User talk:Likelihoodist

Hello DragonflySixtyseven:

Yesterday I received an email saying that "User:Likelihoodist was reviewed by DragonflySixtyseven." but I could not find any input from you.

I would be most grateful if you could give me some feedback on a page that I recently submitted to Wikipedia. This page, PS Power and Sample Size was rejected by Hasteur on October 1, 2013. He/she wrote "Cluster of 9 references all together does not inspire confidence. Fix it." I left him the message given immediately above this one asking him for additional guidance and summarizing the evidence concerning the notability of this page. Unfortunately, he has not responded to date. Would you be willing to look briefly at this submission, and my response to Hasteur? Power calculations are of great interest to an admittedly specialized audience. However, Wikipedia does publish a great deal on information on topics that are only of critical interest to specialized experts. I would be deeply indebted to you if you could provide me with some insight into the degree of notability required of such pages for publication in your encyclopedia.

Sincerely

Likelihoodist (talk) 22:45, 9 October 2013 (UTC)

I patrolled your page. I went through the enormously-backlogged list of newly-created pages and confirmed that your page was okay: not spam, not an attack page, not a copyright violation, not any of the other reasons for which I would delete someone's page without asking. Then I clicked "patrolled" to remove it from the list of "pages that have not yet been patrolled", and moved on to the next entry. That's all. DS (talk) 23:08, 9 October 2013 (UTC)

Hello Hasteur:

I have edited and resubmitted the PS Power and Sample Size in response to your directions "Cluster of 9 references all together does not inspire confidence. Fix it." I would be most grateful if you would let me know whether these edits satisfactorily address your concerns.

Best wishes,

Likelihoodist (talk) 23:33, 22 November 2013 (UTC)

PS Power and Sample Size, which you submitted to Articles for creation, has been created.

You are more than welcome to continue making quality contributions to Wikipedia. Note that because you are a logged-in user, you can create articles yourself, and don't have to post a request. However, you may continue submitting work to Articles for Creation if you prefer.

• If you have any questions, you are welcome to ask at the help desk.
• If you would like to help us improve this process, please consider leaving us some feedback.

Thank you for helping improve Wikipedia!

—Tom Morris (talk) 17:21, 27 December 2013 (UTC)

Just having a look at the above volley of discussions: I'm sorry that you had a harsh time of it submitting your article via articles for creation—or "the crap pile" as I informally refer to it. There's a lot of really terrible articles that gets submitted via AfC and so editors often aren't wild about ploughing through it. The backlog is enormous and so reasonable articles can get lost amongst the never ending parade of junk. Thanks again for writing the article and sorry if the process was rather slow and dispiriting. —Tom Morris (talk) 17:29, 27 December 2013 (UTC)

Hello Tom:

Many thanks. The delay was largely my fault. I initially interpreted Hasteur's rejection sentence "Cluster of 9 references all together does not inspire confidence. Fix it." as meaning that he was not confident that my references documented sufficient notability for the PS program. It was unclear to me how I could fix this so I put my page aside and went on to other labors. It was only months later when I reread his sentence that I realized that he might be only asking for a minor edit related to my having cited nine references at one time. This was trivial to fix, but by this time, Hasteur was working on other pages. He, like yourself and the thousands of other Wikipedia editors donate extraordinary amounts of time to Wikipedia and it is not surprising that his response was terse. But I guess what he intended as a conditional acceptance, I interpreted as a flat rejection.

Currently, Wikipedia has a page on statistical power and a page on sample size calculations. Neither of these pages say anything about software for making these calculations. The PS and PASS pages describe such software but are orphans. Would it make sense for me to add a new page on power and sample size calculation software? This page would link to the PASS and PS pages and would also briefly describe the other most popular commercial and free programs that are available. The sample size and power calculation pages would both be linked to this new software page. Do you think that such an addition might be well received by the Wikipedia editors? Would you recommend that I just go ahead and do this or should I submit same as another contribution for review by the Wikipedia editors?

With best wishes for 2014

Bill Dupont Likelihoodist (talk) 22:13, 27 December 2013 (UTC)

I'd recommend the best solution might be to add a small section to the articles in question pointing to the software, rather than creating an intermediate page. I'd recommend you just go ahead and edit the pages directly (Wikipedia encourages boldness). If you want me to have a look at your edits after you've done them, just drop me a note on my talk page and I'll have a look (although I'm not an expert on mathematics or statistics, so I can't give feedback on that side of things). I'm away for a few days, but am certainly happy to have a look on my return.

Thanks again for your contribution. —Tom Morris (talk) 23:13, 27 December 2013 (UTC)

Cool! Thanks. I would certainly appreciate your edits. I'm OK with the math and stats, but the Wikipedia culture is still new to me. (It may be a week or two before I can find the time to do this.)

