5
$\begingroup$

My question is related to this post about duplicate questions and I hope it does not seem trivial:

When can I be sure that a question is in fact a duplicate?

I was wondering about this when I flagged a question about p-values (in an application to an output table) for which the same content was dealt with in a more general previous question about p-values. To me it seemed that the "new" question was only new because of this table, so now I am a bit puzzled as to when a question can actually be seen as a duplicate.

Perhaps a function for moderators to comment on why a flag was declined would be useful, too. In this way they could help users to improve their flags which ultimately also helps their work.

$\endgroup$
  • 3
    $\begingroup$ You should note that high reputation users (>3000 rep) review these and cast close votes on duplicates, a lot of what goes on here is down to us, not the moderators. Frequently these things can be a bit of a grey area; I often find myself in two minds. Hopefully there's some wisdom of crowds effect going on. $\endgroup$ – Glen_b Aug 23 '13 at 3:19
9
$\begingroup$

I believe you are referring to Significance: p-value or t-value? and I would have vote to leave it opened (unfortunately, my buggy phone browser led me to just skip it). The suggested near duplicate asks for more general definitions of t- and p-values in the context of null hypothesis statistical testing, which the present post doesn't: In my view, it asks how to gauge 'statistical significance' from a (nice) table where only regression coefficients and t-values are given. The latter accounts for sampling variation and provide all information since sample size are given (as noted by Glen).

I must admit that oftentimes it's hard to find a good duplicate (either because there are many, or because the two questions are just slightly different). In this particular case, I think the OP is concerned with the interpretation of results from two-stage regression summarized in a tabular way. This might be of interest for future readers.

No moderator declined your flag. Don't forget that users community can express their views freely. When moderators do reject an edit or flag, they generally explain why if it's not already available in our FAQ. But, please, please keep assuring the quality of this site by flagging and editing as you see fit.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .