Talk:Sexual effects of circumcision/Archive 2

Archive 1

Archive 2

Draft mini-table

I think maybe the plan was to break up the table and put little floating tables in each section. Below is a table format that allows the table to be less wide, while containing the same information. I'm using some of the rows from the "penile sensation" part of the table to demonstrate.

Original format

Study	Design	Peer reviewed	Sample size	Finding	Significant¹
Penile sensation
Fink (2002)	Cross-sectional, adult circumcision patients	Yes	40	Worse after circumcision	Almost; p = 0.08
Masood (2005)	Not stated; adult circumcision patients	Yes	88	Better after circumcision in 38%, worse in 18%	Yes; p = 0.01
Masters (1966)	Neurologic testing; subjects matched for age	No	70 (35 c, 35 uc)²	No difference found	Not stated
Bleustein (2003)	Quantitative somatosensory testing	No	79 (36 c, 43 uc)²	No difference found when controlled for other variables	No; p = 0.08

Narrower format

Penile sensation
Study design	Sample size	Finding
Peer-reviewed
Cross-sectional, adult circumcision patients (Fink 2002)	40	Worse after circumcision. Almost significant; p = 0.08
Not stated; adult circumcision patients. (Masood 2005)	88	Better after circumcision in 38%, worse in 18%. Signficant; p = 0.01
Not peer-reviewed
Neurologic testing; subjects matched for age. (Masters 1966)	70 (35 c, 35 uc)²	Not stated; no difference found.
Quantitative somatosensory testing (Bleustein 2003)	79 (36 c, 43 uc)²	No difference found when controlled for other variables; p = 0.08

Presumably text can be placed here beside the table. I've set the width to "350px"; this can be adjusted. Comments welcome. ☺Coppertwig (talk) 00:23, 28 February 2009 (UTC)

Very nice indeed! I think these "mini-tables" will work very well as a "sidebar" to each section. I think we could compress the mini-tables further by putting the sample size into the 'study design' column. I'm tempted to suggest subtle use of colour to code findings, too: for example, using pastel reds for findings of significant harm, pastel green for findings of significant benefit, and grey for non-significant results. But I'm not sure how well that would fit with the MOS. Jakew (talk) 12:17, 28 February 2009 (UTC)

I think it would be difficult to use colours like that in a NPOV way: that is, I don't oppose it, but I think it may be difficult to achieve consensus (and I don't know if the MOS would have something to say about it). Feel free to give it a try if you like, though; if it can be done in a sufficiently NPOV way, it would be useful to the reader.

An alternative to colours would be to have a middle column with simple symbols, such as "+" for apparent benefit from circumcision, "-" for apparent harm, and "0" for no significant result. These could be criticized as over-simplified, as could the colours; but just to the right of them would be the column explaining the findings in a little more detail. Another alternative would be to include the symbol at the beginning of the text of the last column. If such symbols would be criticized as over-simplified, I think colours could be criticized even more so, since they allow the reader to make a snap decision with a glance at the whole table without reading any details. ☺Coppertwig (talk) 16:40, 28 February 2009 (UTC)

Better/worse

Penile sensation
Study design	Better (+) or worse (-) with circumcision	Finding
Peer-reviewed
Cross-sectional, 40 adult circumcision patients (Fink 2002)	-	Worse after circumcision. Almost significant; p = 0.08
Not stated; 88 adult circumcision patients. (Masood 2005)	+	Better after circumcision in 38%, worse in 18%. Signficant; p = 0.01
Not peer-reviewed
Neurologic testing; 70 subjects matched for age, 35 c, 35 uc² (Masters 1966)	0	Not stated; no difference found.
Quantitative somatosensory testing. 79 subjects (36 c, 43 uc)² (Bleustein 2003)	0	No difference found when controlled for other variables; p = 0.08

Well, here's a try; but already there are difficulties: how to classify "Better after circumcision in 38%, worse in 18%": does that count as "better after circumcision" or as "no result" or what? Would increased time to ejaculation be classified as + or as -? Red or green? Or would different colours or symbols be used for that?