Likelihoodist (talk) 23:28, 27 December 2013 (UTC)

Your recent article submission to Articles for Creation has been reviewed. Unfortunately, it has not been accepted at this time. The reason left by WikiOriginal-9 was:

This draft's references do not show that the subject qualifies for a Wikipedia article. In summary, the draft needs multiple published sources that are:

• in-depth (not just passing mentions about the subject)
• reliable
• secondary
• independent of the subject
  

Make sure you add references that meet these criteria before resubmitting. Learn about mistakes to avoid when addressing this issue. If no additional references exist, the subject is not suitable for Wikipedia.

 The comment the reviewer left was:

Not enough significant, independent coverage.

Please check the submission for any additional comments left by the reviewer. You are encouraged to edit the submission to address the issues raised and resubmit after they have been resolved.

• If you would like to continue working on the submission, go to Draft:Odds ratio for a matched case-control study and click on the "Edit" tab at the top of the window.
• If you do not edit your draft in the next 6 months, it will be considered abandoned and may be deleted.
• If you need any assistance, or have experienced any untoward behavior associated with this submission, you can ask for help at the Articles for creation help desk, on the reviewer's talk page or use Wikipedia's real-time chat help from experienced editors.

WikiOriginal-9 (talk) 22:08, 5 November 2023 (UTC)

I would be most grateful if you could clarify the reasons for your rejecting my article as well as suggestions as to how my entry can be improved. Please see the questions written below.

Your stated reasons in your original rejections are as follows:

1. Does not qualify for a Wikipedia article.

Odds ratios for matched case-control studies is a standard topic in epidemiologic statistics. As such, I would think that this would make my article notable. This topic is covered in many textbooks on epidemiology. This includes Gordis Epidemiology (ref 1 on my submitted page) which is a standard text that is widely used for teaching elementary epidemiology. Perhaps I should also have referenced Rothman et al.’s Modern Epidemiology (See reference 15 in the Wikipedia article entitled “Odds ratio”). Rothman is a renowned American epidemiologist (see the Wikipedia article entitled “Kenneth Rothman (epidemiologist)” Their text, which covers this topic, is arguably the most authoritative text on advanced epidemiology available today.

Wikipedia does have an article on Odds ratios, which covers odds ratios for independent case-control studies but does not mention odds ratios for matched studies. It also has an article entitled “McNemar’s test”, which describes a test of the association between two dichotomous variables in a matched study. This test is also used to test the hypothesis that the odds ratio from a matched case-control study equals one. However, this page does not mention odds ratios or give the maximum likelihood estimate for this statistic from these studies. The lack of any entry on odds ratios for matched case-control studies is a notable omission from Wikipedia that I believe should be filled.

Should I cite Rothman et al.? Do I need to reference additional text books on this topic? If so, how many? Do I need to reference additional peer-reviewed papers that use this method? If so, how many? Is there some other type of reference that I should add.

2. In-depth entry

Celentano et al. (ref 1 on my submitted page), Rothman et al. and Breslow and Day all cover this topic in detail. Celentano et al. discuss calculating odds ratios from a matched-pairs case-control study on pages 251–253 and 290 – 291. They do not give a proof as to why this odds ratio estimate is correct. Rothman et al. cover this topic on pages 287 – 288 and reference Breslow and Day 1980 (ref 2 on my submitted page). Breslow and Day provide the derivation of the odds ratio from matched 2x2 tables that is given in my article. I don’t see how this topic can be covered in greater depth by these authors.

Should I clarify that the derivation that I give in my article is due to Breslow and Day and not a proof that I thought of myself?

3. Reliable sources

The references that are given in my article are authoritative. As mentioned in his Wikipedia web page https://en.wikipedia.org/wiki/Kenneth_Rothman_(epidemiologist) , Rothman is a professor of epidemiology at Boston University and a distinguished Fellow at RTI International. The fact that his text is cited by Wikipedia’s “Odds ratio” page speaks to the reliability of his textbook. (His coauthor Sander Greenland is also one of the world’s leading epidemiologists. See the Wikipedia page entitled “Sander Greenland”. N.E. Breslow and N.E. Day were/are renowned 20th century biostatisticians. See their Wikipedia pages entitled “Norman Breslow” and “Nick Day (statistician)”. Leon Gordis was a professor of epidemiology at the Johns Hopkins Bloomberg School of Public Health. In short, the references on odds ratios from matched case-control studies could not be more reliable.

Do you have any concerns as to whether these authors are reliable?

