sas.kenya@GMAIL.COM wrote: > >Hi All, > Hope this is the right forum to ask this ...... I'm an MSc Applied >Stats student and hope to use data from the project I am employed in for my >thesis/project. Just wondering ...... should I do multivariable analysis or >logistic? Is there a "gold standard" when analysing data for academic >purposes. > >Thanks, >Richie.

Well, I guess this is a reasonable choice of forum.

There is no gold standard in analysis. You should know this by now.

The choice of analysis should be made based on your data, your meta-data, your process, your overall goals, your data scope, your hypothesis tests, ....

Logistic regression is typically a multivariate process, so I guess you are asking about multivariate techniques like cluster anlaysis, factor analysis, PCA, PLS, and so on. There is no magic rule that says "use cluster analysis here and use factor analysis there". You just have to study the data and figure out what questions need to be answered.

If your major prof has a bee in his bonnet about one particular type of analysis, I would recommend that you focus on that, instead of getting yourself in trouble with him from the start.

HTH, David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330

_________________________________________________________________ Don’t miss your chance to WIN $10,000 and other great prizes from Microsoft Office Live http://clk.atdmt.com/MRT/go/aub0540003042mrt/direct/01/