I have an idea: no middle column, but in the right-hand column just put "better" and "worse" in bold font. ☺Coppertwig (talk) 16:54, 28 February 2009 (UTC)

You're quite right: while it was a nice idea, it seems difficult to implement while complying with NPOV and NOR. Ok, let's scrap the middle column. Jakew (talk) 19:40, 28 February 2009 (UTC)

The Mucous membrane Issue & Circumcision

Why is there no discussion or references in this article about the glans (head) of the penis being a natural mucous membrane? This is one of the main medical arguments against circumcision, because when the foreskin is sliced away the glans ceases to be a mucous membrane like it is in its natural uncircumcised state. Circumcision is quite simply the process of turning the glans from a (mostly) internal part of the body (a natural mucous membrane) in to a totally external part of the body (non-mucous membrane). This is bad because mucous membranes like the glans are supposed to be naturally moist, while a circumcised glans loses its ability to be moist because the foreskin has been sliced away and thus the retention of smegma and other natural secretions is lost, leading to gradual toughening [keratinization] of the skin of the glans which in turn leads to reduced sensitivity of the glans over time. There are plenty of sources and references regarding this issue - so why is it not mentioned in this article?—Preceding unsigned comment added by 172.163.80.90 (talk • contribs)

Please see the glans sensitivity section of the article. The postulated keratinisation of the glans is discussed (in a comparative study, Szabo and Short found no difference in the keratinisation of the glans), as is the suggestion of reduction in glans sensitivity (the article cites 3 sources reporting no significant differences, and one reporting a reduction). Jakew (talk) 08:36, 17 April 2009 (UTC)

I agree the section is sparse and lacking other views. Feel free to incorporate properly sourced material. Garycompugeek (talk) 13:44, 17 April 2009 (UTC)

Is too much weight is given to Williamson et al. by quoting them?

Williamson et al. conclude: "Not least among the considerations is the worth of sexual preference for male circumcision within the American culture as a valid reason for continuing the practice."

Sexual_effects_of_circumcision#Female_arousal

Given that it is just one study to state as fact (albeit in quotes) that there is a sexual preference for the circumcised penis and that that is a good reason to perform the operation (obviously isn't) are two very big statements to hinge on one study. I say leave out the quote , the first sentence is good enough. Tremello22 (talk) 17:51, 23 June 2009 (UTC)

13th century French rabbi said circumcision promoted premature ejaculation, decreased libido and a more spiritual disposition in men

It is worth quoting the views on circumcision by Rabbi Isaac ben Yedaiah, a 13th century French rabbi:

"A man uncircumcised in the flesh desires to lie with a beautiful-looking woman who speaks seductively to attract him. He vexes his mind to be with her day after day, growing weary in his attempt to fulfil his desire through lovemaking with her."

"She too will court the man who is uncircumcised in the flesh and lie against his breast with great passion, for he thrusts inside her a long time because of his foreskin, which is a barrier against ejaculation in intercourse. Thus she feels pleasure and reaches an orgasm first. When an uncircumcised man sleeps with her, and then resolves to return to his home, she brazenly grabs him, holding on to his genitals and says to him, "Come back, make love to me." This is because of the pleasure that she finds in intercourse with him, from the sinews of his testicles—sinew of iron—and from his ejaculation—that of a horse which he shoots like an arrow into her womb. They are united without separating, and he makes love twice and three times in one night, yet the appetite is not filled."

"And so he acts with her night after night. The sexual activity emaciates him of his bodily fat, and afflicts his flesh, and he devotes his brain entirely to women, an evil thing. His heart dies within him; between her legs he sinks and falls. He is unable to see the light of the King's face, because the eyes of the intellect are plastered over by women so that they cannot now see light."

"But when a circumcised man desires the beauty of a woman, and cleaves to his wife, or to another woman comely in appearance, he will find himself performing his task quickly, emitting his seed as soon as he inserts his crown. If he lies with her once, he sleeps satisfied, and will not know her again for another seven days. This is the way a circumcised man acts time after time with the woman he loves. He has an orgasm first; he does not hold back his strength. As soon as he begins intercourse, he immediately comes to a climax."

