Research

mETHODOLOGY
Chapter 5
THE CIRCLE OF SCIENCE
 Evolution in Experimental Psychology
Ferdinard Bernhard Ueberwasser - 1787 - 'Instructions'

WilliamWundt - 1832 - 1920 - Chronoscope - Introspection - 1st Lab

JohnWatson - 1875 - 1958 - Behavioural Psychology

Cognitive Psychology | Psychoanalysis

Terminologies
Variables - independent | dependent | Control
Demand Characteristics
Ecological validity
Informed Consent
Deception | Debriefing
Right to withdraw
Validity
Reliability
TYPES OF RESEARCH METHODS
Experiments
Self-reports: Questionnaires & Interviews
Case Studies
Observations
Correlations
Longitudinal Studies
 experiment
Important Terms Main Features
Independent Variable Compare the data
Dependent Variable Variables - Manipulated | Measured
Control/Extraneous Variable Cause & Effect
Operationalisation of Variables Standardisation - all participants have
Standarised Procedure & instructions same experience - Replicate - Reliability
 TYPES OF EXPERIMENT

LABORATORY FIELD
 Laboratory Field

Place Laboratory Real-world Settings

Controls Controlled Variable Poor control

HighControls Less Controls

IV impact on DV IV impact on DV
Validity
Poor ecological validity High ecological validity
Demand characteristics Poor demand characteristics

Reliability Standarised - Replicate Poor replicate

Informed Consent - Debrief Informed Consent - x

Ethics
Right to Withdraw Right to withdraw - x
 Experimental Designs

Independent measures Repeated measures Matched pairs

Different participants
Different participants in Same participants in matched across different
Meaning
different groups different groups variables in different
groups

Participant Variables -
Participant variables - X Participant Variables - X
Participant & Demand similar
Demand characteristics - Demand Characteristics -
Characteristics Demand characteristics -
High Low
Low

Yes
Order | Fatigue | Practice Counterbalancing - order in
None None
Counterbalancing which participants does
both the conditions
 Experimental Designs
Special type of design: Randomised Control Trial (RCT)

Gold standard technique - Clinical & health research

The Placebo Effect - Expectancy effect
Single-blind | Double-blind | Triple-blind

Evaluating RCT's
Validity: Experimenter bias | Demand characteristics
Reliability: Inter-rater reliability
Ethics: Informed Consent | Psychological harm
 IMPORTANT TERMS
Independent Variable - a factor/ variable under investigation in an experiment that
is manipulated to create 2 or more conditions/ levels. OR a variable that is expected
to be responsible for changes in the dependent variable

Dependent Variable - a factor/variable in an experiment that is measured and is

expected to change under the influence of the independent variable

Experimental condition - one or more of the situations in an experiment that

represent different levels of the IV and are compared

Control condition - a level of the IV in an experiment from which the IV itself is

absent and is compared to one or more experimental conditions
 IMPORTANT TERMS
Experimental design - the way in which participants are allocated to levels of the
IV

Independent measures design - an experimental design in which a different group

of participants is used for each level of IV

Repeated measures design - an experimental design in which each participant

performs in every level of the IV

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 1 member
of each pair performs in a different level of the IV
 IMPORTANT TERMS
Confounding variable - an uncontrolled variable that acts systematically on one
level of the IV so could hide or exaggerate differences between levels and
therefore ‘confound’ or confuse the results making it difficult to understand the
effect of the IV on the DV

Demand characteristics - features of experimental situation which give away the

aims
ACTIVITY
A researcher wants to investigate how the level of stress affects a person’s ability to
solve puzzles. One group of participants is asked to complete puzzles while
experiencing a high-stress situation (e.g., a time limit and loud noises), while
another group solves puzzles in a calm, quiet environment without a time constraint.

Independent Variable: ______________________

Dependent Variable: ______________________

Control Variables: _______________________

A psychologist examines the impact of positive reinforcement on children’s
behaviour. One group of children is given praise every time they clean their room,
while another group is given no praise for the same behaviour. The researcher
measures the frequency of cleaning behaviour over a week.

Independent Variable: ______________________

Dependent Variable: ______________________

Control Variables: _______________________

A researcher wants to study how watching violent TV shows affects aggression in
children.

Independent Variable: ______________________

Dependent Variable: ______________________

Control Variables: _______________________

A psychologist is studying how the presence of pets affects stress levels in adults.

Independent Variable: ______________________

Dependent Variable: ______________________

Control Variables: _______________________

A researcher is investigating whether mood affects memory recall.

Independent Variable: ______________________

Dependent Variable: ______________________

Control Variables: _______________________

A researcher is examining whether social media use influences self-esteem.

