I have a question regarding the comparisons of D-Efficiency measures for different cbc designs.
In the manual it says: This value (pth root of the determinant) also depends on the number of respondents x tasks, so when comparing two designs, it is important to hold the number of respondents x tasks constant.
I wonder whether it means that respondents and total tasks should by constant when comparing to designs OR respondents and tasks per respondent should be constant.
Can I compare the efficiency of two designs when number of random tasks per respondent, respondents, none option percentage is held constant but the number of questionnaire versions varies? So for example compare designs with only 1 vs 25 questionnaires?
Thanks very much!