
Bar Chart 6

0
0 1 2 3

Rating on a scale of sadness (0-3)

Histogram
Observations

12

Frequency of
participants
Bar Chart 6

0
0 1 2 3

Number of complaints

Histogram
Biological measures
12

Frequency of
participants
Bar Chart 6

0
60-70 71-80 81-90 91-100

Pulse meter reading

Histogram
Thought box:
Scenario: A psychologist observes a volleyball game using a structured
observation method with prede ned behavioral categories. They collect
data by:

1. Counting the number of males and females on the team.

2. Counting the number of times a player disagrees with the referee.

3. Counting the number of times a player says a word of

encouragement to another player.

Question

What kind of data did you collect? Is it discrete or continuous? Provide a

reason.
fi
Solution

Answer: Discrete Data

Reason:

1. You are counting:

◦ The number of males/females, disagreements, and words of
encouragement are all numerical counts.

◦ These counts are represented as whole numbers (e.g., 3 disagreements,

2 words of encouragement).

◦ Discrete data consists of speci c, separate values, and in this case, you
cannot have fractional counts like 3.5 females or 2.7 disagreements.

2. Fixed values only:

◦ Each behavior or characteristic (e.g., disagreement or gender) can only
be recorded in whole numbers. There are no intermediate values in the
data collection process.
fi
Thought box
Scenario

A psychologist uses a pulse meter to measure

anxiety by recording pulse rates (measured in beats
per minute) for di erent participants. The pulse rate
range is between 40-140 beats per minute.

Question

Is the data discrete or continuous? Provide a

reason.
ff
Answer: Discrete Data

Reason:

Whole Numbers: Pulse rates are usually recorded as

whole numbers (e.g., 72, 73), with no decimal values.

Counting: Since you are counting the number of

beats per minute in xed values, this makes it
discrete data.
fi
Thought box
Scenario

A psychologist collects data on stress levels by

measuring cortisol concentration from saliva
samples. The cortisol levels are recorded in
micrograms per deciliter (µg/dL) and can have values
such as 12.3 µg/dL or 18.7 µg/dL.

Question

Is the data discrete or continuous? Provide a reason.

Answer

Continuous Data

Reason:

• Decimals Allowed: Cortisol levels are measured

on a scale that allows precise values, including
decimals.

• Measurement: The data represents a continuous

variable because it can take any value within the
range, re ecting a measurement rather than a
count.
fl
 Bar Chart Histogram
Discre /Ca gorical Data Con nuous Data

DV

IV DV
ti
te
te
 12

9
Number of people

0
CBT Psychoanalysis Art therapy Group therapy
 12

9
Number of people

0
0-10 11-20 21-10 31-40

Grade Intervals
In Experiments:
100

75

50

25

0
April May June July
Central tendency
A measure of central tendency (also referred to as
measures of centre or central location) is a summary measure
that attempts to describe a whole set of data with a single
value that represents the middle or centre of its distribution.

The 3 most common measures of central tendency are the

mean, median and mode.

• The mean is the sum of all values divided by the total

number of values.

• The median is the middle number in an ordered data

set.

• The mode is the most frequent value.

 Measures of central tendency

Mode Nominal Preferable for nominal data Bar charts

Typical and (gender, eye colour, race, etc)
value Discrete data
scale

Median Ordinal scale Preferable for when data ordinal Bar charts
(requires an order) (has an order) OR is not
symmetrical
Middle
value -is not impacted by outliers
-does not represent outliers

Mean Discrete data Preferable for when data is Bar charts

discrete and symmetrical
Average Cannot use for
value qualitative data -includes/represents outliers
(nominal or ordinal) -can be skewed by outliers
 (1) Mean: the average
It is calculated by adding up all the scores and dividing by the number of
scores in the data set. Mean is generally considered the best measure of
central tendency and the most frequently used one. However, there are
some situations where the other measures of central tendency are preferred.

Advantages

• The mean provides representative results as it takes into account

outliers as it takes into account all values in a data set.

Disadvantages

• The important disadvantage of mean is that it is sensitive to extreme

values/outliers, especially when the sample size is small. Therefore,
it is not an appropriate measure of central tendency for skewed
distribution.

