Jump to content

Talk:Abortion–breast cancer hypothesis/GA1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

GA Review

[edit]

The article is, for the most part, very well-written and thorough researched.

Red XN Prose. Mostly good, I made a few minor copyediting changes, but it reads very well. A few sentences need to be fixed, mainly because I cannot understand exactly what's attempted to be said here in order to fix them myself:

  • This sentence is incomplete and needs to be finished: "The results indicated a increased 1.9 (1.2 - 3.0) RR for induced abortion and 1.5 (0.7 - 3.7)"
  • This sentence doesn't make much sense as written: "The study was funded by Family Health International, a pro-choice NGO and although the study started with 49,000 "after 11 years she had fewer than 5,000" who stayed in the study."
  • "The first section of Table 1 in the Melbye study:" -- needs a source.
  • "The following are induced abortion results from Table 4 of the Michels study, with parity distinguished between nulliparous (no children) and parous (had children):" -- needs a source.

Green tickY Factually accurate. Most everything that is required to be cited, or that could be challenged, is cited. The sources appear to meet WP:RS guidelines.

Green tickY Broad in coverage. This article is very thorough, and covers both the pro & con aspects of the issue very well. At 81 Kb, the article is getting a bit long. Going forward, perhaps some of the descriptions on studies could be moved to linked articles, and descriptions here could be paraphrased? But I don't think this would hold the article from GA at this time.

Red XN WP:NPOV. Quite surprisingly, I do not see any major NPOV issues with the article. Like I said, it covers the pro & con aspects quite well.

I still think that the pro & con is covered pretty well (good enough for GA, but another user has raised some other NPOV concerns, so see below. Dr. Cash (talk) 02:03, 28 June 2008 (UTC)[reply]

Green tickY Stability. The vast majority of edits in the edit history are by a single user, RoyBoy (talk · contribs). So no, there are no major stability issues.

Green tickY Images. Article's a little light on images, and could use an eye-catching image for the top-right corner (though this is not a requirement for GA). The images that are present are appropriate tagged with image copyright tags and fair use rationales.

So, the article mostly meets the GA criteria, and can be listed once the issues with the aforementioned sentences are cleared up. I will put it on hold until that is done.

Cheers! Dr. Cash (talk) 22:41, 27 June 2008 (UTC)[reply]

I don't really agree with the NPOV assessment (specifically the Proponents section, as I've already commented). Some concerns:
  • The proponents section. Some ideas on what to add:
1. any conflicts of interest on the part of prominent opponents should also be listed.
2. The qualifications, related research, etc... of the proponents are notable (why would anyone believe them?).
3. Some characterization of what support (if any) the proponents have from other notable scientists or other nominally neutral parties would be nice.
Hopefully my tweaks have addressed these concerns. - RoyBoy 17:04, 29 June 2008 (UTC)[reply]
Made one other edit to discussion of Dr. Brind's background, just language concerns, not so much content. Somedumbyankee (talk) 19:34, 29 June 2008 (UTC)[reply]
Good call again, very RS with pertinent info. - RoyBoy 01:07, 4 July 2008 (UTC)[reply]
  • The Howe study could use a bit more discussion about why the possible RR values are so broad (0.7-3.7 particularly) and why it was thought to be inconclusive.
The "With more data the confidence interval becomes smaller" in Abortion-breast_cancer_hypothesis#Epidemiological_studies addresses this issue. Does it need to be reiterated so quickly? The specific patient pair numbers for this is 65/35 case/control for induced and 17/11 case/control for spontaneous. - RoyBoy 17:04, 29 June 2008 (UTC)[reply]
Hmm. I'd just put a few words on it like "this is an example of the problem of small study size blah blah" No Great American Novel, clearly. Somedumbyankee (talk) 19:34, 29 June 2008 (UTC)[reply]
My goodness, found a significant mistake on my part. I have 1451 case patients for Howe, but in actuality they are 100 case patients WITH abortions and 63 controls. Damnit! All this time misrepresented the statistical significance of the Howe study. Fixed, and explained CI in relation to dataset. - RoyBoy 01:55, 4 July 2008 (UTC)[reply]
  • "Highly regarded study" (Melbye) is kind of weaselish, especially since rejecting that study is the cornerstone of Dr. Brind's arguments.
Not a precise characterization. Brind does consider the analysis flawed, but that does not change the quality of their dataset which is widely acknowledged (Brind included) as being one of the most complete in the world. Moreover, as referenced by the now fixed PP link, it was received by notable publications and organizations as authoritative. Even if we take Brind's criticism as valid, that doesn't change the broad and positive coverage the Melbye study received. Should one, two or several news references be added? - RoyBoy 17:04, 29 June 2008 (UTC)[reply]
One is plenty. Somedumbyankee (talk) 19:34, 29 June 2008 (UTC)[reply]
Added detailed NYTimes article. - RoyBoy 01:37, 4 July 2008 (UTC)[reply]
  • For clarity, the two Lindefors-Harris studies should be identified by something other than date since they're about two different things (actual study 1989 vs. analysis of response rates 1991), unless they are the same study and I'm very confused (probably also a problem).
Good call, transitioned to Lindefors-Harris "response bias study". - RoyBoy 17:04, 29 June 2008 (UTC)[reply]
  • One of the cites used is Planned Parenthood (currently #45, used 3 times). I would avoid using this organization whenever possible when talking about the science since they have a clear position in the overarching abortion debate and listing then with neutral sources may distract readers from the presentation.
Seems like an excellent precaution. Replaced one with "WEED", and removed the other as it reiterates Sanderson's position. Leaving the Melbye ref since Planned Parenthood is named and it seems important to clarify what science pro-choice organizations focus on. - RoyBoy 17:04, 29 June 2008 (UTC)[reply]
The article has some language which sounds like it wants to disprove the link, (for example one that I changed). It's minor, but it has a certain sense of "gotcha!" which I don't care for. Somedumbyankee (talk) 23:44, 27 June 2008 (UTC)[reply]
LOL, indeed. There has been some contentious editing in months past, not all of which I have been able to spot and remove. - RoyBoy 17:04, 29 June 2008 (UTC)[reply]

