Hello,
I generated a D-efficient design in Ngene with 24 choice situations, divided into four blocks of six choice situations each.
In the Excel output exported from Ngene, the choice situations assigned to the same block are not listed consecutively. When I first tested the survey and was assigned to block 1, SurveyEngine displayed the first six rows of the output/Excel file. However, these were not the six choice situations that Ngene had assigned to block 1.
I therefore sorted the Excel file by the Ngene-generated block-column, so that all six choice situations belonging to each block were placed together. I did not change the attribute levels, the alternatives, the choice situations or the block assignments. I only changed the order of the rows. After this adjustment, each respondent is randomly assigned to one block and receives the six choice situations that Ngene assigned to that block.
Could you please confirm whether this is the correct way to implement the Ngene-generated blocks in SurveyEngine? Could you also confirm that sorting the rows in Excel does not affect the D-efficiency of the full design?
Thank you very much for your help.
Best regards,
Lauren