An experiment has an independent variable and a dependent variable.

In order to find out whether the

IV did affect the DV, we always need a comparison condition – a condition where there is a different
level of the IV. The way that the two levels of the IV are delivered is called experimental design.

TYPE OF EXPERIMENTAL DESIGN

Type of experimental design Evaluation: Limitations Method of dealing with the
limitations
Repeated measures design The order of the conditions Researchers may use two
All participants receive all may affect performance (an different to reduce a practice
levels of the IV; for example: order effect). For example, effect - though the two tests
each participant does a task participants may do better on must be equivalent. This can
with the TV on and then a the second test because of a be done by constructing a test
week later each participant practice effect or because they of say, 40 items and randomly
completes the task without the are less anxious. Alternatively, allocating items to Test A and
TV on. in some situations participants Test B.
may do worse on the second
We compare the test because of being bored The main way that order
performance (DV) of the with doing the same test again effects are dealt with is using
participant on both tests. (boredom effect) counterbalancing.

When participants do the In order to avoid participants

second test they may guess the guessing the aims of a study, a
purpose of the experiment, cover study can be presented
which may affect their about the purpose of the test.
behaviour. For example,
some participants may
purposely do worse on the
second test because they want
it to appear as if they work
less well in the afternoon.
Independent groups design The researcher cannot Randomly allocate
Participants are placed in control the effects of participants to conditions
separate (independent) participant variables (I.e. the which (theoretically) distribute
groups. Each group does one different abilities or participant variables.
level of the IV. Group A does characteristics of each
the task with the TV on, participant). For example, Random allocation can be
Group B without the TV. participants in Group A might done by putting the
happen to have better participant names in a jar and
We compare the memories than those in drawing out the names so that
performance (DV) of the two Group B. This would act as a every other person goes in
groups. confounding variable. Group A.

Independent groups design

needs more participants than
repeated measures design in
order to end up with the same
amount of data.
Matched pairs design It is very time-consuming and Restrict the number of
A compromise is to use two difficult to match participants variables to match on to make
groups of participants but on key variables. The it easier.
match participants on key researcher probably has to
characteristics believed to start with a large group of Conduct a pilot study to
affect performance. participants to ensure they can consider key variables that
obtain matched pairs on key might be important when
variables. matching.
 Then one member of the pair
is allocated to Group A and It is not possible to control all EVALUATION:
the other to Group B. The participant variables because STRENGTHS
procedure is then the same as you can only match on You can work out the
for independent groups. variables known to be strengths of each
relevant, but it could be that experimental design by
It is important to realise that others are important. For looking at the limitations of
the characteristics for example, in a memory the other design.
matching must be relevant to experiment you might match
the study. In other words, you on memory abilities but later
wouldn’t need to match find that some of the
participants on gender if you participants had been
were testing memory – unless involved in a teaching
there was some evidence that programme to boost memory
gender was a potential skills and you should have
confounding variable. matched on this.
COUNTERBALANCING
Counterbalancing ensures that each condition in a repeated measures design is tested first or
second in equal amounts. If participants do the same memory test first in the morning and then in
the afternoon, we might expect them to do better on the second test because they have had some
practice – or they might do worse because they are bored with the task. These are called order
effects, which can be dealt with by counterbalancing.

There are two ways to counterbalance order effects. In each case, we have two conditions:
• Condition A - test done in the morning.
• Condition B – test done in the afternoon.

Way 1. AB or BA
Divide participants into two groups:
• Group 1: each participant does A then B.
• Group 2: each participant does B then A.
Note that this is still a repeated measures design even though there are two groups of participants,
because comparison will be made for each participant on their performance on the two conditions
(morning and afternoon).

Way 2. ABBA
This time all participants take part in each condition twice.
• Trial 1: Condition A (morning)
• Trial 2: Condition B (afternoon)
• Trial 3: Condition B (afternoon)
• Trial 4: Condition A (morning)
Then we compare scores on trials 1 and 4 with trials 2 and 3. As before, this is still a repeated
measures design because we are comparing the scores of the same person,

Key terms
Counterbalancing – An experimental technique used to overcome order effects when using a
repeated measures design. Counterbalancing ensures that each condition is tested first or second in
equal amounts.
Experimental design – A set of procedures used to control the influence of factors such as
participant variables in an experiment.
Independent groups design – Participants are allocated to two (or more) groups representing
different levels of the IV. Allocation is usually done using random techniques.
Matched pairs design – Pairs of participants are matched in terms of key variables such as age and
IQ. One member of each pair is allocated to one of the conditions under test and the second
person is allocated to the other condition.
Order effect – In a repeated measures design, an extraneous variable arising from the order in
which conditions are presented, e.g. a practice effect or fatigue effect.
Random allocation – Allocating participants to experimental groups or conditions using random
techniques.
Repeated measures design – Each participant takes part in every condition under test, i.e. each
level of the IV.