One of the courses I maintain has its student evaluation built as an Ungraded Survey. A few of the questions use the multiple-dropdown format. We've recently discovered a "problem" with the reporting for these questions in the Student Analysis .csv export file, in that it only logs selections, not failures to select.
For example, let's assume a question has four dropdowns — each of which has No, Maybe, and Yes as the options. If a student makes a selection on all four, the cell for that question in the .csv file might look like No,Yes,Maybe,Yes. That tells us that the student selected "No" for the first dropdown, "Yes" for the second one, and so on.
But if the student only makes selections for two of the four, the cell would look more like No,Yes. That is, there's no indication which of the four dropdowns the student made selections for. That makes it impossible to extract accurate data for that student, for that question, from the .csv.
For all kinds of reasons, it would be ideal to find a way to generate a "failed-to-select" cue in the .csv — using the preceding example again, something along the lines of No,[No Answer],[No Answer],Yes.
Does anyone have experience with this? If so, did you find a solution? Thanks in advance for potentially saving us many hours of retooling and/or tracking down individual responses inside Canvas.
Marc Faris, Instructional Designer