Research Methods

AS-level
Table of contents

01 Aims
03 Variables

02 Hypothesis
04 Correlations

05 Experiments
 Time to refresh

● What do you remember

about research methods
from other subjects?
○ Think about chemistry,
physics, biology
 Think about..
How do these
methodological concepts
relate to psychology?
 Experiments

● It’s a research that allows

us to explore the cause and
effect of something
 01
‘Aims’
 Aim

● First step in every study

● The aim is the purpose of
the study
● What am I trying to find
out?
Example 1

What is a
potential
aim?
Example 2

What is a
potential
aim?
02
Hypothesis
A hypothesis (plural hypotheses) is a
precise, testable statement of what
the researcher(s) predict will be the
outcome of the study. It is stated at
the start of the study
 Hypothesis

● It is..
○ An assumption made by
the researcher
○ It is testable
Example 1

What is a
potential
hypothesis?
Example 2

What is a
potential
hypothesis?
 Three types of hypothesis

● Non-directional
● Directional
● Null hypothesis
 So far..

● In a study, there is an aim

and hypothesis
● Aim  what we are trying to
find out or the “purpose”
● Hypothesis  what we predict
03 Variables
Operationalize
 Operationalization – why it matters

● Turning abstract concepts into

measurable observations
● Without operationalizing the
abstract, research is not
possible
● Measurement is key!
 Operationalization – why it matters

● It’s about turning something

subjective to objective
● We want to measure relevant
concepts and apply consistent
methods
 Operationalization

● Everything can be
operationalized
○ But not every operationalization is perfect..
○ Key word for the future  reliability
 Operationalization

● How you operationalize a study

can change everything!
We
operationalize
variables!
 Variables

● Two main types of variables

which exist
● Independent variables
● Dependent variables
 Variables

● Independent variables
● Dependent variables
● What is the difference?
 Independent variable

● The variable the

experimenter manipulates or
changes
● Assumed to directly affect
the dependent variable.
Dependent variable

Variable being tested and measured in an

experiment and is “dependent” on the
independent variable
 Affecting
Independent Dependent
variable variable
No chocolate

Affecting
Eating
Concentration
chocolate

Eating
 Independent variable

● There’s usually two or more

levels or conditions of an IV
● One of these conditions can be
a ‘control’ condition or an
‘experimental’ condition
 Independent variable

● In an experimental condition,
the participant is ‘exposed’
to a variable or ‘treatment’
● In a ‘control’ condition,
there is an absence of the
experimental variable
 Two
experimental
conditions

VS.
Experimental Control

VS.
● In research, we want to see
how the IV affects the DV
● Based on this, we make a
conclusion about the cause
and effect
04 Correlation
 Correlations

● Let’s say we want to see the

relationship between the
independent variable and
dependent variable
● To do this, we use a
scatterplot
Independent Dependent
variable variable
Attendance Scores on a
to class test
 Correlations

● Correlations tell us the

relationship between two
variables
● To see when one variable
increases (or decreases), what
happens to the other one?
Think of..

● Two variables which are

positively correlated
● Two variables which are
negatively correlated
● Two variables which are not
correlated
 Correlations

● Correlational studies are

different from experimental
studies
● How?
 Correlations

Experiment Correlation
● Attempts to see how the ● Attempts to see how two
IV affects the DV variables are related
● Researchers manipulate ● IV is not manipulated
IV ● Cannot infer causality
● Infers causality

Causality = cause and effect

Correlations does not infer causation!