4. Secondary source

I’m not sure how this criteria applies to a statistical methods page. I do reference the paper by McEvoy et al. that provides an interesting application of this method.

What type of additional references would be considered an appropriate secondary source?

5. Independent sources

All of my references are independent from me. My name is Bill Dupont (see https://www.vumc.org/biostatistics/person/william-d-dupont ). I am not a personal friend of any of the authors cited in my article or given above. Also, I am not a co-author of any paper written with these scholars.

In summary I am puzzled as to why you rejected my article. It appears to meet the criteria for publication in Wikipedia and would be a worthwhile contribution to the pages that have already been published on epidemiologic methods. I would be most grateful for any advice that you can give me as to how to improve my article to make it suitable for publication in Wikipedia.

Thank you for reading the above an considering the questions that I have asked. Likelihoodist (talk) 03:18, 10 November 2023 (UTC)

Hello Likelihoodist -- I don't have the expertise to assess the notability of this topic, but is there a reason you are using the Articles for Creation service? It's optional for editors with a moderate amount of experience who are not directly connected to the topic. The reviewers are generalists and it's extremely unlikely you will find anyone who can assist with this kind of technical material. ETA To address your questions, generally two or three reliable independent sources with significant coverage is ok. Academic books, university-level textbooks, review-type journal articles are all ideal, as long as they are not written by the originator of the concept. Regards, Espresso Addict (talk) 12:18, 10 November 2023 (UTC)

Thank you for your reply. There is no particular reason why I am using the Articles for Creation service. I thought that I was just submitting a conventional Wikipedia article in the appropriate way. Your guidance as to which service I should use (and how to use it) would be greatly appreciated.

It would be really helpful if I could have my article reviewed by a someone who is familiar with epidemiologic statistics. (I do not know if my current reviewer has this expertise.) Is there some other way that I should submit my article? Is there any way that I can request a reviewer who is familiar with the topic that I am discussing? Likelihoodist (talk) 16:48, 10 November 2023 (UTC)

I'm sorry that the system is working so poorly in this case. Articles for Creation, in my opinion, is not good at handling technical topics; reviewers get offered a random article to process rather than articles selected by topic. You could request a reviewer from a relevant Wikiproject, but beware a lot of them are now inactive. WP:Wikiproject Medicine is active but I'm less familiar with the mathematics projects; WP:WikiProject Statistics looks inactive but WP:WikiProject Mathematics is active. Just drop a note on their talk page and briefly explain the problems you have been having (editors are always overcommitted; longer comments tend to get less attention).

As an autoconfirmed editor, there's nothing stopping you from creating articles in mainspace or moving articles into mainspace yourself, and generally I recommend this for any specialised or technical topic, although the new-page patrollers are also non-specialists and can be poor at handling technical topics. In the longer term, once you are more familiar with the encyclopedia's processes, perhaps you would consider reviewing relevant drafts yourself? (The editor who was doing most of this work, David Goodman, died earlier this year.)

I hope this is helpful, and again apologies that you are having such difficulties contributing valuable content. Regards, Espresso Addict (talk) 00:28, 11 November 2023 (UTC)

Your advice is really helpful and greatly appreciated. I think I will try approaching Wikiproject Medicine as I would guess that some of these editors are epidemiologists. On reflection, I do feel that I should have referenced Rothman et al. and mentioned a different approach to deriving the odds ratio that they provide.

I had forgotten that I was an autoconfirmed editor. Going ahead and posting my article is also a tempting option.

Enjoy your coffee and best wishes, Likelihoodist (talk) 18:23, 11 November 2023 (UTC)

FYI, I decided that I did not want to argue with reviewers who were not necessarily either statisticians or epidemiologists. Instead, I submitted my entry as an addition to the Wikipedia odds ratio article. I did this on Nov 30, and my entry has not been edited since then. I infer that this means that whoever monitors the odds ratio article is ok with what I have done. This gets my contribution into Wikipedia.

What is mildly unfortunate is that it is unlikely to be found by Google, and hence will be harder to find by people who are looking specifically for odds ratios from matched case-control studies. It also makes the odds ratio article longer than, is perhaps, ideal.

With best wishes for the holidays. Likelihoodist (talk) 21:53, 16 December 2023 (UTC)

Hello, Likelihoodist. This is a bot-delivered message letting you know that Draft:Odds ratio for a matched case-control study, a page you created, has not been edited in at least 5 months. Drafts that have not been edited for six months may be deleted, so if you wish to retain the page, please edit it again or request that it be moved to your userspace.

If the page has already been deleted, you can request it be undeleted so you can continue working on it.

Thank you for your submission to Wikipedia. FireflyBot (talk) 23:05, 6 April 2024 (UTC)