Independent Variable: ______________________

Dependent Variable: ______________________

Control Variables: _______________________

Let's test ourself
 Self-reports - Questionnaire
Self -report: Method of research - responds to either open or closed
questions
Questionnaire or Interviews
Quantitative or Qualitative Data

Questionnaire:

Printed - Paper pencil or digital

Closed or open questions
 Closed & Open Questions
Closed -

1. Do you find mindfulness meditation relaxing?

Yes / No

2. Mindfulness meditation helps you to relax.

0 = Never 1 = Sometimes 2 = Most of the times 3 = Always

Open -

1. How mindfulness meditation is relaxing for you?

2. Does mindfulness meditation helps you to relax? Yes/No

If yes, answer the following questions, else continue from question 5
 Closed Open

Qualitative | Rich &

Data Quantitative
detailed - deeper insight

Advantages
Measures of central
tendency | Graphs & --
charts
Analysis
---------------------
Hard - huge data | Time
consuming | Subjective

Limited choice - Poor Subjective - Poor

Disadvantages Reliability & Validity
validity reliability

Forced choice |
Answers Cannot express --
opinion
 What to avoid when creating questionnaires

Double barrelled questions

Ambiguous questions - Pilot studies
Technical terms
Emotive Language
 Evaluating Questionnaires
Advantages Disadvantages

Relatively quick & easy Less willing to give elaborate

answers
Large number of participants -
representative, depending upon Limiting the depth
sampling techniques

Less likely to give socially Might give socially desirable

desirable answers compared to
interviews
postal questionnaires
Refers to questionnaires that are posted out to the sample

Advantages Disadvantages

Relatively quick & easy Low return rate - effort of posting back

Large number of participants - Return - unrepresentative of target

representative, depending upon population - especially helpful,
sampling techniques motivated, particular interest in topic

Can be posted anywhere in the world expensive - post large numbers,

especially return postage is included
 Self-reports - Interviews
Researcher speaks directly to the participants - Ask questions - Record
responses

- Face to face or telephonic

- Closed or open ended

- Structured | Unstructured | Semi-structured

 Structured Unstructured Semi-Structured

Pre-determined Main topics pre-determined Few questions pre-determined

Fixed list of questions Does not follow certain list of Any order
questions
Same order Add new questions
 Evaluating Interviews
Advantages Disadvantages

In-depth data More likely to give socially desirable

answers - fear of being judged

Semi-structured or unstructured - good Risk of researcher bias

insight - discover thoughts & ideas - never
considered

Face-to-face - Non-verbal communication | Face-to-face - impractical - remote areas

Body Language Uncomfortable - sensitive topics

Telephone interview - Quicker | Cost Telephone - not comfortable - less details

effective - Saves resources - larger sample Researcher cannot use non-verbal signals
 Correlations
Correlation: Relationship between two variables i.e. co-variables
A study exploring the link between academic performance and
career success.
A study examining the correlation between income levels and
crime rates.

Operationalise co-variables
Prior to experiment or hard to manipulate variables
Data - quantitative

Analyse data - Scatter plot graph

- Positive | negative | no correlation
- Strength of relationship
 Correlations
Individuals with more social support tend to experience lower levels of
depression

Individuals with higher levels of education tend to have higher incomes

There is no meaningful relationship between a person's hair color and

their IQ

As sleep increases, fatigue tends to decrease

Higher stress levels tend to be associated with higher levels of anxiety

 Evaluating Correlations
Advantages Disadvantages

Evaluate new areas prior to Shows strength of relationship,

experiment not cause and effect

Investigate where manipulation Does not indicate why

is not possible or ethical relationship has occurred

Presence of third variable cannot

be ruled out. It might have
influenced
 Observations
Observation: Watching people's behaviour & recording it

Features:

Role of observer - Overt or Covert

- Participant or non-participant

Nature of observation - Structured or unstructured

Place of observation - Natural or Controlled
 Evaluating Observations
Advantages Disadvantages

Structured observations - more reliable Observer bias - subjective

than unstructured Unstructured - difficult to observe all the
Unstructured - Pilot study - limit the behaviours - subtler ones - missed
observations Overt - Demand characteristics
Overt - Ethical - no deception Covert - Less ethical - deception
Covert - less demand characteristics Participant observer - get too involved, lose
Participant observer - observe behaviour objectivity
closely Non-participant - Watching from distance -
Non-participant - More objective some behaviour would be missed
Naturalistic - Own settings - high Naturalistic - Hard to control variables
Ecological validity Controlled - Behaviour less natural - miss key
Controlled - Easily replicated - reliable behaviour - rigid schedule
 Case Studies
Case-study: Detailed investigations of an individual or small group