Third variable?
Third variables

Hot weather
Reverse causation

Depression Smoking
Reverse causation

Depression Smoking
Reverse causation

Depression Smoking
05 Experiments
 Experiment

● An investigation
which that allows us
to look for a cause-
and-effect
● IV  DV
 Types of experiments

● Laboratory experiment
○ Not in natural environment
(usually in a lab)
○ Strict control
○ There is manipulation of the
IV and measurement
 Types of experiments

● Laboratory experiment
○ Not in natural environment (usually
in a lab)
○ Strict control
○ There is manipulation of the IV and
measurement
○ Dependent variable measured
○ Looking for a causal relationship
 Types of experiments
● Field experiment
○ Done in the natural environment
○ Independent variable still
manipulated
○ Dependent variable is more
natural
○ Moderate control
 Types of experiments
● KEY WORDS:
● Standardization: keeping
procedure for each participant
the exact same
● Replication: keeping procedure
for between studies the exact
same
● Controls: ways to keep
confounding variables constant
 Types of experiments

● KEY WORDS:
● Reliability: extent to which
a procedure, task, or measure
is consistent  produces the
same results!
● Validity: measuring what you
actually intended to measure
VALIDITY in the middle
 Types of experimental designs

● Independent measures design

● A separate group of
participants used for each
level of IV
● Independent groups
● Exposed to the IV once
 Types of experimental designs

● Repeated measures design

● The same group of people
participate in every level
of IV
● Repeating their performance
● Part of multiple IVs
Independent vs Repeated measures
 Matched pairs design

● Participants are matched

with another participant who
is similar
● Age, gender, personality,
etc.
 What makes a good experiment?

● When variables such as age,

personality, gender, etc. affect
the DV, these are called 
confounding variables
 What makes a good experiment?

● Participants don’t know what the

study is about 
● Demand characteristics 
information about the study that
gives away what it’s about
● Participants in both groups are
similar
○ Participant variables  when
individual differences affect
DV
Advantages and disadvantages
 Independent measures design - advantages

● Participants are in one

condition once  demand
characteristics
● No order effects
● Participant variables can be
minimized by random
allocation
Independent measures design - disadvantages

● Participant variables can

distort results
● More participants are needed
 Repeated measures design - advantages

● Participant are in every

condition  less
participant variables
● Counterbalancing reduces
order effects
 Order effects

● Order effects occur from

participating in a condition
repeatedly
● Practice effects – improvement in
performance because of experience
● Fatigue effects – decline in
performance because tired/bored
 How to fix order effects  Counterbalancing

● Instead of one group being in

Condition A and then Condition B
 make two groups!
● Group 1 will do condition A first,
then condition B
● Group 2 will do condition B first,
● Then condition A
 Randomization

● Randomizing the order of the

condition for every participants
● Relevant if you have more than 2
conditions
 Repeated measures design - advantages

● Participant are in every

condition  less
participant variables
● Counterbalancing reduces
order effects
● Fewer participants needed
 Repeated measures design - disadvantages

● Order effects can distort

results
● See the condition more than
once, more exposure to
demand characteristics
 Matched pairs design - advantages

● Participants see one level

of condition  less demand
characteristics
● Participant variables 
less likely
● No order effects
 Matched pairs design - disadvantages

● Availability may be limited

● Limited by matching process,
matching criteria must be
chosen
 Summary (so far)

● We think of a study, and we

have an aim
● We choose an independent and
dependent variable (and
operationalize them)
● We choose what type of
experiment to do, and choose
the design
 Summary (so far)

● We make a hypothesis about the

experiment
● With hypothesis, it’s
important to be specific!
Hypothesis example

● Girls have more intelligence

than boys
● Girls will score more points
on the IQ test than boys will
 Next time

● Self-reports, case studies,

observations, etc..

● Until next class, read pages

5-18, takes notes and if
prepare questions for next
class if needed
 Design a mini-experiment
● Based on what you know, design
an experiment in a school ● Independent measures,
setting repeated measures, or
● Think about what is: matched pair design?
○ Your aim  what are you ● How much validity does
trying to find out? your study have? And
reliability?
○ Your hypothesis  what do
● What are the strengths
you predict?
of your study? And
○ Your variables  what is
weaknesses?
your independent variable?
And dependent variable?