Richard,

Would you be kind enough to elaborate your thinking behind the following paragraph of your reply to Cindy (Earthstars). In particular, when you would and would not use a matched analysis.

>You want that kind of analysis when the pairs are SIBS, for instance, but not (in my opinion) for most other matches. >Instead, you can control for the important covariate by entering it as a covariate, or by using it as a factor in the >ANOVA design.

Thanks,

Gene Maguin