....another beginner


New Member
Hello everyone, :wave:

I'm a beginner with stats but also a beginner with forums, so not even sure if I have posted this in the correct place - it took me 10 minutes to find out how to create a new thread.

Anyway, I'm doing some research at the moment that is looking at teachers' attitudes. I have administered attitudes scales to 25 teachers before they completed a training course and again after they completed the training course.

I also administered the same attitude scale to a comparison group (who were not involved in the training) at the same time as the test group completed the pre-training attitude scales. However, I did not administer the scales again to the comparitive group after the training. This comparison group was by no means meant to be a control group- it is just to look at baselines of the scales and make broad and general comparison.

I'm now looking to analyse the data now and I have been reading lots around spss and which stats to use.

I want to find out if there has been an overall significant increase in teachers' attitudes as a result of attending the training course.

I would very much appreciate some input from soemone who knows about these things, I find it very confusing. I have been trying to read about it but think I have just been getting myself more confused! :shakehead

If anyone could help, that would be amazing!
Hi enid,
Based on what you wrote I would suggest you seek to consult with a statistician directly. This is not to say that that the members of this board don't "knows about these things" or "which stats to use" but because there will be a lot of specific details involved in the course of the analysis and interpretation of your data. I believe the advice on which buttons/menue items you should use in SPSS will not be helphul in your case.


Fortran must die
I agree with the above post but I would also point out that the comparison group won't be usable in all liklihood because you don't have a before and after score. In addition 25 cases won't be enough in most statistical methods.