• Mean cannot be calculated for nominal or ordinal data

(2) Median: the middle score of a data set
Advantages
• It is not distorted by outliers/skewed data (small number of outlying
scores). This is a bene t because it is less swayed by a small number of
extreme scores; therefore, it is usually the preferred measure of central tendency
when the distribution is not symmetrical.
• The median is more informative than the mode because it considers the rank
order of the values of the scores in a data set rather than just how many they
are.
• It can be determined for ratio, interval, and ordinal scale.
• It is easy to compute and comprehend.

Disadvantages
• As it does not take into account outlying values it is less representative
than the mean.
• Unlike the mode, the median cannot be identi ed for categorical nominal data,
it is only used with numerical data on a linear scale (i.e. points in a sequence).
• Unlike mean, median is not amenable to further mathematical calculation and
hence is not used in many statistical tests.
fi
fi
(2) Median: the middle score of a data set

Advantages

• It is the only measure of central tendency that can be

used for data measured in a nominal scale (discrete
categories)

• It can be calculated easily

Disadvantages

• Although the mode considers all the scores in the

data set, it doesn't consider their values, so it is less
informative than the median or mode.
 Dataset Example: Likert Scale Ratings for "Satisfaction with Therapy"

Participants rated their satisfaction with therapy on a scale of 1-5, where:

1 = Very Unsatis ed, 2 = Unsatis ed, 3 = Neutral, 4 = Satis ed, 5 = Very Satis ed.

Participant Rating

1 5

2 4

3 3

4 4

5 2

6 3

7 5

8 1

9 4

10 3
fi
fi
fi
fi
Dataset Example: Likert Scale Ratings with Two Outliers

Participants rated their satisfaction with therapy on a 1–5 scale:

1 = Very Unsatis ed, 2 = Unsatis ed, 3 = Neutral, 4 = Satis ed, 5 = Very Satis ed.

Participant Rating

1 5

2 4

3 3

4 4

5 2

6 3

7 5

8 1

9 4

10 1

11 (Outlier) 20

12 (Outlier) 25
fi
fi
fi
fi
Median is preferred to mean when:

• There are few extreme scores in the distribution.

• Data are measured in an ordinal (has order) scale.
• the median is often preferred because it is robust to outliers and
provides a better representation of the central tendency, especially
for ordinal data.

• Mode is the preferred measure when data are measured in a

nominal scale (qualitatively).

• As it does not take into account outlying values it is less

representative than the mean
 Measures of spread

Range The di erence between

the maximum value and
the minimum value
(plus 1)

Standard It is a Measures how close The smaller the SD

Deviation measure of the values of a data set the closer the value
variability are to the mean are to the mean i.e.
less spread out

The greater the SD

Are the clustered near the more dispersed
the mean OR dispersed the values are from
widely from the mean the mean i.e. more
spread out.
ff
Graphical representation

Characteristic Bar Chart Histogram

Data Type Categorical or Nominal Data Continuous Data

X-Axis Categories or distinct groups Range of values (intervals)

Y-Axis Count, percentages, averages Frequency (count or density)

