Wikipedia:Articles for deletion/Multifactor design of experiments software

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was keep. Editors are encouraged to discuss further improvements to the quality of the article itself, at the article's talk page. Additionally, possible merge discussions may take place as well, at the article's talk page. -- Cirt (talk) 05:58, 21 September 2010 (UTC)[reply]

Multifactor design of experiments software[edit]

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

Seems like a neologism, though it's unclear as it's not explained what exactly the topic is, with POV and editoralising in place of proper encyclopaedic content. Not clear from the references either that there's a topic here. JohnBlackburne^words_deeds 09:56, 13 September 2010 (UTC)[reply]

Delete and consider starting again. I'm sure we could write a decent article on algorithms for experimental design, including their software implementations. But this is not it. Most of the article simply recaps the basic ideas of experimental design, with just three or four sentences mentioning software at the end, and they are mostly puffery. --Avenue (talk) 14:45, 13 September 2010 (UTC)[reply]

Clarification: software for experimental design is a reasonable topic (although I think algorithms is better than just software), and I would reconsider my !vote if someone goes to the trouble of tracking down a suitable source (not just vendor's web sites) and works it into the article. There are some useful leads at Optimal_design#Finding_optimal_designs, for instance. But I think Wikipedia is better off without the current version, which fails WP:NOTADVOCATE. ~~There is nothing here even worth merging back to our Design of experiments article, hence my !vote to delete, not merge.~~ I also think we would be better off expanding the software/algorithmic coverage of our existing articles on Design of experiments, Response surface methodology, and Optimal design (which is minimal at present) before starting a whole new article. --Avenue (talk) 21:57, 13 September 2010 (UTC)[reply]

The article is still a bit of a mess, but it has improved enough that I think deletion is no longer appropriate, and I've struck my delete !vote. I'll wait and see how it progresses before deciding whether I'd prefer it to be kept or merged. --Avenue (talk) 12:04, 14 September 2010 (UTC)[reply]

Merge into design of experiments. The only citation that addresses software issues, as opposed to DOE in general, is to http://www.statease.com/software.html. The article still reads like a puff piece, and I don't have much hope this will improve if it's kept as a standalone article. If it's merged, the material should be seen by more knowledgeable editors and probably be put into better perspective. I would have no objection to the article being recreated later if there is a consensus at Talk:design of experiments that enough material has been gathered for a separate article. --Avenue (talk) 15:19, 19 September 2010 (UTC)[reply]

Comment I don't understand the "neologism" comment. "Multifactor design of experiments" is obviously not a neologism. Can "software for X" be considered a "neologism" if X is some standard thing, and software can be applied to doing X? The problem with the article in its present form is that it doesn't say much about the software. Presumably that could be remedied by adding the information and otherwise adapting the article to Wikipedia's usages. Michael Hardy (talk) 15:52, 13 September 2010 (UTC)[reply]
comment A neologism is a made up word or phrase and as best I can tell "Multifactor design of experiments software" is made up as a topic. A search returns nothing but WP and mirrors, and the ref or external links seem to have nothing on it, so as a topic it gets no mention anywhere and is therefore completely un-notable.--JohnBlackburne^words_deeds 16:23, 13 September 2010 (UTC)[reply]
- Your search was done wrong. You searched for "Multifactor design of experiments software". You should have searched for "Multifactor design of experiments". The article seems intended to be an account of software for doing multifactor design of experiments (otherwise sometimes called "factorial designs" or "factorial experiments"). Michael Hardy (talk) 03:31, 14 September 2010 (UTC)[reply]
Keep There is a misunderstanding here. "Multifactor design of experiments software" is not a neologism, it is a description (and not a particularly good one) of a topic of statistical interest. This is made clear when the article starts "Software for the design of multifactor experiments". Here[1] would be a better search and there are other appropriate wordings too. I am not well qualified to judge the article which I find a bit thin when it gets to the point and more references would be helpful. However, as a topic, it certainly deserves a place. Thincat (talk) 17:27, 13 September 2010 (UTC)[reply]
Speedy Keep It is not a neologism. Design of experiments is a statistical method, which is used amongst others in Six Sigma and is detailed here: Design of Expriments. DOE comes in two types, Single DOE and Multi DOE. So software that create mulitfactor DOE statistical models is perfectly acceptable. scope_creep (talk) 18:53, 13 September 2010 (UTC)[reply]
Comment The article needs expanded and wikified.scope_creep (talk) 18:56, 13 September 2010 (UTC)[reply]
Keep: The article is in bad shape, admittedly, but the arguments don't stick. It does seem that the article is more about multifactor experiment design than the software used in such -- but in that case the article needs a new name, not deletion. CRGreathouse (t | c) 18:08, 13 September 2010 (UTC)[reply]

