Controlling design issues Flashcards
Counterbalancing
Dividing the pt’s into two groups and have each group perform a slightly different task, so that any mistake in the task is cancelled out.
It eliminates confounding varibles from an experiment by giving slightly different treatments to different pt groups.
Randomisation
Randomly allocate pt’s to experimental conditions by putting all names on paper, then place in a top hat and shuffle. Extract at random pulling each name into different experimental conditions. This way you will have a variety of individual differences that will eventually balance out across conditions.
This reduces Pt variability and reduces sample bias.
Single blind techniques
When the PT’s are deliberatly kept ignorant of either the group to which they have been assigned or key info about the materials they’re assesing, but the experimenter knows about this.
Possible sources of human bias and influence from the pt’s and for researchers, they encourage alternate reporting and data analysis.
Double blind method
Neither pt nor researcher knows which treatment or intervention pt;s recieve until trial is over.
reduces exprimenter effects, situational factors and participant factors.
Standardised procedures
Procedures used in research are kept the same. Changes in data can be attributed to the IV.
Reduces pt’s guessing the study and ambiguity and situational variables.