Bars touch or have minimal

Gaps Between Signi cant gaps between bars
gaps
Bars (no connection)
(continuous distribution)
fi
1.10 ETHICAL
CONSIDERATIONS
PAPER 2 RESEARCH METHODS
Ethical considerations
privacy
-pieces of advice that guide right to withdraw - researchers should not enter a
psychologists to consider the welfare of physical space or emotional territory
participants and wider society - participants should be able to leave a that the participant would like to keep
study whenever they wish to themselves
- although incentives can be given -
- self-reports: participants should be
researchers cannot use their position of aware of their right to not answer
authority to encourage the participants to questions
valid consent remain
- - observations: people should only
be watched in situations they would
- - participants must be given su cient
information about the procedure so that expect to be on public display
their decision is informed (without lack of deception
leading to demand characteristics)
- - consent must be freely given - participants should not be deliberately
- - by a competent individual who misinformed i.e. deception should be debrie ng
understands what is being asked (might avoided
exclude people with mental health - when deception is essential to avoid the
e ects of demand characteristics
- giving a full explanation to
problems, learning di culties and participants at the end of this study
children) - they should be told the real aim as soon so they leave in at least as positive
- - presumptive consent can be taken as possible and be allowed to remove their a condition as they arrived
where informed consent cannot be results if they want
- -all participants who are aware of
being studied should be thanked
and given the chance to ask
questions
con dentiality - they should be explained the aim of
protection from harm
the study to ensure they do not want
- all data about participants should be to withdraw their data
-participants have the right to be protected
stored securely and not published without - if participants were impacted
from psychological (stress,
consent negatively the researcher must ensure
embarrassment, self doubt) and physical
- their identity should be protected by for they are retuned to their previous
(risky behaviours, injections) harm
example anonymising individuals or condition
- they should not be exposed to any risk
greater than they would be in their normal allocating numbers to participants
life
ff
fi
fi
ffi
ffi
Ethical guidelines relates to the use of animals
Pain, su ering and distress
(2) species -research causing death, su ering or
Animals are used in psychological
research because disease, pain, injury, psychological or
- the chosen species should be the one physiological distress or discomfort
1. they are convenient models least likely to su er pain or distress
(processes of learning) should be avoided
- -other factors: sentience, wild or captive -designs that improve rather than
2. o er a way to carry out procedures animals, previous experience of worsen the animals experience should
that cannot be done ethically on experimentation be used
humans (isolation or brain surgery) - attention should be paid to the
3. as interesting examples in their own animals daily care and veterinary
(3) number of animals
right (whale communication) needs and any costs to the animals
should be justi ed by the scienti c
-only the minimum number of animals
bene t of the work
Bateson’s cube needed to produce valid and reliable results
should be used.
- -Researchers aim to ensure that the -to minimise the number, pilot studies,
means justify the ends i.e. the animal reliable measures of the DV, good housing
su ering caused by planned experiment experimental design and appropriate data
is outweighed by the bene ts analysis should all be used -isolation and crowding can cause
- - the certainty of bene t is high, the animals distress
research is good and su ering is low -caging conditions should depend on
the research can be argued to be (4) Procedure the social behaviour of the species
worthwhile. -overcrowding can cause distress and
- research is controlled by legal aggression (harm)
requirements or guidelines from relevant -housing with enough space to move
-Guidelines from the British organisations freely and with su cient food and
Psychological Society Guidelines for - -for research on animals to be e ective water for their biological and
Psychologists Working with Animals the animals experience should within ecological needs
constraints be a normal and positive one -arti cial environment should recreate
the natural environment important for
(1) replacement welfare like warmth
-researchers should consider replacing reward, deprivation and averse stimuli -cage cleaning should be balanced
animal experiments with alternatives, such against stress
as videos from previous studies or - - alternatives to deprivation and averse
computer simulations stimuli should be used where possible
ff
ff
fi
fi
ff
fi
ff
ffi
fi
ff
fi
ff
ff
fi
 Animal Ethics
• Some experiments may have a medical bene t to animals, which gives them an ethical advantage.
• For example, Fagen et als ndings assisted in treating tuberculosis in elephants, which would be a highly bene cial
nding. Moreover, one of the goals of Fagen et al. was to provide evidence for Elephant trainers to avoid using
punishment (negative reinforcement) and instead focus on reward (positive reinforcement).
• Less punishment for the Elephants would reduce su ering, and improve their condition.
justi cation • Researchers must aim to ensure that in any research, the means justify the ends, i.e. that the animal su ering caused by
the planned experiment is outweighed by the bene ts. One way to consider this question is to use Bateson’s (1986)
cube. When the certainty of bene t (e.g. to humans) is high, the research is good and the su ering is low, the research is
worthwhile.
• Overall, Fagen et al. provided a High certainty of Medical Bene t with a High Quality of Research and Low Levels of
Animal Su ering

• Animals should be protected from pain, e.g. relating to surgery using appropriate anaesthesia and analgesia, and
killed (euthanised) if su ering lasting pain.
harm
• This does not relate to the Fagen et al. study as the elephants were not at risk of catching a life threatening
disease in fact, the study was designed to prevent elephants from developing tuberculosis.

