Experimental Design Flashcards
what is an experimental design?
3
the ways that the two levels of IV are delivered
how participants are allocated to different IV conditions or levels in the experiment
a set of procedures used to control the influence of factors such as participant variables in an experiment
three types of experimental design
3
independent groups
repeated measures
matched pairs
independent groups design
3
participants are placed in separate, independent groups
each group does one level of the IV — e.g. group A does the take with the tv on, group B does the task with the tv off
then compare the performance (DV) of the different groups
repeated measures design
3
all participants take part in all levels and experience all conditions of the IV
EXAMPLE = each participant does a memory test with the TV on, then maybe a week later the same participants do a similar test with the TV off
then compare the performance (DV) of the participants in the two different conditions
matched pairs design
4
match participants on key characteristics that may potentially affect the DV, these characteristics must be relevant to the study (e.g. age or IQ)
one member of the pair is allocated to group A, the other is allocated to group B
each group does one level of the IV — e.g. group A does the take with the tv on, group B does the task with the tv off
then compare the performance (DV) of the different groups
2 strengths of independent groups design
2
AVOIDS ORDER EFFECTS — avoids order effects such as practise or fatigue because participants only participate in one condition - they won’t get as tired or bored as if they were doing several conditions
REDUCES INVESTIGATOR BIAS — each participant has an equal chance of being assigned any group as it’s done randomly, reduces investigator bias
2 limitations of independent groups design
2
TIME CONSUMING — more participants are needed than repeated measures design in order to end up with the same amount of data, thus its more time consuming and will cost more money
PARTICIPANT VARIABLES — participant variables (age, gender, abilities, etc) may affect results and it’s not always possible to equally separate participants based on this, for example group A might happen to have better memories than those in group B, this would act as a confounding variable and interfere with results
2 strengths of repeated measures design
2
FEWER PEOPLE NEEDED — fewer people are needed as they all take part in all of the conditions, saves time
REDUCES PARTICIPANT VARIABLES — participant variables are reduced because each group takes part in all of the conditions
2 limitations of repeated measures design
2
ORDER EFFECTS — high chance of order effects, order of the conditions may affect performance, participants may do worse in second condition because they’re more tired or bored (boredom effect) OR they might do better because they know what to do, have had practise and are less anxious (practice effect)
DEMAND CHARACTERISTICS — participants may do second test, guess purpose of experiment and change their behaviour (they might do purposefully worse, etc), known as demand characteristics and the ‘screw you effect’ which ruins results and reduces validity
2 strengths of matched pairs design
2
REDUCES PPT VARIABLES — reduced participant variables because researcher has tried to pair them up so that each group has an equal spread of people with similar abilities and characteristics
AVOIDS ORDER EFFECTS — avoids order effects because each of the participants only participate in one condition, counterbalancing is not necessary for this design
2 limitations of matched pairs design
2
TIME CONSUMING — very time consuming as it’s hard to closely match pairs based on key characteristics, researcher probably has to start with a large group to ensure they can match enough people based on key variables
NOT POSSIBLE TO CONTROL ALL PPT VARIABLES — not possible to control all participant variables, you can only match variables known to be relevant, but it could be that you did not realise an important characteristic until after the experiment, e.g. you might match on memory abilities but later find that some ppts have been involved in a teaching programme to boost memory skills and you should have matched on this
method of dealing with limitations of independent groups
2
randomly allocate participants to conditions which would theoretically distribute participant variables evenly
random allocation can be done by putting the participant names in a hat and drawing out the names so every other person goes in group A, the rest in group B
method of dealing with limitations of repeated measures
3
researchers may use two different tests to reduce the practice effect, though the two tests must be equivalent — can be done by constructing a test of 40 items and randomly allocating items to Test A abs Test B
counterbalancing can help deal with order effects
in order to avoid participants guessing the aims of a study, a cover story can be presented about the purpose of the test
method of dealing with limitations of matched pairs
2
restrict the number of variables to match on to make it easier
conduct a pilot study to consider key variables that might be important when matching
what is counterbalancing?
3
ensures that each condition in a repeated measures design is tested first or second in equal amounts
can help tackle order effects
2 ways to counterbalance order effects
• AB or BA
• ABBA
