# What should our FAQ contain?

In line with the link to the 7 Essential Meta Questions of Every Beta above. What should our FAQ contain?

• A suggestion: I think each answer should be phrased as a potential faq that is proposed to be included along with a potential answer. Please also make the answers CW so that all of us can improve the question and the corresponding answer. The votes will then determine what the community consensus is reg the faq content.
Jul 22 '10 at 1:46
• Could you please make this CW so that we can start the process of re-writing the questions as proper FAQ entries? Thanks
Jul 29 '10 at 15:13
• This needs to be CW. Amos could do this but he seems to be absent. However, each answer can be separately made CW by the original answerer. I have done that for my answer. Can others please do it for theirs? Then we can start editing these. (Those with >1K rep can edit the answers anyway.) Aug 3 '10 at 1:14
• I've now converted this to CW. Please edit the various answers so we can construct the FAQ together. Aug 5 '10 at 0:47

## Mathematics support

You can use LaTeX expression in questions and answers. Simply put the expression within dollar signs. For example, $\sqrt{\alpha+\beta}$ produces $\sqrt{\alpha+\beta}$.

Some R code contains dollar signs and to prevent these being interpreted as LaTeX expressions, please enclose all code in single left quotes like this: fit$x <- model$y.

How to include references to published papers and books.

Suggested wording:

## How to include citations

1. When citing scholarly papers, please use the author-date system. For example, "Cox (1992)". It is not necessary to provide a complete reference provided you link to a suitable site that contains the details.

• Otherwise link to a public repository such as JSTOR, arXiv, RePEc or SSRN.
• Otherwise link to the publisher's page or some other source that provides a full citation and (if available) a link to the PDF.

2. When citing books, please use the author-date system and provide a link to the book on amazon.com or Google books.

That the scope contains Statistics, Machine learning, Data mining and Visualization, as goes form this discussion: Is machine learning a part of statistical analysis?

Suggested wording:

## What kinds of questions can I ask here?

This site is for statisticians, data miners, and anyone else doing data analysis. If you have a question about

• statistical analysis
• data mining and machine learning
• data visualization
• statistical and data-driven computing (e.g., questions about R, SAS, SPSS, Stata and Minitab)

## What kind of questions should I not ask here?

Avoid asking questions that are subjective, argumentative, or require extended discussion. This is not a discussion board, this is a place for questions that can be answered!

If you want to talk about the site itself, please don't do it here. Visit our meta-discussion site where you can talk about things like what questions are appropriate, what tags we should use, suggest a feature, or generally discuss how stats.stackexchange.com works.

• @Rob, thanks for extension. Looks pretty good.
– user88
Aug 2 '10 at 15:20
• It should possibly differentiate mathematics from research-level mathematics on MathOverflow. I'm not sure if you want to reach out of the Stack Exchange 2.0 community, though. I think I read somewhere where Jeff (or another developer) said that eventually, SE2.0 sites can move questions between each other. Aug 12 '10 at 23:44

Something about "community wiki". I know what a wiki is, but I have no idea why the phrase "community wiki" appears on this site, and when I should check that option.... Is that a separate site? More editing allowed in this site? Confusing.

## Community wiki questions

Community wiki questions and answers can be edited by any user with at least 100 reputation points. These are designed for the community to pool information about a topic. Usually, a community wiki question will be subjective and open-ended. For example, "What is the best introductory statistics textbook?" Also, please note that votes for any posts marked "community wiki" do not generate reputation.

• It is strange that the existing FAQ doesn't explain that clearly. Aug 3 '10 at 1:17

An explanation of whatever answer we come up with in How to answer R questions and the more general How much programming here.

• This is now part of the scope statement covered in another answer. Aug 6 '10 at 7:53

What statistics resources provide basic instruction? / What resources should I review before I ask a basic question?