• In the Fagen et al study chopped bananas were used as part of the SPR training which is based on the principles of
positive reinforcement aka rewards and purposely sought to prove the e cacy of SPR over punishments therefore
Reward,
promoting and maintaining the welfare of the elephants.
deprivation
and • The one aversive stimuli used was the syringe and saline, but it was justi ed as it was necessary for the diagnosis of
aversive tuberculosis and therefore in the long run bene cial. Furthermore the elephants were gradually desensitised to the
stimuli
syringe and saline through gradual exposure, starting with a drop and building up in small increments (ranging from
1-15ml) to reach a tolerance of 60ml of saline for sample collection.
fi
fi
ff
ff
fi
fi
fi
fi
fi
ff
fi
fi
ffi
ff
ff
fi
 • Only the minimum number of animals needed to produce valid and reliable results should be used.
To minimise the number, pilot studies, reliable measures of the dependent variable, good
number of experimental design and appropriate data analysis should all be used.
animals
• Only a minimum number of animals should be used to produce valid results. In Fagen et al. only
ve animals were used, four baby elephants and one older elephant. The data might not be
generalisable but it makes the study ethically sound as least amount of animals faced the suffering.

• Fagen et al.'s research is based on elephants, a species highly sentient and socially complex,
which could experience signi cant distress in captivity.
species and • However, their study aims to use positive reinforcement (without the use of punishment) to teach
strain elephants to perform a trunk wash procedure that could medically offer great bene t as it would
allow their caretakers to timely diagnose and therefore treat the elephants with tuberculosis
therefore minimising their suffering and even death.

• The elephants’ diet consisted mainly of fresh grasses and dhana (packets of grain, nutritional
supplements, and grasses).
• The elephants had access to the river for water during their grazing time but were otherwise not
food offered water outside of the training protocol.

• These husbandry conditions are the standard practice at the stable and no alterations were made
for the purposes of the study.

• Researchers should consider replacing animal experiments with alternatives, such as videos from
previous studies or computer simulations.
replacement
• The elephants could not have been replaced as it was their illness being treated through the
experiment.
fi
fi
fi
 • Research causing death, disease, injury, physiological or psychological distress or discomfort should be
avoided. Where possible, designs which improve rather than worsen the animals’ experience should be used to
normal rather than early deprivation). Alternatively, naturally occurring instances may be used (e.g. where stress
arises naturally in the animal’s environment or lifetime).
• Given the limited freedom permitted by their chains, the elephants could clearly indicate a preference to not
participate in training sessions by walking to the other side of their stalls or simply turning away from the
trainer.
• Mahouts were present at all sessions and stood on the periphery for the safety of the trainer, but they were

procedures: clearly instructed not to speak to or signal the elephants in any way during the sessions to maintain the integrity
pain and of the training. The mahouts complied with this request. Their presence also ensured safety due to the fact that
distress elephant were familiar with the mahouts
• During research, attention should be paid to the animals’ daily care and veterinary needs and any costs to the
animals should be justi ed by the scienti c bene t of the work
• Saline was slowly introduced to the elephants through the process of desensitisation to reduce the distress
caused by saline washing; over a series of repetitions the syringe was brought closer to the elephant, made
contact with its trunk and then with little increments of ml the saline was given to the elephants. This process

• Between testing , animals should be housed with enough space to move freely.
• Elephants went into the jungle to graze under the control of their mahouts from 5 a.m. to 7 a.m. and 10:30 a.m.
to 4 p.m. each day, and they were leg-chained to posts in open stalls for the remainder of the day and night.
• Leg chains were normally placed with both front legs chained together or on a single front leg, with a chain
approximately 6 ft to 8 ft long (1.8 m to 2.4 m) between them and the post. This setup allowed enough laxity in
housing the chains for the elephant to shu e in a diameter that was 6 ft to 8 ft around her.
• The arti cial environment needed to recreate aspects of the natural environment important for the welfare and
survival.
• The elephants spent most of the day grazing in the jungle miming their natural environment.
• They spent the rest of the day leg-chained in a stable
fi
fi
ffl
fi
fi
Why is it important to get consent
from children and parents/
guardians?
Consent is agreeing to participate

Parents

Informed consent requires that they have enough

information to decide for their child;

Children are not fully able to give informed consent /

children may not understand enough;

Children

Child consent necessary as they have feelings too

and should be asked if they are comfortable with
being part of the study
1.11 EVALUATING RESEARCH
PAPER 2 RESEARCH METHODS
 • can it be generalised to other cultures?

