Dear All,
I would be grateful for any help you can give me with the following enquiry.
I have undertaken an experiment of mixed design:-
Between subject factors are age at two levels (young vs older adults) and test format at two levels (recall vs recognition questions)
Repeated measure factor is memory performance (accuracy and quantity) measured on two occasions under free and forced report instructions.
Memory performance is a further repeated measures factor (type of content at two levels - action vs description) - all participants provided a score on each.
The participant sample total is 195 - in each of 4 groups (53, 48, 46, 47 participants in each group)
My problem with outlier scores has arisen on the performance measures at forced report (time 2) - specifically when analysing the data for the type of content DVs.
Using the "explore" function on SPSS - the K-S statistic is significant for the overall performance measure - however there are no outlier scores identified.
Separate normality tests on the two content types have shown the K-S statistics to be significant - additionally different participants have recorded extreme scores on the different content types.
I am considering "transforming" the scores but would appreciate advice on whether this is the correct action to take and if so which form of transformation would be most appropriate
Many Thanks
Hazel
Hazel McMurtrie
Postgraduate Research Student
Department of Psychology
University of Strathclyde
Graham Hills Building
40 George Street
Glasgow G1 1QE
Tel (0141) 548 4392
http://www.strath.ac.uk/psychology/people/research/mcmurtriehazelms/#d.en.157784
|