Hello all,
I have a datafile with subscales totals from various questionnaires. Some subscales are from the same questionnaire, so if a participant didn't fill that questionnaire in, it has a missing value on two (or more) scales from the same questionnaire.
I would like to know whether I can put all of the subscales in when I run a missing value analysis, or whether that would be unadvisable. If I put all of the scales in the analysis, the Little's test is not significant, so that allows me to impute the missing values with EM. But I am doubting whether it is appropriate to enter the subscales, since they are obviously related to each other.
Hope somebody can help
-scisci
I have a datafile with subscales totals from various questionnaires. Some subscales are from the same questionnaire, so if a participant didn't fill that questionnaire in, it has a missing value on two (or more) scales from the same questionnaire.
I would like to know whether I can put all of the subscales in when I run a missing value analysis, or whether that would be unadvisable. If I put all of the scales in the analysis, the Little's test is not significant, so that allows me to impute the missing values with EM. But I am doubting whether it is appropriate to enter the subscales, since they are obviously related to each other.
Hope somebody can help
-scisci