"She has no pleasure from him when she lies down on when she arises, and it would be better for her if he had not known her and not drawn near to her, for he arouses her passsion to no avail, and she remains in a state of desire for her husband, ashamed and confounded, while the seed is still in her "reservoir." She does not have an orgasm once a year except on rare occasions, because of the great heat and the fire burning within her. Thus he who says "I am the Lord's" will not empty his brain because of his wife or the wife of his friend. He will find grace and good favor; his heart will be strong to seek out God. he will not fear to behold that which is beyond, and when He speaks to him, he will not turn away."

quoted by Marc Saperstein in Decoding the Rabbis: A Thirteenth-Century Commentary on the Aggadah. Cambridge, Mass, and London, England: Harvard University Press, 1980: pp.97-98.--Orlando F (talk) 18:34, 12 August 2009 (UTC)

I'm sure that Rabbi Isaac ben Yedaiah was (and perhaps is) viewed as an authority on the subject of Judaism, but it seems strangely inappropriate in a medical/scientific topic to rely on an opinion piece by a 13th century rabbi. Frankly it seems beneath the standards of reliable sourcing. Jakew (talk) 19:39, 12 August 2009 (UTC)

As an Orthodox Jew who has spent decades studying halakha, I am not certain that this one particular Rishon would be considered authoritative. The rambam, the Ramban, Rashi, the Rosh, the Rif, the Ravviah, the Rivash, Rabbenu Tam, the Riva, the Maharil, etc. would be more appropriate; although I would like to see the original text and from whence it is quoted. This way, I can try and see if it is brought as authoritative by later authorities. -- Avi (talk) 20:14, 12 August 2009 (UTC)

I can say that it is unlikely that Marc Saperstein is any kind of authority on what is considered normative Jewish tradition. And I do not believe wikipedia should be quoting random 13th century people without some indiciation why that are notable enough. If we had some quote from Maimonedes or Rashi, that would be much different. -- Avi (talk) 20:18, 12 August 2009 (UTC)

"I believe one of the reasons for circumcision was the diminution of sexual intercourse and the weakening of the sexual organs; its purpose was to restrict the activities of this organ and to leave it at rest as much as possible. The true purpose of circumcision was to give the sexual organ that kind of physical pain as not to impair its natural function or the potency of the individual, but to lessen the power of passion and of too great desire." – Moses Maimonides, quoted by L. deMause in The History of Childhood, Psychohistory Press, New York, 1974, page 24, quoted by James DeMeo in Saharasia, OBRL, Greensprings, Oregon, USA, 1998, page 119. See also more Maimonides quotes–Orlando F (talk) 02:27, 18 August 2009 (UTC)

While I own a copy (multiple) of the יד החזקה, the quote you bring is sourced on CIRP to the מורה נבוכים, for which I will have to borrow one to check the accuracy of the translation into English, but at least Maimonides was a towering figure in Judaism and a better source than Rabbi Isaac ben Yedaiah for this purpose. -- Avi (talk)

Kim and Pang 2007

Is missing or not factual in this article. Specifically [[1]] needs to be added to Penile sensitivity and sexual sensation, Foreskin sensitivity, and Sexual practice and masturbation. The 20% statement in Satisfaction is not representative; pleasure decreased for 48%. Finally, the study is missing from the table.Zinbarg (talk) 01:31, 11 January 2010 (UTC)

Already cited. See ref 28. Jakew (talk) 10:44, 11 January 2010 (UTC)

It's cited, but not used well in the text. A main finding is that circumcision often reduces masturbation pleasure, but I don't see that in the text.Zinbarg (talk) 18:14, 20 January 2010 (UTC)

Masood 2005

The source states: "Sixty-four percent had no problem with premature ejaculation before or after circumcision. Only 13% reported improvement in premature ejaculation, whereas 33% found it worse." [2]

My version: "In a study of 150 men circumcised for benign disease, Masood et al. reported that 13% reported improvement in premature ejaculation after circumcision, 33% reported that premature ejaculation became worse, and 64% had no problem with premature ejaculation before or after circumcision"

I stated that as previously written, the statement implied that all men in the study had a problem with premature ejaculation. Jakew reverted back to this version: [3] "[...] disease, Masood et al. reported that 13% reported improvement in premature ejaculation after circumcision, 33% reported that premature ejaculation became worse, and 53% reported no change."

