Pilot Studies & Experimental Design Flashcards
What is meant by a pilot study?
Small-scale prototypes of study carried out in advance of full research to find out if there any problems with:
Experimental design - do participants have enough time to complete tasks
Instructions for participants - are instructions clear?
Measuring instruments - including behavioural categories in observational research & questions when using questionnaires - allow for categories & qs to be checked & modified where necessary
Pilot studies ensure time, money & effort aren’t wasted on flawed methodology
What are repeated measures?
Participants take part in each condition of experiment
Data obtained from both conditions then compared for each participant to see if there was difference
Evaluate repeated measures
Strength - fewer participants required
E.g. - fewer participants recruited as they go through both conditions
Ex. - design less costly & time consuming - reduces possibility of participant variables such as individual differences playing part in results, meaning effect of IV on DV trusted with greater confidence
Limitation - potential order effects
E.g. - parties who experience practise effects may perform better in 2nd conditions as they know what’s expected of them, whilst parties who experience fatigue may perform worse in 2nd condition
Ex. - can skew results - we don’t have full confidence IV has had particular effect on participants
How do we address the issue of practise effects in repeated measures?
Counterbalancing - half parties take part inn condition ‘A’ then condition ‘B’ with other half doing vies versa
Any order effects experienced by those who started in Condition A should be offset by those who started in Condition B
What are independent groups?
Uses 2 separate of participants - 1 group in each condition of experiment
Participants allocated to their group by random allocation, ensuring each parti has equal chance of being assigned to each
Evaluate independent groups
Strength - avoids order effects
E.g. - parties only do 1 condition in experiment - less likely to get bored & give up, reducing impact of order effects
Ex. - we have more confidence results down to effect of IV on DV rather than other factors
Limitation - participant variables may affect results
E.g. - difference in age, sex or social background
Ex. - these things may affect results by acting as EVs on DV which means psychologists can’t be certain that IV caused changes measured
What are matched pairs?
Pairs of participants matched from sample in terms of key variables, e.g. age, IQ
After matching takes place, participants treated much like those in independent measures
One member of each pair placed in experimental group & other member in control group
Evaluate matched pairs
Strength - order effects (such as practise/fatigue) less of issue compared to repeated measures design
E.g. - participants only take part in 1 condition of experiment & thus they’re likely to become bored & give up
Ex. - these EVs won’t affect DV & thus results as they don’t skew them
Limitation - More participants required
E.g. - different participants take part in different conditions of experiment, making design more expensive & time consuming for researcher. Very difficult trying to find close pairs
Ex. - individual differences may still play role in measurements of DV reducing certainty that IV affected change