My supervisor has given me questionnaires to use in my study and I'm tidying them up for appearances sake. However, although I'm aware of the authors that devised these questionnaires, I'm not sure that they are the originals, that is, word for word same. If some small words are missing, or if they have been changed slightly, can I still use these in the study? I've been on the Internet and am having trouble tracking them down.

I suppose you can, Delta, if you can't find the originals - however, I would acknowledge this as a potential limitation. Usually when a questionnaire is changed in some way, it is revalidated, as even slight amendments may change the measurement properties, such a construct validity. A slight change of wording may completely change the meaning of quest

questions (sorry my laptop is going funny!). Questions arranged in a different order can also do this. Anyway, sorry to rabbit on. I presume it's just the wording and not response format that has been changed?

Thanks for answering.

On some of the questionnaires, it's a slight varation on the wording of the questions and on another it is the response format and that is causing me some concern.

I'm meant to be predicting things on the basis of these. I'm with you, my instinct tells me I shouldn't be using them but it's 'difficult' to discuss such things with my supervisor. I'm starting to encounter a lot of confusion and concerns about this study. I took the study on because it was my impression the study was ready to run but this seems far from the case, indeed, I feel it is ill thought out, and nearly seven months in nothing of substance has been achieved. I was very motivated and always pushing for things to happen but now I'm just marking time, think I'm wasting my time and realising the only good thing about it is it has relieved me from unemployment. Sorry for the moan...

You and me both, Delta. I'm having great difficulty with my work - it's not going forward and I don't know what to do about it. With regards to your questionnaires, you could pilot them and try and collerate the scores with constructs similar to what you are trying to predict. It may take a bit of extra time, but may go some way to compensating for your problem. Or you could have validation by application, where subjects you know to have more of what you are trying to predict a 'higher score'. The use of improperly or non-validated questionnaires is not uncommon in clinical research (I can give you loads of examples), so it's not the end of the world. Why not try and get a focus group of 'experts' together to discuss the content and form of the questionnaires - that will lend some credence.

I hope things pick up for us both and thanks for the suggestions which I may follow up, if necessary. I have started to contact the authors to try and access the originals and I just hope they get back to me.

You've been extremely helpful, thank you.

Cool, Delta. My PhD is actually in the field of outcome assessment, so if you think I may be able to help, please PM me.

Many thanks and I may take you up on your kind offer.