Dear Aleya Aziz Marzuki

I had a similar problem and was told to rescale the DV score, then proceed with a beta-regression.

Best wishes,

Francesco

Yes. The easiest is probably the p.adjust() function will run various corrections using the p adjust (rather than alpha adjustment) approach. This produces the same results as alpha correction. It takes a vector of p values as input and offers a number of methods.

e.g.,

pvals

Article Constructing confidence intervals for selected parameters

Sorry, I missed the later part of the question. The simple answer is that you can't really adjust confidence intervals using step-down or steppe procedures such as the Hochberg, Holm or Hommel procedure because they don't apply a transformation to the estimate of error but to decision process. The simplest alternative is probably to use the Bonferroni or Sidak correction if you need an adjustment (because you can just change alpha in the confint procedure).

Another option is to plot standard errors rather than a CI and indicate significant differences by annotating your graph or table. This allows you to give an indication of the relative precision of the estimates.

Has this article something to with what you want to know?

Summary Plots With Adjusted Error Bars: The superb Framework With an Implementation in R

Denis Cousineau, Marc-André Goulet, and Bradley Harding

Advances in Methods and Practices in Psychological Science  (latest issue; it is an APS journal) 📷

In figures showing data, error bars conveying confidence intervals provide limited information about the precision of estimated results. For instance, confidence intervals do not allow the reader to compare results between groups, between repeated measures, when participants are clustered, and when the population size is finite. Thus, inferences from such error bars can be at odds with conclusions derived from statistical tests. Here, Cousineau and colleagues propose adjusting confidence intervals so that they reflect the experimental design and sampling strategy used. To facilitate the creation of plots with error bars reflecting the adjustments, the researchers developed superb, an open-source library for R.

Hi Peter Prudon. That looks like an interesting article--thanks for pointing it out. But I just did word searches for , , and , and all of them came up empty. So it does not appear to address Aleya Aziz Marzuki's question.

Thanks for finding out. That spares Aleya the trouble. I am not familiar with statistics at this level, but stumbled on the article just after having read her question.