Since almost all non-trivial experimental designs are multifactor, I think our Design of experiments article already fills that niche. --Avenue (talk) 21:57, 13 September 2010 (UTC)[reply]

comment we already have an article on factorial experiment which is a redirect from factorial design. Looking at the sources "factorial design" is the term used in this, this and this (the first sample from this EL), so it seems an article on the design of such experiments already exists.--JohnBlackburne^words_deeds 18:29, 13 September 2010 (UTC)[reply]

This is out of my area. Are these the same topic? I've read the article you refer to, but it wasn't clear to me (especially due to the stubbiness of the current article) whether they were distinct; I thought likely so. If not, I may change my ~vote. CRGreathouse (t | c) 18:55, 13 September 2010 (UTC)[reply]

It's out of my area too but I think so: the links above are all from the article, i.e. they are its sources (though they're in External links), and all use "factorial design" none use "multifactor design". Some also use fractional factorial design and we've and article on that too. And both this and factorial experiment say the theory is the work of Ronald Fisher.--JohnBlackburne^words_deeds 19:05, 13 September 2010 (UTC)[reply]

Multifactor experiments also include Response surface designs. --Avenue (talk) 21:57, 13 September 2010 (UTC)[reply]

comment The following makes me think even more that this topic is an inadvertent fork of Factorial experiments:

"Factorial experiments Use of factorial experiments instead of the one-factor-at-a-time method." (from Design of experiments#Principles of experimental design, following Ronald A. Fisher)

"Before Fisher's multi-factor DOE breakthrough, the common experimentation method was conducted using OFAT (one-factor-at-a-time)" (from this article)

Again this is not my area of experience, but everything I've read now including the links in this article suggests that "factorial" is the word commonly used instead of "multifactor".--JohnBlackburne^words_deeds 22:09, 13 September 2010 (UTC)[reply]

Generally they are called factorial experiments, but "multifactor experiments" is a self-explanatory alternative. I think the article should be kept if it can be made into an account of the software. But last time I looked at it (some hours ago), it wasn't anything like that. I'll look again tomorrow..... Michael Hardy (talk) 03:36, 14 September 2010 (UTC)[reply]

It seems the main contributor to the article has clarified this by updating the article: it does seem now to be on factorial experiments which we have an article on. Half of it is a section on software but it says nothing about it the software as a category, and gives no indication why the topic "factorial experiment software" is notable; in general just because a topic XXX is notable does not mean "XXX software" is, and we have very few such articles.--JohnBlackburne^words_deeds 09:22, 14 September 2010 (UTC)[reply]

Keep: while there is presently the problem of too much overlap with other articles, the article is still evolving quite rapidly, and does have a stub tag on it. It would be good if helpful comments can be made on the article's talk page, recalling that the main contributor to the article has only recently started here. Melcombe (talk) 08:36, 14 September 2010 (UTC)[reply]
Note: This debate has been included in the list of Software-related deletion discussions. -- • Gene93k (talk) 19:51, 14 September 2010 (UTC)[reply]
Comment the article is heavily weighted towards one product StatEase's DesignExpert, I've removed quite a bit of that material, which might be more appropriate in an article on the company or product. With that removes I'm not sure that there is enough material for an article which might work better as a section in another article.--Salix (talk): 09:28, 15 September 2010 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.