The 53% appears nowhere in the text of the article, and unfortunately the CIRP copy does not reproduce the table Jakew mentions. Jakew, could you describe what the table says that apparently contradicts the 64% figure above, and can you explain why you reverted to the words "no change" that imply that all of them had premature ejaculation?

I noticed that the statement is misleading as it implies that the percentages are of 150 men, when the source says "The mean total IIEF-5 score of the 84 patients at baseline was..." I'd suggest a reword based on that. Blackworm (talk) 00:07, 21 January 2010 (UTC)

Certainly, Blackworm. Table 2 has four columns, labelled "Questions", "Better after circumcision n (%)", "Worse after circumcision n (%)", and "Same as before n (%)". The relevant row has the following information: "Premature ejaculation", "4 (13)", "10 (33)", and "16 (53)". This, then, is the source of the figures 13, 33, and 53.

A footnote to the table notes that "54 patients did not have premature ejaculation before or after surgery". This means, I believe, that the figures listed in the table exclude men who did not have this problem before or after circumcision. I would guess that these 54 are the 64% that Masood et al. refer to in their text, since 54 is 64% of 84 (54 + the 30 listed in the table).

I think it would be misleading to include the 13/33/64 figures, since they refer to different things. The first two figures (13 and 33) refer to percentages of the 30 men who had PE before and/or after circumcision, whereas the last (64) apparently refers to a larger group of men. (The sum of the figures is 110, not 100 as one might expect. The other set of figures sum to 99, which is probably due to rounding.)

It seems to me, though, that saying "in a study of 150 men..." is also misleading, since we should express that it is actually a study of a subset of those men. The simplest solution would seem to be deleting "150". Any thoughts? Jakew (talk) 12:14, 21 January 2010 (UTC)

Thank you for the explanation. Indeed, I should have noticed something wrong. I think your interpretation is the most likely. I'll remove the "150" to address the other issue. Blackworm (talk) 07:18, 22 January 2010 (UTC)

I took the library of rewording to make the prior issue clear: "In a study of men circumcised for benign disease, Masood et al. reported that of those who stated they had prior premature ejaculation, 13% reported improvement after circumcision, 33% reported that it became worse, and 53% reported no change." I hope this has your approval. Blackworm (talk) 07:22, 22 January 2010 (UTC)

Looks good to me. I hope that the library is safe and comfortable in your possession. Jakew (talk) 10:35, 22 January 2010 (UTC)

Penile sensivity graph

The following graph has moved from Circumcision to Penis, and now to discussion here for the time being. What do others think about it's location here? Why, and why not?--Studiodan (talk) 14:37, 24 January 2010 (UTC)

Introduction and Sensitivity graph

The article requires an neutral introduction. The current vague sentence is POV.

The article must include that sensitivity graph.Zinbarg (talk) 14:41, 12 February 2010 (UTC)

Please explain what is "POV" about the introduction. Jakew (talk) 14:59, 12 February 2010 (UTC)

Sorrell's is in wrong section (glans), and info poorly conveyed.Zinbarg (talk) 14:43, 12 February 2010 (UTC)

If you read the article, you'll see that Sorrells et al. is cited four times: twice in the "glans sensitivity" section, once in the "foreskin sensitivity" section, and once in the "ejaculatory function". Jakew (talk) 14:59, 12 February 2010 (UTC)

What about including a proper introduction?

Sorrell's should be in the Penile sensitivity and sexual sensation section.

The ejac function is a misquote ... it's not Sorrell's making that claim, it's Waldinger.

Where's the Sorrell's graph? Objective data is always neutral.

You have misquoted Boyle because et al his focus is more on sensation loss from damage or removal of the frenular delta and band. You've got a stupid vague sentence.

Where's Taylor???

Many of the survey studies had a very low rate of survey response. They are also frequently not statistically relevant results. The table should include a column stating relevance. You'd end up throwing half out. The table is therefore POV misleading (no significance means no finding, not no impact in "finding.").