G
GENERALISABILITY
- can the ndings of the
• how large was the sample size?
• how representative (age, gender, ethnicity) was the
study be generalised to
sample?
the general population? • how realistic was the setting/measure?

RELIABILITY • how standardised was the study?

R
- consistency • are there any extraneous or confounding variables?
- if the study was • is it replicable?
repeated how likely are • how many controls were there?
you to get the same • how objective were the measures?
results?
• does it provide evidence for a theory?

A APPLICATION • does it help people resolve a problem?

- how can a study be • can the information be used to explain a real life
phenomena?
used to support theories?
•

VALIDITY

V
accuracy • how likely is it that the IV caused the change in the
DV (internal validity) - depends on controls
- did the study accurately
• to what extent can the results be applied to real life
measure what it set out to situations (ecological validity)? - depends on
measure? mundane realism

did participants give informed consent?

E
ETHICS
were participants deceived during the study?
- did the researchers
could participants be at risk of psychological harm?
follow the ethical
were participants given the right to withdraw and
guidelines?
debriefed after the experiment
fi
 Objectivity: the impact of an unbiased external viewpoint on, for
example, how data is interpreted.
• A strength of quantitative (QT) data gathered from close ended
questions and structured observations that are based on
quanti cation.

Subjectivity: the effect of an individual's personal viewpoint on for

example how they interpret data.
• An issue with qualitative data (QL) gathered from open ended
questions and unstructured observations, that require
interpretation.
fi
 Generalisability
Kinds of generalisability Impacted by

Ecological validity Mundane realism

- whether results can be - how realistic was the
generalised to the real world experimental setting?

Representativeness
Population validity
External validity - whether results of one sample - how large was the
i.e. generalisability of sample?
can be generalised to other
ndings - how representative was
populations the sample?

Can be tested with

Temporal validity
longitudinal studies or
- whether results of one time
through replication. Is
can be generalised to other
impacted by many
times factors.
fi
Reliability - in experiments

• tools like pulse meters, MRI scans are

objective and therefore reliable
measures • self reports: close ended questionnaire
+ structured interviews + structured
- how objective were observations provide
the tools used to • objective and quantitive data and
measure the DV or IV therefore improve the reliability of a
study

Reliability
consistency

procedure

- how standardised was

the procedure?
• if within each level of the IV the
procedure is consistent the study will
have reliability
lower in eld lower in the eld
experiments because it is
because harder to control
standardisation extraneous
is di cult variables
ffi
fi
fi
 Validity
the extent to which the researcher
can be sure that the changes
internal validity measured in the DV were caused
by the manipulation of the IV.
addresses whether causality
was established
Improves with control of
extraneous variables (participant +
situational), control groups,
demand characteristics, order
e ects, experimenter bias, and
Validity social desirability
accuracy

ecological validity
the extent to which the results of
the study can be applied to real
addresses whether the
world contexts
results are true/applicable to
the real world

improves of the study has high

mundane realism i.e. is able to
measure participants natural and
normal responses
ff
 Validity - in experiments

internal validity
the extent to which the researcher can be sure that the
changes measured in the DV were caused by the
addresses whether causality
manipulation of the IV.
was established

high in lab lower in eld

a ected with control of extraneous variables
(participant + situational), control groups, demand
characteristics, order e ects, experimenter bias, and
because of because of low
high controls controls social desirability
Validity
accuracy

ecological validity
the extent to which the results of the study can be
addresses whether the applied to real world contexts
results are true/applicable to

lower in lab higher in eld

improves of the study has high mundane realism i.e.

because low is able to measure participants natural and normal
because of
mundane responses
high mundane
realism
ff
fi
fi
ff