This is a question for people doing demographics / symptom assessment research: My team is using several instruments to quantify patient symptoms (Neck Pain Disability Index, McGill Short Form Pain, DASS 21). Right now we have patients fill them out on paper forms. My question is: if we convert these forms to be collected on a computer (e.g. with check boxes), can we compare our results to those collected by other researchers using the same forms collected on paper? I am not sure if it is ok to assume the responses would be equivalent.
Dear Bryn,
Even slight changes to the original questionnaire (e.g. visual appearance, layout) may affect its measurement properties and participants' responses.
You may find useful information on this topic in Dillman, D.A., Christian, L.M. (2005). Survey Mode as a Source of Instability in Responses across Surveys. Field Methods, 17(1), 30-52.
I would suggest you to check the comparability of your computer and paper-based surveys within a pilot study, before you fully implement the computer-based assessment.
Even if the two survey modes do not show perfect agreement, your pilot study may provide you reasonably accurate conversion equations.
Cheers,
Zeljko
Hi Bryn,
Have you looked at the NIH PROMIS? These forms are designed for computerized administration and are highly robust. The administrators are very helpful with any questions. It has been very useful in our research.
Dr. Chuck Woodfield
http://www.nihpromis.org