You've produced a huge pile of POV.Zinbarg (talk) 15:35, 12 February 2010 (UTC)

Let me address these questions one by one:

• "What about including a proper introduction?" That's a good idea. However, you've tagged the introduction with {{POV-section}}. I agree that it's too short, but that doesn't mean that it is POV. If there are neurality problems, please explain what they are. Otherwise, the tag should be removed.

• I asked for an introduction, and you rejected my proposal as anti-circ.Zinbarg (talk) 16:39, 12 February 2010 (UTC)

• I've removed the POV-section tag, since you haven't provided an explanation of why it violates WP:NPOV. Jakew (talk) 10:50, 17 February 2010 (UTC)

• "Sorrell's should be in the Penile sensitivity and sexual sensation section" Sorrells et al. don't really say much about sexual sensation, nor about penile sensitivity in general. They do make some specific claims about the sensitivity of the glans and foreskin, so it makes sense to cite what they say in those subsections.

• Sorrell's is all about penile sensitivity and sensation, so it belongs in Penile sensitivity and sexual sensation. Zinbarg (talk) 16:39, 12 February 2010 (UTC)

• I see little point in repeating myself, so please see above. Jakew (talk) 10:50, 17 February 2010 (UTC)

• "The ejac function is a misquote ... it's not Sorrell's making that claim, it's Waldinger." See the 4th paragraph of Sorrells' "introduction" section. The full sentence reads: "A recent multinational population survey using stopwatch assessment of the intravaginal ejaculation latency time (IELT) found that Turkish men, the vast majority of whom are circumcised, had the shortest IELT." (emph added)

• Sorry, I'm wrong. Zinbarg (talk) 16:39, 12 February 2010 (UTC)

• Okay. Jakew (talk) 10:50, 17 February 2010 (UTC)

• Where's the Sorrell's graph? Objective data is always neutral. Illustrating Sorrells et al. but not other studies would give Sorrells undue weight, as has been explained in several other discussions.

• It's objective and relevant. WP:UNDUE is silly Jakew. Propose graphs you think are needed instead of rejecting one because you don't like it. Zinbarg (talk) 16:39, 12 February 2010 (UTC)

• Silly or not, it is Wikipedia policy. And since adding the graph causes a policy violation, the onus is upon those who wish to add the material to ensure that NPOV is maintained. Jakew (talk) 10:50, 17 February 2010 (UTC)

• "You have misquoted Boyle because et al his focus is more on sensation loss from damage or removal of the frenular delta and band. You've got a stupid vague sentence." As far as I can tell, these are exact quotes from Boyle et al.

• Quote you selected is a vague generalization, instead of citing his specific impacts. POV Zinbarg (talk) 16:39, 12 February 2010 (UTC)

• Sorry, your objection is too vague to address. It might be more productive for you to suggest a specific change. Jakew (talk) 10:50, 17 February 2010 (UTC)

• "Where's Taylor???" Refs 17, 18, and 19, all cited in the "foreskin sensitivity" section, as you could easily determine for yourself.

• Very biased or diluted Taylor. Please read [[4]] Maybe that's part of the intro though. Zinbarg (talk) 16:39, 12 February 2010 (UTC)

• An interview with an anti-circumcision website is not a reliable source, hence that could not be cited in the article, and whatever he says in it is irrelevant. What matters is what he says in reliable, peer-reviewed sources. Jakew (talk) 10:50, 17 February 2010 (UTC)

• "Many of the survey studies had a very low rate of survey response. They are also frequently not statistically relevant results. The table should include a column stating relevance."' How are you defining "relevance"?

