I am using the macro intracc from SAS and developed by R. Hamer, but I am not sure if I use it correctly. I want to assess the reliability between a test-retest of a questionnaire. My data are the following: 146 subjects have filled in twice a questionnaire containing 146 items.
My first idea was that rater was the repetition variable (i.e. if data come from the test or from the retest, so repetition can take either the value 1 or 2) and target was item (146 different items), but each of them being rated 146 times (so in total 21316 values). It seems to me that something is not taken into account, in particular it doesn't take into account that each item have been rated by the same subjects (in other words, the variable subjects appears nowhere).
Has anyone used this ICC to measure the reliability of a test-retest for a questionnaire and could advise me?