2
$\begingroup$

The question Regression with percentage response variable (ratio of two counts but the counts themselves are not available) in R is very close to a duplicate of What are the issues with using percentage outcome in linear regression?. However, IMO the best (hint to an) answer are @Glen_b comments to the second question. Glen_b also answers the second question, but he doesn't go into some gory details I would be interested in.

In particular, I understand that a logistic GLM is the best choice, but the denominator is not the same among all data points (because students answered to different tests having different total numbers of questions). The suggestion is to condition on the denominator, which would mean making the total numbers of questions an additional predictor, right? How does one deal with a predictor which is not continuous, but it's "more than ordinal"? 20 questions are not only more than 10, they are twice as much, and 3+10 questions is a meaningful operation, etc.

I feel this question is different enough to be asked as a new question. Or should I just wait to see if Glen_b (or someone else) would like to answer the current question? Apologies if this sounds too personalised, but if that's the case, then it's our of esteem and not because I want to pester someone.

$\endgroup$
  • 1
    $\begingroup$ The suggestion is to condition on the denominator, which would mean making the total numbers of questions an additional predictor, right? - no, that's wrong (but not a discussion for Meta). I think this Q is a duplicate but of another thread. Maybe this? stats.stackexchange.com/questions/26762 There are actually lots of threads about this topic. $\endgroup$ – amoeba Feb 15 '17 at 8:55
  • $\begingroup$ Or this stats.stackexchange.com/questions/29038 $\endgroup$ – amoeba Feb 15 '17 at 8:59
  • $\begingroup$ I agree with amoeba, I am sure this has been answered before $\endgroup$ – Glen_b Feb 15 '17 at 9:00
  • 1
    $\begingroup$ I voted to close as a duplicate, and also tidied up and cross-linked some of the mentioned questions. $\endgroup$ – amoeba Feb 15 '17 at 9:31
  • $\begingroup$ @amoeba ok, ok, I wasn't planning on discussion on Meta, I just wanted to give you guys an idea of the question I was considering to ask, so that you could tell me if it was a duplicate or not. Duplicate, got it. And thanks for the other threads. $\endgroup$ – DeltaIV Feb 15 '17 at 12:36
  • 1
    $\begingroup$ Well this led to an interesting discussion with NickCox in the comments over there, so thanks for drawing attention to these questions anyway :-) $\endgroup$ – amoeba Feb 15 '17 at 12:38
  • $\begingroup$ @amoeba you're welcome. It looks like while you were writing this comment I wrote a very similar one below the original question :) $\endgroup$ – DeltaIV Feb 15 '17 at 12:50
  • $\begingroup$ @amoeba, are you voting to close this meta Q? I think this is on topic for meta.CV. $\endgroup$ – gung Feb 17 '17 at 13:14
  • $\begingroup$ @gung No, of course I don't! This Meta Q is on-topic on Meta. I meant that I voted to close stats.stackexchange.com/questions/261790 as a duplicate (and it is closed by now). $\endgroup$ – amoeba Feb 17 '17 at 13:15
  • 1
    $\begingroup$ By the way, @gung, you might have seen my long conversation with Nick Cox in the comments under stats.stackexchange.com/questions/261790. In particular, Nick seems to prefer one approach to dealing with 0-to-1 response variables that is not mentioned in your accepted answer to stats.stackexchange.com/questions/29038 (and by now the 261790 is closed as a duplicate of 29038). I urged Nick to post a new answer to 29038 so that we have a more comprehensive thread there, but he did not reply and did not do it yet. I don't know if you want to extend your answer... $\endgroup$ – amoeba Feb 17 '17 at 13:18
  • 1
    $\begingroup$ @amoeba, it was in the close vote review queue. I voted to leave open. $\endgroup$ – gung Feb 17 '17 at 13:25
  • 1
    $\begingroup$ @amoeba, why was it in the review queue? I don't know. Why leave open? B/c I think it's on topic for meta.CV. $\endgroup$ – gung Feb 17 '17 at 13:38
  • 1
    $\begingroup$ @gung Ah! Sorry! I thought "it" in your comment above referred to the thread on the main site, not to this meta thread. Confusion :-) $\endgroup$ – amoeba Feb 17 '17 at 13:40
  • 1
    $\begingroup$ I flagged to close as off-topic as this question is too localized (perhaps more a chat subject); and OP already got an answer for the specific situation he/she was asking about. Very likely this post will remain unanswered (@amoeba; @gung). $\endgroup$ – Andre Silva Feb 17 '17 at 21:56
  • 3
    $\begingroup$ @AndreSilva I would suggest DeltaIV to self-answer this question and to accept this answer; it's enough to briefly state that this was based on a misunderstanding and the issue is resolved. $\endgroup$ – amoeba Feb 17 '17 at 21:58
3
$\begingroup$

Upon suggestion by amoeba, I self-answer my question: it was based on a misunderstanding which has been clarified.

$\endgroup$

You must log in to answer this question.

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