Help need re survey data evaluation

[y_serenity]

Dear Forum Members,

I am currently evaluating a survey I conducted as part of my PhD. My background is in languages and education and my knowledge of statistics is very limited. I am looking for help with two questions that I have been unable to solve for quite some time now. Here it goes:

To begin with, I identified the target group of the survey (teacher instructors at German universities who teach either language practice (LP) or teaching methodology (TM) to teacher trainees doing an English degree) and found 414 participants who fit all the requirements. I wrote to all of them and 212 took part in the survey which was administered online.

The purpose of the survey was to find out to what extent these instructors were familiar with a certain teaching resource and to what extent they were using it for their classes.

QUESTION 1:
Ok, in Question 1, the participants were asked to state which area they are teaching in: LP and/or TM. Thus, the total of the participants could be divided into three groups: LP (48.6&#37 TM (35.8%) and LP+TM (15.6%).

The remaining questions of the survey could now be looked at either as total results or the results for the individual answers could be looked at in the sense what is each groups contribution to this answer. I have attached an Excel sheet to demonstrate what I have done to do this.

My question is - is this correct? Obviously because the groups are of different size, in numbers their contribution to each question would be dependent on this. I would like to be able to make a statement like: It appears that Group LP is more likely to ... etc.

I am on the right track here?

QUESTION 2:
Because the population for the survey was closed and small enough to contact everyone, does that have an impact on things like calculations for representativity? Do I have to look into things like p-value etc?

I would greatly appreciate any help, pointers in the right directions, etc.
Thank you in advance.
y_serenity

[JohnM]

It appears that with Question 1, you're over-thinking it a bit. If in general you want to compare the three groups (LP, TM, LP+TM) on the various responses to the questions, then do exactly that......or are you trying to do something else?

On Question 2, you did not get 100% participation (only around 51%), and on top of that, it's not clear that you were able to "locate" the entire population (there may be more than 414), so yes, you will need to apply inferential methods and use p-values, etc.

In terms of representativeness, you could just compare demographics of the 212 to demographics of the 414 to see how well they match up in percentages.

[y_serenity]

John,

first of all, thank you for taking the time and looking at my stuff.

I am still not quite sure what to do about those three groups though. You suggested: "If in general you want to compare the three groups (LP, TM, LP+TM) on the various responses to the questions, then do exactly that"

I thought that's what I was doing but obviously I went wrong somewhere. Could you point out to me what your approach would be?

Regards,

y_serenity