These issues seem valid as well, and should be addressed prior to GA. The planned parenthood reference definitely needs fixing; when I click on it I get a 404 not found error, so I cannot even verify the content. Dr. Cash (talk) 02:02, 28 June 2008 (UTC)[reply]

Excellent and invaluable analysis, both of you. I'll be working on it this weekend. The PP link is curious as I fixed that recently from a previous move. Will look into it. - RoyBoy 22:10, 28 June 2008 (UTC)[reply]
  • This sentence is incomplete and needs to be finished: "The results indicated a increased 1.9 (1.2 - 3.0) RR for induced abortion and 1.5 (0.7 - 3.7)"
Fixed. - RoyBoy 16:14, 29 June 2008 (UTC)[reply]
  • This sentence doesn't make much sense as written: "The study was funded by Family Health International, a pro-choice NGO and although the study started with 49,000 "after 11 years she had fewer than 5,000" who stayed in the study."
Tweaked. - RoyBoy 16:14, 29 June 2008 (UTC)[reply]
  • "The first section of Table 1 in the Melbye study:" -- needs a source.
Added. I sourced it originally, but removed it as redundant. - RoyBoy 16:14, 29 June 2008 (UTC)[reply]
  • "The following are induced abortion results from Table 4 of the Michels study, with parity distinguished between nulliparous (no children) and parous (had children):" -- needs a source.
Added. - RoyBoy 16:14, 29 June 2008 (UTC)[reply]

Are there any outstanding issues? - RoyBoy 02:45, 4 July 2008 (UTC)[reply]

There are some language things that I think could be better, like "posits" (correct, but unusual and distracting word choice). There are some extra commas here and there that I'm seeing and some awkward wordings (q.v. "In early pregnancy, levels of estrogen increase, leading to breast growth in preparation for lactation.") I'd definitely work on the copy editing before considering it for FA, but I know the "needs copy editing" thing is incredibly annoying because it sunk my last GA submission. I don't see it as a game-ender, and I'd defer to the primary reviewer as to whether that's enough of a problem to keep it on hold. SDY (talk) 03:44, 4 July 2008 (UTC)[reply]

I think that the issues raised by both of us are largely addressed, and that the article meets the Good Article criteria and can be listed. I still recommend a good copyedit; while I've made a few minor copyediting changes, there could still be some minor issues and the reviewers at FAC are very thorough -- they're going to go through this with a fine-toothed comb, so make sure all the commas are in the right place, and all other punctuation is in place. But overall, I think that the prose is very "good", and there are no major glaring issues that a non-English major is going to catch,... So I think the article is therefore a "good article". Cheers! Dr. Cash (talk) 16:27, 24 July 2008 (UTC)[reply]

Woot! Your collective feedback made this possible, thank you Derek, SDY, Mastcell and in small part OrangeMarlin as the controversial past tense tweak was necessary in retrospect. - RoyBoy 22:30, 24 July 2008 (UTC)[reply]