Hi all, This question is more stats related than R related, but as I will be doing these analyses in R, there may be some programming questions to follow. I figured it best to start here. To keep things relatively simple, assume that I recently fielded a survey over three weeks. I have three dichotomous variables that indicate what week a respondent responded in. I would like test to see if the percent responding in each week is significantly different. (For example, 24% responded in week 1 and 28% in week 2. Were there significantly more respondents in week 2?) Here is the data layout: ID Week1 Week2 Week3 1 1 0 0 2 0 1 0 3 1 0 0 4 0 1 0 5 0 1 0 6 0 0 1 7 0 0 1 .... It seems like this could be done as difference of percentages test, but I am not sure how to account for the third option. Any help is appreciated. Bryan [[alternative HTML version deleted]]