Reporting Beta + unstandardized CIs?

#1
Hello,

I did a multiple regression analysis and for now always only reported the standardized effect sized (Beta). Since I heard that it would be good practice to also include the CIs, I wonder how this is done in practice - should I just present the standardized effect size next to the UNstandardized CIs? (I heard that standardized CIs do not really make a lot of sense?). On the one hand I think that may look strange to some people, on the other hand I think that only additional information is given, and no contradicition is introduced. But I am not sure how people interpret it.

I searched the Internet vigorously, and while I did find plenty of information on how to report effect sizes and confidence intervals, I did find no notion of my exact problem. Neither did I find any paper that reports both (reporting CI is really not common in my field, but I would like to counter this trend - as most guidelines for the authors actually proclaim that they want to see them).

I appreciate any help or hints!

Thank you!
 
Last edited:
#2
It is possible and more than useful and more important than them necessitated by various modern report guidelines to state CIs too. I think reporting both B (non-standard coefficient) and beta, B's standard error, and its CI range altogether is very good. Some of these can be re-calculated from the others, but I prefer a rich report.

Neither did I find any paper that reports both (reporting CI is really not common in my field, but I would like to counter this trend - as most guidelines for the authors actually proclaim that they want to see them).
Same as in my field. However, I personally always report all the above items.
 
#3
Hello,

thanks for your input! Reporting both Beta and B + CI might indeed be the option that causes least confusion. I just thought I could maybe skip the "B-part", in order to not flood my results section with too many numbers (the "normal" way of just putting them into a table is not really applicable in my case, as the table would be really too large than).

Thank you again!