SPSS Confussion, what test

Hi there,

We (my girlfriend and myself) are trying to work out a methodology/test we can run on the load of data that she has collated.

Basically she has a set of data that has is pre and post operation on patients with back injury.

We have looked at many variables from BMI to pain threshold scale (e.g. a scale of 1 to 10). In amongst all this we have age, height, weight, etc. also occupation and type of operation. What I'm trying to say is we have a lot.

So my/our problem is as follows

1/ She did not decide on a stats test before she collected the data, hence the reason she has so much.

2/ We are trying to filter out the relevant data that will give us meaningful results.

3/ I have knowledge of this type of analysis but from 5 years ago, I do not have the memory to recall what I should be looking for in data sets and which will lend its self to stats. analysis.

So I am asking if any of you guys could point us in the right direction, our Hypothesis is that the operation has a positive effect on the patients quality of life. We also have information on the amount of activity the patients done pre and post opp. which would indicate quality of life for example.

If this is making sense and someone can draw even the slightest bit of interest from this post please could you suggest a direction for us to start looking.

My first instinct was multi variance ANOVA (?) plugging in all the data. Would this lend itself to pre and post analysis.

I can read output etc. but the bible of choosing an analysis test has been lost over the years.

Thank you in advance

Kind Regards

Samuel....and MissyP ;-)


TS Contributor
Shouldn't it be like
H0= m1=m0 (m0 being pre op mean, and m1 post op mean)

You can compare the mean of the movements, and if z cal does fall in critical region you reject H0, and accept H1 .Assuming you have enough data to support normal dist.
I might be wrong:p
So you need to screen the variables you have,right?

Could you send me the data? You can contact me thru the Contact form in my signature.
Is that normal to send data? Would it help if we sent you the types of data er aren't too sure about sending data, no offence but it is of a medical nature.


S and S