Experimental Designs Flashcards
Experimental designs
Refer to the way in which participants are used in experiments.
Independent groups
Two separated groups of participants experience two different conditions of the experiment. If there are two levels of the IV this means that all participants experience one level of the IV only. Performance of two groups compared.
Eg water group and energy drink group
Repeated measure
All participants experience both conditions of the experiment
Following this the two mean scores from both conditions would be compared to see if there was a difference.
Matched pairs design
Participants are paired together on a variable or variables relevant to the experiment. For example in a memory study participants might be matched on their IQ. 1,2 go together, then 3,4 etc. Then one participant from each pair would be allocated to a different condition of the experiment. This is an attempt to control confounding variable of participant variables and often necessitates the use of a pre-test.
Then run in same way as independent groups design after matching.
Evaluation-independent groups
Participants who occupy the different groups are not the same in terms of participant variables. May act as confounding variables reducing validity of the finding. Participants should be randomly allocated to reduce this.
Independent groups are less economical than repeated measures as each participant contributes a single result only.
Strength of using independent groups are order effects are not a problem. Participants are also less likely to guess aims.
Evaluation of repeated measures
Each participant has to do two tasks and order of tasks maybe significant. Researches he’s counterbalancing instead .
Order effects also arise because repeating two tasks could create boredom or fatigue that might cause deterioration of performance on second task. Alternatively participants performance may improve if skill based task. Order effects act as confounding variable.
More likely participants work out aim of study. Therefore demand characteristics happen
Strength is that participant variables are controlled (higher validity) and fewer participants are needed (less time and money spent recruiting)
Evaluation of matched pairs design
Participants only take part in sing condition so order effects and demand characteristics aren’t problem.
Participants can never be matched exactly so there may still be be participant variables-affect DV
Matching may be time-consuming and expensive, particularly if pre-testing is required. Less economical than other designs.
