7
$\begingroup$

As one can see, the functionality of tags has been extended lately; blog post about it. One of the most important changes are the tag-wikis; now every tag can have a small description of a topic it covers.
Because it is pretty hard to gather enough reputation to edit wikis, in this question you can suggest wikis; those upvoted enough will be added to tags by me or other mods.

$\endgroup$
  • $\begingroup$ uhh, you're looking for answers of the form 'tag foo' for some foo? $\endgroup$ – shabbychef Aug 30 '10 at 23:11
  • 1
    $\begingroup$ Rather a complete wiki text candidate, like "foo is something like bar, but also something completely different; nevertheless it is widely used..." $\endgroup$ – user88 Aug 31 '10 at 7:25
  • $\begingroup$ OK, I don't get it, then. why don't we just start writing the tag wikis? if anyone can edit them, then won't they converge to what the community wants? what am I missing--why have this intermediate step? $\endgroup$ – shabbychef Sep 1 '10 at 18:04
  • 1
    $\begingroup$ They can only be edited by users with 2000+ reputation, or by the top 10 answerers for the tag. At this point, we've only got three users who meet the first criteria, and there hasn't been enough time for the people who are really interested in a tag to rise to the top of its answerers (for some of the more rare tags). This way, the mods don't have to write everything but we still have the tag wikis filled in from early on. $\endgroup$ – Matt Parker Sep 2 '10 at 13:41
  • 2
    $\begingroup$ gotcha. so the point here is to make a mild end-run around the restrictions on who may edit tag wiki. Can we just lower the bar on editing tag wikis, at least temporarily? 2K+ reputation seems rather high given how young the site is. $\endgroup$ – shabbychef Sep 2 '10 at 19:20
  • $\begingroup$ @shabbychef This way is easier; limits are already reduced to what is needed on SO, so SE team will say it is ok. Also this site is somewhat different about the ease of judging is something is true. $\endgroup$ – user88 Sep 3 '10 at 6:32
4
$\begingroup$

mixed-models:

"Mixed models" refers to a class of models that are variously known as: mixed-effects models, multilevel models, hierarchical linear models, [these things have about three different names per discipline, feel free to add your favorite]. This class of models was developed to account for correlation that may occur within nested data. A classic example is the estimation of test scores of students: if test scores are correlated within classes, schools, districts, etc., mixed models allow the modeler to simultaneously estimate the differences between individual students and between the groups to which they belong (with the possibility of including covariates at all levels).

StatsExchangers often recommend the following resources for learning more about mixed models: Venables and Ripley (2002), Laird and Ware (1982), Baayen (2008), Raudenbush and Bryk (2001), Gelman and Hill (2006), and Singer and Willett (2003).

Mixed models are available in the following statistical packages: R (lme4 and nlme), SAS (PROC MIXED), MLwiN, Stata (xtreg and xtmixed), and HLM.


Please edit to improve - I just wanted to get the ball rolling on this one. Thought I would throw out the book and package recommendations as those are likely to be asked often, but I'm not sure how we could come to a consensus about what books and packages should go there. However, maybe the upkeep of those lists could be left to whomever has the rep and the interest, with disputes hashed out here on meta.

$\endgroup$
  • 1
    $\begingroup$ Added in the Stata modules. Also I have found the MlwiN online tutorials very helpful, which can be found here cmm.bristol.ac.uk/learning-training/index.shtml . Also should the free Goldstein book be included in the suggested reading materials? Although it is not as beginner friendly as Gelman's or Bryk and Raudenbush, google scholar does say it is cited by over 4,000 papers. Here is one link to it but their may be more appropriate ones, cmm.bristol.ac.uk/team/HG_Personal/multbook1995.pdf $\endgroup$ – Andy W Sep 3 '10 at 3:13
3
$\begingroup$

SPSS (Statistical Package for the Social Sciences) is a proprietary cross-platform general-purpose statistical software package. SPSS's homepage

Versions 17 and 18 of SPSS were renamed with the acronym PASW (for Predictive Analytics SoftWare), but with version 19 it has been renamed to SPSS (see Wikipedia for a history of the name changes). One unique aspect of SPSS software compared to other popular propriety software packages (such as Stata or SAS) is the built in functionality to call Python or R commands within syntax. Otherwise it is largely comparable to other general proprietary and freeware packages (such as R), although it differs in some advanced statistical capabilities and aspects of data manipulation.

Suggested readings on using SPSS and learning the command syntax are two online PDF's.

Other useful print versions are

Other forums entirely devoted to the software are (and suggested material to search when encountering a problem with SPSS are):

Other suggested webpages are

PSPP is a free-ware, open source alternative largely mimicking the look and functionality of SPSS.

$\endgroup$
  • $\begingroup$ Also as a note I entirely stole the opening line written by onestop for the Stata tagwiki. $\endgroup$ – Andy W Jan 10 '11 at 16:51
  • $\begingroup$ Now set as [spss] tag wiki. $\endgroup$ – user88 Jan 10 '11 at 17:46
  • $\begingroup$ In fact, it was renamed PASW before the company was bought by IBM, and it's now reverted back to SPSS! See first para of en.wikipedia.org/wiki/SPSS. $\endgroup$ – onestop Jan 17 '11 at 23:40
  • $\begingroup$ @onestop , do you mind editing the tag wiki directly? I don't have enough reputation to do it myself. I was considering putting in a quip about them updating versions far too frequently! $\endgroup$ – Andy W Jan 18 '11 at 1:03
  • $\begingroup$ @Andy W: Will do. I didn't realise I could edit it either, but it appears I can. (I just can't create new tag wikis now we're out of beta as I haven't earned any tag badges). $\endgroup$ – onestop Jan 18 '11 at 10:06
  • $\begingroup$ ... or rather I'll try to, but at present I'm getting a site error message when I click 'about the spss tag'. I'll try again later. $\endgroup$ – onestop Jan 18 '11 at 10:09
  • $\begingroup$ It's working better today, so I've replaced the tag wiki text with your edited answer above. $\endgroup$ – onestop Jan 19 '11 at 11:54

You must log in to answer this question.