• You list the significance level, and many are no. But you also say no finding as though it really means something. It is misleading. If a study finding isn't relevant, it shouldn't be included. You know stats better than most. You mislead. Zinbarg (talk) 16:39, 12 February 2010 (UTC)

• A result that is not statistically significant is meaningful result. How do you suppose that researchers find out that A has no effect on B? Excluding studies that find no statistically significant results would make it impossible to present results finding no association, and that would be unacceptable under WP:NPOV. Jakew (talk) 10:50, 17 February 2010 (UTC)

• "You'd end up throwing half out. The table is therefore POV misleading (no significance means no finding, not no impact in "finding.")." A finding that is not statistically significant can mean either a) that there is no association, or b) that the study failed to find a true association. Jakew (talk) 15:55, 12 February 2010 (UTC)

• Agree, so why do you have studies that are one of those (no, or failed), but you have it as a "finding." POVZinbarg (talk) 16:39, 12 February 2010 (UTC)

• Similarly, a statistically significant result can mean either a) that there is an association, or b) a false positive. The result may not be conclusive, but it's still a result. Jakew (talk) 10:50, 17 February 2010 (UTC)

• Note: I've reformatted Zinbarg's comments and have had copied his signature to make it clear who is responsible for what text. Jakew (talk) 10:38, 17 February 2010 (UTC)

Why don't you label the col "no association, or the study failed to find a true association"Zinbarg (talk) 00:06, 17 February 2010 (UTC)

Because it is not our place to reinterpret the results of published studies. Jakew (talk) 09:59, 17 February 2010 (UTC)

But that's (wp nor) exactly what you have done with your table.Zinbarg (talk) 00:36, 1 March 2010 (UTC)

In what way? I'm fairly sure that the table simply tabulates data and interpretations already published. Jakew (talk) 11:13, 1 March 2010 (UTC)

Denniston study

The change in citation came because what the article originally cited a letter to the editor (also written by Denniston), which in turn cited the Denniston study. I changed the citation so it directly cited the study instead of citing a letter that cited the study. The change from 58% to 63% comes from, if you look at the results section (specifically page 48), you will notice that 24 of the respondents cited decrease sensitivity as a disadvantage after their circumcision. This is out of a total of 38 respondents. 24 out of 38 is 63%. —Preceding unsigned comment added by Trinu (talk • contribs) 17:16, 16 April 2010 (UTC)

I reverted this change because Denniston's article in an anti-circumcision book falls short of Wikipedia's requirements for reliable sources in medical topics. Denniston's 2004 letter in Urology is at least in a peer-reviewed journal, and as such is more suitable for inclusion. Jakew (talk) 17:28, 16 April 2010 (UTC)

Misleading Summary

The table is misleading to most are wiki audience; basically anyone without a good background in statistics. Looking down the table, there are many "no finding" from studies that are not statistically significant. In college, I was taught to ignore (not even read) studies with lousy (low) t stats. Including those studies in the table causes confusion, and is likely misleading. I'm afraid most readers see no finding as no impact. We should remove those with low t scores, or place them in a different section of the table. Until then, it's best to just remove the table.Zinbarg (talk) 03:49, 26 May 2010 (UTC)

Actually, Zinbarg, that is how studies report no differences: they test for a statistically significant difference and report its absence. I've restored the table. Please do not remove it again. Jakew (talk) 08:41, 26 May 2010 (UTC)

And studies with very low t stats are considered irrelevant. The only time we read such studies in college was to design better studies. Please help fix the table and don't force our Wiki audience to weed through junk (at best, assuming they aren't mislead).Zinbarg (talk) 17:39, 26 May 2010 (UTC)

Could you explain in English what you mean by "studies with very low t stats"? I've asked you before, but you didn't answer. Jakew (talk) 17:51, 26 May 2010 (UTC)

From your table key: "If stated, author's analysis is used. Otherwise, significance is considered to be p <= 0.05." I think we used a much higher number (t >3?) in college. Looking at some old notes, I'm not sure how your "p" is related to my t [[5]] score. Regardless, the studies and the table acknowledge "not significant." Including "not significant" (and at the far end of the table) is confusing and probably misleading.Zinbarg (talk) 18:36, 26 May 2010 (UTC)

