15
$\begingroup$

There are some questions on site which are also about professional ethics. Some examples: Ethics of publishing covariate matrix, Likert Scale - Treatment of Neutral or Mid-Point of 5-Pt Scale (and others)

Should we make a new tag ethics for this, or is there already some tag which includes this? I thought that could be used, maybe. Thoughts?

$\endgroup$
6
  • 9
    $\begingroup$ To me careers does not hint at ethics, so a new tag sounds fine to me. $\endgroup$
    – Nick Cox
    Sep 23, 2019 at 8:18
  • 5
    $\begingroup$ I have conflicting feelings about this, so I'm writing a comment and not an answer. On the one hand, I think that modern statistical methods can be extremely dangerous if they are unmoored from a sense of ethics and social responsibility, and I think that ethics in statistics should be a core part of a statistical education. On the other hand, I don't know how well-suited stats.SE is to provide ethical advice; a question mainly about ethics in statistics would fall outside of how I understand on-topic questions. $\endgroup$
    – Sycorax Mod
    Sep 24, 2019 at 3:13
  • 5
    $\begingroup$ Ethics can have very different specific meanings in statistics. This includes topics such as medical ethics, in which we have a well defined goal of minimizing harm to test subjects, influence of social biases into ML models and p-hacking are three totally different and largely unrelated statistical ethics topics. $\endgroup$
    – Cliff AB
    Sep 24, 2019 at 3:27
  • 5
    $\begingroup$ How to preserve confidentiality in handling data and results is one of several areas in which an ethical imperative is clear and the discussion is all about how to do it. For such questions a tag would make sense and a thread need not get mired in moral philosophy. $\endgroup$
    – Nick Cox
    Sep 25, 2019 at 7:26
  • 1
    $\begingroup$ In my opinion, a lot of posts that would get the "ethics" tag should be posted in academia.stackexchange.com or workplace.stackexchange.com. $\endgroup$
    – TrynnaDoStat
    Sep 26, 2019 at 18:53
  • $\begingroup$ You can add a tag, but the real questions to me are will it take off in terms of people using it and people searching for it? Naturally the same applies to any tag. My small principle is that if a tag does harm. people will notice and raise an issue. As the most obvious person for tag editing is yourself, Kjetil, I would say try it. $\endgroup$
    – Nick Cox
    Sep 30, 2019 at 15:38

2 Answers 2

Reset to default
5
$\begingroup$

I was just going to ask the exact same question and found that I already upvoted this question here, because great minds think alike.

The question currently has 12 upvotes and 0 downvotes, I would take this as a clear vote that we should indeed have such a tag.

I have created a tag wiki based on the comments above; everyone please take a look.

I have also started retagging questions, everyone is invited to contribute here. (Let's retag slowly, no more than a few questions every day, so as not to bump-flood the homepage. No need to retag closed questions, unless they can be improved to be candidates for reopening.)

$\endgroup$
2
$\begingroup$

I would suggest to formulate the tag information and the tag wiki in a similar way as we use tags for software. E.g. the R-tag:

https://stats.stackexchange.com/questions/tagged/r

Use this tag for any on-topic question that (a) involves R either as a critical part of the question or expected answer, & (b) is not just about how to use R.

And in the help section

  • Programming, ask on Stack Overflow. If the language is statistically oriented (such as R, SAS, Stata, SPSS, etc.), then decide based on the nature of your question: if it needs statistical expertise to understand or answer, ask it here; if it's about the implementation of an algorithm, routine data processing, or details of the language, then please refer to the collection of links to resources we maintain.

For ethics we could have an analogues approach. The questions should require statistical expertise and require statistics to answer, rather than being just about a more general ethical question that is related to statistics but not neccesarily about statistics.

An example question of such category could be:

Was Amazon's AI tool, more than human recruiters, biased against women? which is not a question purely about the ethics (e.g. whether or not bias is good) but about how statistics can lead to the ethical problem.

$\endgroup$
1
  • 1
    $\begingroup$ +1 for raising the issue. On the other hand, maybe this is something that doesn't need clarification for the moment. But if clarification is needed, adding a couple of examples of off-topic questions may be useful for people who ask questions. $\endgroup$
    – J-J-J
    Nov 22 at 10:03

You must log in to answer this question.

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