To help preemptively direct people with basic questions to review other available resources, maybe we should have a FAQ that provides a list of such resources. Or maybe the answer can just link to an ongoing CW resource list.

Whatever policy we end up adopting with respect to homework questions.

If it is decided that the scope of Statistical Analysis includes helping hard working students understand their homework then we should outline how to ask good homework questions in the FAQ.

Following this question on meta.math.SX and in particular the excellent post by Pete L. Clark that inspired it, perhaps some equivalent question-optimisation hints should be in the FAQ here? Possibly as some sort of hybrid with or counterpart to the existing SX boilerplate What if I don't get a good answer?

For example, the question might be along the lines of:

## How should I ask my question to increase my chances of getting good answers?

And the answer would probably hit at least the following points:

• Use as descriptive a title as possible
• Use appropriate tags
• Be specific: what do you want to know?
• Put the question in context: why do you want to know?
• Be clear: explain acronyms, define any non-standard jargon, link to external sources.
• Show that you care: what have you already done?
• ...

(I can expand these into something less bullety if this seems like a worthwhile question, but everyone else should feel free to beat me to it!)

It should contain info about moderators. Such as who they are, what they do, and where they're based.

Following mbq's comment, I would suggest just a list of the current (and past?) moderators. As mbq suggested, further information would be found in the user's page.

• I don't think it goes along with the not-we-but-the-community philosophy. On the other hand this is for what user profiles are.
– user88
Jul 28 '10 at 9:34

This is a proposed entry to the faq regarding community norms. The text in blockquotes is meant to be what we will include in the faq. Feel free to edit as necessary.

The first entry is discussed here: Should we encourage downvoters to leave a comment for their downvote?

What are the community norms?

The following are some norms that we as a community will follow. There will be occasions when we will not be able to follow the norms because of various reasons but to the extent possible we will try to follow thenm.

These norms are not cast in stone. If you wish to discuss any of these norms please open a meta thread at http://meta.stats.stackexchange.com.

1. When we downvote a question we will leave a comment explaining to the OP the reasons for the downvote.

Leaving a comment when downvoting achieves several objectives.

• It increases the chances that that the OP does not leave the site out of frustration.

• It increases the chances that the question actually gets edited and improved.

• Keeps the quality of questions high.

What are some good books / videos / software and resources to learn more about statistics / data mining / machine learning?

Consider browsing the questions and answers under the tags below. There are a number of useful resources and recommended material gathered therein.

For example:

• I am not sure if we should have this under faq. Isn't the faq about how to use the site rather than a faq about statistics per se?
Aug 23 '10 at 10:51
• @Srikant: Yeah, that occurred to me after. I just did it on impulse because I think we're going to see this kind of question frequently. Maybe just move to this to a CW question on the site? Or I can just delete it.
– ars
Aug 23 '10 at 14:53
• I would prefer you do not delete it as you have taken the effort to collect the useful links here. I suggest that we keep it here and if we feel that the volume of qns on the main site about this topic go up then we can shift this as a CW qn to the main site. We can then close all such as qns as dupes of this one. This would also serve as a useful resource to close qns as exact duplicates of the others that have been asked already.
Aug 23 '10 at 15:14

The FAQ should explain how this site works. What does it mean to vote up or down a questions or an answer? Why should we do so? When it is possible to edit someone else's answer? When does it make sense to comment on a prior answer vs. write a new one? All this basic stuff really should be explained. Some is obvious, some not so obvious to some. This stuff must be written up somewhere for StackOverflow.

• This is already in the FAQ. I'm assuming that we start with the existing FAQ and add to it. Jul 30 '10 at 4:14
• The FAQ seems to have been updated since I wrote this, and it does now explain voting up and down answers. But it doesn't explain what it means to vote up or down questions. The tool tip says that voting up means the question is "useful and clear", but some elaboration would be useful. Browsing over the site, it doesn't seem that people are voting on questions frequently or consistently. Aug 3 '10 at 15:39