Hello all.

I am using SPSS to analyze associations between genetic polymorphisms (i.e. AA, AG or GG genotypes) and certain variables, both scale and nominal. I would like to add a trend analysis, like the Cochran-Armitage trend analysis but can't find it or anything like it anywhere in SPSS v.13, is this at all possible to do in SPSS and if so, how?

Thanks in advance.

This is from the SPSS-L mailing list:

The linear by linear association chi^2 statistic in CROSSTABS is one version
of a trend test. A logistic regression is another option. SPSS doesn't offer
specifically the Cochran-Armitage trend test, which is based on a linear
probability model rather than a linear logistic model, but running a linear
regression will produce a p-value that's reasonably close to what you'd get
if you went to the trouble of calculating that statistic.

David Nichols
Principal Support Statistician and
Manager of Statistical Support
SPSS Inc.

Great, thanks a lot :)

I want to calculate the p value for trends in odds ratios for a given independent variable in a logistic regression analysis using spss.
How can I do this?

Thanks in advance

Maybe I am misinterpreting your question, but I don't think that there can be such a thing as a trend in an odds-ratio, because the odds-ratio is produced by a 2x2 table. Please correct me if I am wrong, anyone.

Thanks for your mail.
I want to find social inequalities in respect to the presence of dental caries. In this regrad I intend running a logistic regressiion analysis (LRA) with presence of dental caries as the dependent variable and several socio-economic indicators as independent variables. The independent variables have more than two strata. With the LRA it will be possible get the odds ratios for the different strata of a given socio-economic indicator. I want to find the trend in these odds ratios for a given indicator using spss.
I have given an example for clarity

occupation status:
group 1 (lowest)- 2.01
2 1.76
3 1.45
4 1.50
5(highest) 1.00
What I want to find out is the p value for trends in these odds in order to conclude that there is a social gradient with respect to dental caries?
In the literature there are many papers that present the p value for trends in odds using spss in similar situations.
Thanks
Lilani

Oh, I see.
In that case, I guess you might be able to use the linear-by-linear or the logistic regression on the OR's directly, as suggested in the quote from the SPSS-L mailing list brought by Quark earlier in this thread:

This is from the SPSS-L mailing list:

This is from the SPSS-L mailing list:

Dear friends

I think your suggestion is for comparing trend between to form of one variable according to another.
I want to know when we want to estimate the p for trend for one variable suach as age what we should do?
for example we want to know weather number of diabetic patients increases by increase in age or not? could we use spss and how?
regards