Useful for rare or unusual cases

Key feature - Triangulation - information using multiple techniques

Data - multiple sources - extended time - could be longitudinal

Might have qualitative & quantitative data

 Evaluating Case-Studies
Advantages Disadvantages

Rich & detailed information Less objectivity when analyzing data - poor
validity
Unique insight into participants
Hard to replicate
Triangulate the data - high validity
Specific experience - hard to generalise
 Longitudinal Studies
Longitudinal Studies: Same group of individual over an extended time

Features -
Various tasks/tests are given
Participants - re-contacted over time for repeated testing
Testing needs to be standardized - valid & reliable
Varied techniques used - interview, questionnaire
Experimental & control group compared over a period of time
 Evaluating Longitudinal Studies
Advantages Disadvantages

Advantage over cross-sectional Long-term - Attrition, i.e. drop-out

studies

Allows some topics to be studied Difficult to generalise

which otherwise would not be
possible
E.g. Child development

Experiment + Longitudinal Studies -

Avoid participant effects
 Psychometric tests
Psychometric means 'measuring the mind'

Designed to provide valid and reliable measures of individual differences -

aptitude, personality, skills

Tests are standardized on random sample - norms - compare

Normal Distribution Curve

Floor & Ceiling effect

Normal Distribution Curve

Ceiling Effect Floor Effect

 Test Yourself
Sanija is conducting a study to investigate whether there is a difference in
students' attention spans in school depending on whether they eat hot or cold
breakfast.

Suggest an appropriate research methods that Sanija could use to test her
theory.

Suggest how Sanija could operationalize her dependent variables.

 Test Yourself
Andrade wanted to understand if doodling during the task helps them to focus
better by reducing day dreaming. He has two groups for his study and different
participants for both. The procedure is standardized for both except for the
manipulation of the independent variable. He then understands the experience of
each participant post the task.

Identify the research method used.

If he wants to understand the long term effect of same, then which research
method could be used.

Few controls that you would suggest.

 Methodological Concepts
Aims, hypotheses & variables
Controlling of variables
Types of data
Sampling of participants
Ethics & human participants
Ethics & non-human animals
Validity
Reliability & replicability
Data Analysis
 Aims, Hypotheses, Variables

Aim: Statement that explains why the researcher is doing research & what they
hope to achieve

Usually research paper starts with the statement mentioning the aim

The aim would be then simplified/operationalised for the study

Hypotheses: Statement where researchers predict what they think will happen in
the research
Hypothesis is operationalized - variables are defined

Types:
Directional | One-tailed
Non-directional | Two-tailed
Null
 Directional / one-tailed hypothesis

Directional/one-tailed - Researcher predict whether exp. group will perform significantly better
or worse than the control group
Direction of impact of IV on DV is stated - research previously done
E.g.: ---------will improve/worsen--------- or -----better than---------------

1. Individual who have 8 hours of sleep at night have better attention span as compared to/than
individuals who have less than 8 hours of sleep.
There is a positive correlation between number of hours of sleep and attention spans of an
individual.

2. Frame an one-tailed hypothesis for the following -

IV - Attending lectures
DV - Exam results
 Non-directional / two-tailed hypothesis
Non-directional/two-tailed - Researcher states that there will be significant difference but
direction of impact of IV on DV is not stated - new areas of research
E.g.: There will be significant difference/correlation

1.There is a significant difference between number of hours of sleep and attention spans of an
individual.
There is a correlation between number of hours of sleep and attention spans of an individual.

2. Frame a two-tailed hypothesis for the following -

IV - Attending lectures
DV - Exam results
 Null Hypothesis
Null - No impact of IV on DV
E.g.: There will be no difference/correlation-------- or any difference that arise will be due to
chance.

1.There is a no difference between number of hours of sleep and attention spans of an individual.
There is no correlation between number of hours of sleep and attention spans of an individual.

2. Frame a null hypothesis for the following -

IV - Attending lectures
DV - Exam results
 Controlling of Variables
Control all variables except the one which is manipulated
Kept constant across conditions
Changes in DV is caused by IV

Replicate - reliability - standardized procedure - all information same - except information about IV
when required

Participant Variables -
Random allocation of participants
Matched pair design

Situational Variables -
Aspects of environment that could impact participants' behaviour and affect the results
Control as much as possible

Uncontrolled Variables -
Factors not measured or controlled and have unwanted impact on DV
E.g. Noise, Light, Pigeon
 Types of Data

Quantitative Qualitative

Meaning Numerical Data Non-numerical data

E.g. Closed-ended questions Open-ended questions

In-depth understanding of
Easily compared | Objective | Less
Strengths participants behaviour, feelings |
prone to researcher bias
may give greater validity

