Experimental Designs

what types of experimental design are there?

independent groups, repeated measures, matched pairs

independent groups

two different sets of participants used in the conditions of the experiment, assigned by random allocation

strength of independent groups

avoids order effects, e.g. fatigue, practice, and demand characteristics. Maintains internal validity of results

weakness of independent groups- participant variables

participant variables may affect results as diff people take part in each condition so may be individual differences between groups which influence DV

weakness of independent groups- sixe

larger sample size needed- resource intensive in terms of time, cost, recruitment

repeated measures

All participants take part in all conditions of the experiment

strength of repeated measures

controls for participant variables, increasing internal validity

less participant needed so smaller sample size needed- cost effective and efficient

weakness of repeated measures- demand characteristics

exposed to all conditions so participants more aware of aim of study and may alter their behaviour accordingly leading to biased results, reducing validity of findings

weakness of repeated measures- order effects

take part in multiple conditions so performance may improve with practice or decrease with fatigue, so order effects act as confounding variables

what is counterbalancing

An attempt to control for the effects of order in a repeated measures design: half the participants experience the conditions in one order, and the other half in the opposite order.

matched pairs

participants paired based on key characteristics and then randomly assigning one member of each pair to the experimental group, and the other to the control group

strength of matched pairs

avoids order effects and demand characteristics, as only take part in a single condition

weakness of matched pairs- time

time consuming and labour intensive as often requires pre-testing which delays research process. may be impossible to find matches if traits are complex/ hard to measure

weakness of matched pairs- participant variables

unlikely participants are identical in every way as may still be uncontrolled differences which could influence DV, reducing internal validity