Research Methods in Psychology Sampling and Experimental Design

Research Methods
in Psychology

Sampling and Experimental
Designs

Friday, 27 January 2012

Lesson 12: Research Methods Variables and
Hypothesising
Exam Question:

Read the following research question and respond to
the following
Does drinking alcohol effect reaction time?
a) What is the dependent variable? (1 mark)
b) What is the independent variable? (1 mark)
c) What is a possible extraneous variable? (1 mark)
d) Write an operational hypothesis for this research
question. (3 marks)


Model response:

a) DV: Reaction time

b) IV: Alcohol Consumption

c) Many possible responses e.g. age, gender, sleep deprivation,
strength of eye sight, natural skill at task

d) It was hypothesised that Victorian adults aged 20-30 who
drink 3, 375ml bottles of beer, 20 minutes prior to taking the
“Reaction Speed Simulator” test will produce a lower
reaction time (lower percentage score) than those who did
not consume alcohol.


Lesson 13: Research Methods: Sampling, Participant Selection and Experimental Designs

OUTCOMES:

Define population
Define sample
Describe the process of sampling procedures including
random, stratified and random stratified
Describe the process of participate allocation to groups
(experimental and control) including random allocation
Discuss the advantages and disadvantages of different
experimental designs including repeated measures, matched
participants and independent groups
Describe the placebo effect and ways of managing its
occurrence
Describe the experimenter effect and ways of managing its
occurrence

Sampling
Sampling is the selection of participants for a
research.

Population refers to the group which the research
wishes to draw conclusions from.

The term sample refers to the members of the
population that have been chosen to take part in the
research.

Sampling procedures must ensure that the sample is
representative of the population.

Representative Samples

Two techniques are used to
ensure a representative sample:
1)Random Sampling
2)Stratiﬁed Sampling and
Stratiﬁed Random Sampling.


Random Sampling
A sampling procedure in which every member
of the population has an equal chance of
being selected

Examples include:
1) Picking a name out of
a hat
2) Tattslotto
3) Closing my eyes and
selecting a number to
match that number with
student id numbers.


Stratiﬁed Sampling and
Stratiﬁed Random Sampling

Is used when you wish to eliminate the effects of
confounding variables.

The effects of a certain variable can be eliminated as a
possible confounding variable in an experiment.

The variable could be any personal attribute, such as
age, years of education, ethnicity, gender, IQ etc.

Involves six procedures:


1) Identifying a property that we believe may
interfere with the effects of the IV on the value of
the DV.
2) Measuring that property for each member of the
population.
3) Dividing the population into particular strata (groups)
based on the value of that variable.
4) Deciding on the number of participants required for
the experiment.
5) Selecting participants in the same proportions as
exist in the population to make up the sample
(stratiﬁed sample).
6) Selecting a random sample from each stratum, in
the same proportions as exist in the population
(stratiﬁed random sample).

Stratiﬁed Sample


Which to use?
Sophisticated, advanced Psychological
research studies use Stratiﬁed Sampling,
however it is very time consuming and
expensive, therefore majority of research
uses random sampling.

More so common, as the name suggests,
Psychological research uses a sample of
convenience, which although is biased is
quick, easy and cheap!


SAMPLING FLOW CHART


Participant Allocation:
Experimental & Control Groups
The experimental method uses two different groups
called the experimental and control groups.

The experimental group are exposed to the IV, known
as the ‘treatment’.

The control group do not receive the treatment (IV).

The purpose of the experimental group is to show the
effects of the IV on the value of the DV.

The purpose of the control group is to form a basis for
comparison with the experimental group.

Experimental and Control Group
Allocation
It is super important that all participants have an
equal chance of being in the experimental or control
group. That is Random Allocation.

When there is a large enough sample, both the
experimental and control groups will be equivalent on
all participant characteristics therefore the presence
or absence of the IV is the only difference between
them.

E.G. If we had all males in the experimental group and
all females in the control group, then an obvious
extraneous variable will be gender.

Experimental Designs

There are three popular experimental
designs

Repeated Measures Design

Matched Participants Design

Independent Groups Design


Repeated Measures Design
(within participants design)
In a repeated measures design participants experience
both the experimental and control groups.

This is possible by conducting the experiment on two
occasions and then comparing the two results.


ADVANTAGES:
1)Using the same participants means that confounding
variables that are participant depend are eliminated.
2)Allows for fewer participants to be used than with
other designs.

DISADVANTAGES:
1) Time consuming - drop outs
2) Confounding variables such as Order Effects:
a) Participants may perform better on the task when
doing for a second time (practise effect).
b) Participants may do worse the second time because
of fatigue or boredom.

Counterbalancing

Used to overcome order effect.

In counterbalancing, half the participants
will ﬁrst perform the task with the IV
present (experimental condition) and then
perform the task with the IV absent (control
condition). The other half of the participants
will experience the conditions in the reverse
order.


Matched Participants Design
Enables the researcher to identify a variable that is likely
confound and to eliminate the effects of this variable from the
experiment.

Participants can be ranked in accordance with their scores on this
variable and then allocated to the respective groups.


E.G. A sports coach developed a new game plan that
would help the team reach the playoffs. He decided to
test this by giving the experimental group the
instruction but not the control group. Because
individual skills would be a confounding variable, he
decided to ‘match’ the groups. The two highest skilled
players will be randomly allocated to either the
experimental or control group, the third and fourth
most skilled will then be randomly allocated to either
and so on and so forth until all players were allocated
to a group resulting in the same mean skill percentage
in both groups.


Advantages: The variable on which the
participants are ‘matched’ will not inﬂuence
the results because its effects will be the
same in the experimental and control groups.

Disadvantages: It is very time consuming
(and therefore expensive) to ﬁnd out the
value of this variable for every participant.
Also, if one of the pair drops out, the scores
for the other must also be eliminated.


Independent Groups Design
(between participants design)
Allocates participants to the
experimental or control group at random


Advantages: The independent groups design
can be done at once and drop-outs are
unlikely.

Disadvantages: The procedure needs a large
number of participants to ensure that the
spread of participant variables in the sample
will match the spread in the population.


This may effect the DV
Participants
resulting in invalid
expectations
results

PLACEBO EFFECT

Can be eliminated by using single blind procedure,
that is participants are unaware of which group
they are in.


For example: Experimenter
Experimenter treats participants differently
inﬂuences depending on the group they are
in which in turn inﬂuences the
experiment
behaviour of the participant and
effects the results

EXPERIMENTER EFFECT

Can be eliminated by using a double blind procedure, that is,
neither the experimenter or the participants are aware of
whether they are in the experimental or control group


Research Methods in Psychology Sampling and Experimental Design

More Related Content

What's hot (20)

Similar to Research Methods in Psychology Sampling and Experimental Design (20)

More from Crystal Delosa (20)

Recently uploaded (20)

Research Methods in Psychology Sampling and Experimental Design