SAS Support Communities

but what I should write in the code to get the ANOVA ? or Wilcoxon I only know how to get chisq I write proc surveyfreq data=......; tables ...../chisq; run;   But what I should write in the code to get ANOVE and Wilcoxon ... View more

You can use Proc SURVEYREG to fit an ANOVA to complex survey data.   There are no procedures to fit non-parametric models, although I suppose you could apply SURVEYREG to the ranks.  You would have to do a little research to see how this would be different from the SRS-world but it might be a possibility.  I believe this book even discusses it although I don't recall where exactly in the book it is.           ... View more

BTW, I fixed the link in Rob's post -- there was an extra space in the URL.  Works now! ... View more

Sir, I am doing linear regression not logistic regression. ... View more

For purely nominal variables, you can use PROC FREQ and the chi-square statistic to test the association between variables.  If the variables are ordinal (they can be ordered, such as 'small', 'medium', and 'large'), you can use PROC FREQ to compute tetrachoric or polychoric correlations between discrete categories. Use the PLCORR options on the TABLES stmt.   Lastly, for multivariate statistics, you can use PROC CORRESP to study correlations between categorical variables. Roughly speaking, correspondence analysis for discrete variables is similar to principal component analysis for continuous variables, except contributions to the chi-square statistic are used instead of the variance. ... View more

Yes ... View more

http://support.sas.com/kb/22/585.html ... View more

You're not asking for programming help here, you're asking for statistical advice/methodology. There's a little less of that on this forum and I'm probably not the best person to answer anyways.  I've moved your question (and edited the title to correct for spelling mistakes) to the Statistical Procedures forum which is more appropriate.    I would also suggest you review the assumptions for linear regression and determine if it's appropriate for your data. You can also post this on Cross Validated, which is specifically designed for these types of question, but make sure to elaborate your question, including your hypothesis and context of the question may help. Otherwise you're likely to have your question down-voted and rejected.      ... View more