Does not allow full understanding of

participant | Limit how participant Risk of researcher bias |
Weakness
could respond | less valid - none of the Subjective | Less reliable
response could reflect their feeling
 Types of Data

Subjective Objective

Data that can be influenced by

Un-biased, factual
Meaning person's personal thoughts, feelings
Not influenced
or opinions

More of an issue with qualitative data Quantitative Data

Subject to interpretation by the Objectifiable | does not require

researcher interpretation by the researcher
 Sampling of Participants
Sampling - How researchers obtain participants for their research

Target Population - Sample

Sample is the group of people who participate in a study. They are taken from a population and
should ideally be representative of that group so that the findings will be representative
Population - the group, sharing one or more characteristics, from which a sample is drawn.
sample - the group of people selected to represent the population in a study.

Sampling Techniques
Opportunity Sampling

Random Sampling

Volunteer (self-selecting) Sampling

 Opportunity Random Volunteer

Researchers select Every person has an equal Participants put themselves

Meaning participants who are readily chance of being chosen to forward to take part in
available participate research

Quick & easy | Motivated & willing - drop

Would be representative |
May result in larger sample | out rate is low | important in
Strength More generalizable if
Study conducted without longitudinal study - attrition
sample is large
delays - validity

Time-consuming &
May not be representative |
expensive |
Unrepresentative | Only with pro-social
Weakness Potential participant may
May not be generalisable attitudes |
not want to participate |
Personal Interest
Make arrangements
 Ethics & human participants
Serve to protect the participants and ensure they do not come to any harm
Ensure research benefits wider society and can be pursued

Ethical guidelines in psychological research -

Valid Consent
Right to withdraw
Minimising harm and maximising benefit
Lack of deception
Confidentiality
Privacy
Debriefing
 Ethics & non-human animals
Serve to protect the participants and ensure they do not come to any harm
Ensure research benefits wider society and can be pursued

Ethical guidelines in psychological research -

Minimise harm and maximise benefit
Replacement
Species
Numbers
Procedures: pain & distress
Housing
Reward, deprivation and aversive stimuli
 Validity
Refers to the extent to which the researcher is measuring what they think they are measuring and
the extent to which the findings are useful and meaningful

E.g. Army Alpha and Beta IQ Test - testing cultural knowledge rather than IQ

Ecological Validity

Extent to which behaviour that participants demonstrate during research relates to how
they would behave in their real lives

Research lacking in ecological validity - cannot be generalized

 Evaluating studies - Validity
1. Subjectivity/objectivity -
Researcher to required to interpret behaviour - risk for subjectivity
Qualitative data - higher risk of of subjectivity

Data not influenced by researcher's personal thoughts, feelings - objective

Quantitative data - chances of higher validity

2. Demand Characteristics - Any features of research environment which may give away
the study's aim - Deception
Animals - less demand characteristics

3. Generalisability - Generalised to target population if the sample was representative

Population validity

4. Temporal Validity - Refers to whether the results of a study can be generalised to a

different time period
 Reliability & replicability
Reliability - Consistency
Results from each participants in the study | Results from same participants on re-test | Results
from different participants | Same study repeated

Same results in the above scenarios - findings are reliable

Replicability - Psychologists can repeat the study exactly to see if results are consistent -
required to show findings are not due to chance - relies heavily on standardised procedure

Standardisation - Procedure and instructions

All aspects are consistent

Ways of increasing reliability -

Test-retest reliability - same test repeated, scores are consistent
Inter-rater reliability - researchers agree to their scoring of questionnaire/test
Inter-observer reliability - researchers agree in their rating/coding of behaviour during obs.
 Data Analysis
Measures of Central Tendency
Descriptive data
Summarise all data into one score - avearge

Three measures of central tendency - Mean, Median & Mode

Mean Median Mode

Meaning Average Middle Point Most frequent

Arrange data in rank order -

smallest to greatest -
Add all scores and divide middle one Most frequent occurring
How to calculate
by number of scores Even number of data - add data point or score
two middle scores and
divide by two
 Measures of Spread

The Range Standard Deviation (SD)

Give precise information about the

How widely the data is spread spread around mean
between lowest and highest Smaller SD - data points are
Meaning
Larger the range, wider the data is clustered around mean
spread Larger SD - Data points are more
diverse

Mean is subtracted from each data

point to see how far each scores
Subtract the lowest value from the deviates
How to calculate
highest value
Average amount of deviation from
mean
 Graphical Representation of data

Bar Chart Histogram Scatter Graph

Quick representation of Quick representation of all

Quick representation of
difference between scores of data collected - distributed
correlational data
two conditions within each interval

Use to show relationship

Use to show categorical data Use to show continuous data
between

Bars do not touch each other Bars touch each other Line of best fit
Thank
You