How can I do a PRE-TEST???


Hi all,

I hope someone can give me some insight on the following issue:

For my survey I need to generate some categorical data that participants just need to tick off. These are not massively important for my research but function as descriptive statistics or possibly moderators should I find them to be influential on my main constructs and their relationship.

Initially I wanted to carry out a big qualitative research study to generate those items/categories including its own research design. That would have made my study a mixed method study and obviously would involve a whole lot of approval and work. However, I don't think that this is neccessary only to get a few items and it will not stand equal to my actual quantitative study.

hence, I stumbled upon a paper that carried out a pre-test to generate items. It wasn't mentioned in a big fashion, just the facts were stated. So my question is: Do I need to develop a whole separate research design for a pre-test? Has anyone good examples or some literature that would help me clarify what a pre-test consists off and how I can place it as part of my bigger study.

Just wanted to mention that with pre-test I don't mean a pilot study. It would actually entail its own set of questions but on a smaller scale than my large survet that will follow.

Many thanks,


I also wanted to collect some characteristics of participate and case studies. As I didn't know which factors would be important, I used extensive standardised questionnaires. Developing such questionnaires might be a project on its own. The key here is standardised


Dr Jeckyll, thank your for your post. What was the basis of your standardisation? Where did you get the original characteristics from?


Hello - I conducted a retrospective pre test using a tried and ested questionnaire in my MA. I simply designed it so that the partcipants could answer "then" and "now" to each question. HTH x


"I stumbled upon a paper that carried out a pre-test to generate items"

Contact the authors of this paper and ask them for the questionnaire. Most people are happy to help.
This is what I did :)