I have requested 30 participants to answer the same question about 100 images: "How easy is it for you to recognize this image?" on a scale from 0 to 100, where 0 is easy and 100 is difficult.
I want to 1) eliminate extreme scores, so the closer they are to 100, the more unfavourable these images will be to me. 2) I would like to identify participants with extreme scores, so for instance if they were bored and they have different scores than the rest I would like to identify these.
Would I do a simple descriptive analysis to see whether some participants have higher means than the rest (or look at standard deviations)? Also, how can I compare the scores on the images, would a graph with frequencies do the trick?
Thanks for your time!