P values aren't the same as t statistics. P values indicate the significance of a statistical relationship; they range from 0 to 1 and the smaller the p value, the more significant the result. Researchers choose a so-called "alpha", which is a threshold, below which p values are said to be significant, and above which p values are said to be non-significant. 0.05 is the value that is normally used in medical studies. "Not significant" does not mean "does not matter", nor does it mean that anything was wrong with the study. It means that no evidence was found indicating a relationship between the two quantities being studied. That's valuable information, and it should not be excluded. By analogy, it's like telling a mechanic that nothing happens when you turn the ignition key: knowing that helps the mechanic to understand the problem. Jakew (talk) 18:46, 26 May 2010 (UTC)

Unindent. Thank you. No relationship found. No finding. We shoud segregate significant findings from those which found no information. Your mechanic analogy is silly. Several drivers report differing results with the ignition system and the study finds no result significant (sometimes it helps, no change, or doesn't work, with no apparent relationship found) the mechanic diagnosis. It's usually methodology problems leading to the insignificant result.Zinbarg (talk) 19:15, 26 May 2010 (UTC)

No, it is not "no finding". It is a finding that there was no statistically significant relationship, and as such it is an important as any other finding. In fact it is the only possible result of a study if the two quantities are truly unrelated, and for that reason it is very valuable information. Jakew (talk) 19:23, 26 May 2010 (UTC)

The studies and the table call it "no finding." You ignore the fact that frequently there's a methodology problem causing the lack of result. For example a low and/or biased responce to questionare, and poor construction of the questionare. I don't understand why you don't want "no finding" studies separated from those with statistically relevant relationships; because it would help our audience.Zinbarg (talk) 19:29, 26 May 2010 (UTC)

No, the table is headed "Significant", so "No" in a cell means "not significant", not "no finding". Which studies call it "no finding"?

As for methodology problems, those can affect studies finding significant results as easily as those finding non-significant results. I don't think it would make any sense to separate studies finding no statistically significant difference from those finding significant differences. It would mean that instead of being able to see the results of all studies at a glance, the reader would have to scroll down to a different section, so it would be less convenient. Jakew (talk) 19:49, 26 May 2010 (UTC)

There are two columns in question. The column headed "Finding" may say "no difference." I mistakenly wrote "no finding," but I think it's the same thing. The last column is headed Significant, and the rows of studies are rated "yes" or "no" based on that p score or the study parameters.

You're much better at statistics than I. But I know methodology relates to significance. For example, when you sample 1000 patients and only 45% participate, results are weakened. Isn't that weakness reflected in the t stat or p score?

Why not just show the results of statistically strong studies? I think you once wrote that there were 40?+ studies to review. Focus on the best.Zinbarg (talk) 02:39, 27 May 2010 (UTC)

Yes, the "finding" column sometimes says "no difference", but that is not at all the same as "no finding". As the old saying goes, "absence of evidence is not evidence of absence", and a finding that there is no difference is not the same as no finding at all. "No finding" would describe a study that did not investigate the association at all. "No difference" is a finding of a study that did investigate and found that there was no relationship.

Response rates are generally unrelated to p values, because usually statistical analysis is performed with responses that were received. There would be an indirect effect, though, in that the sample size would be reduced to 450, which could affect the power.

What's a "statistically strong" study? How do you determine that without performing original research? Jakew (talk) 09:31, 27 May 2010 (UTC)

Scratch that. I just added up the table and looked at the number of "yes" Significant; there are almost no yes es. What a mess. Why are there so few of those ~40 in the table? Is this another hugh problem with the table? How about adding a column to contain the study conclusion statement?Zinbarg (talk) 02:46, 27 May 2010 (UTC)

There are very few "yes"es because most studies have found no association between circumcision and various hypothetical sexual effects. A likely explanation for that is that circumcision actually has no effect. I think we're running out of space for extra columns, but we do occasionally include quotes of conclusions (where relevant to the item in the table), and I wouldn't object to doing this for more studies. Jakew (talk) 09:31, 27 May 2010 (UTC) (edited 16:25, 27 May 2010 (UTC))

There's a serious problem with reference 39. Bailey doesn't address intromission, and I read some study maybe in the New Zealand Medical I think stated the opposite. And remember Wison's argument (reduced ease of entry cuts down on promiscuous acts).Zinbarg (talk) 02:59, 27 May 2010 (UTC)

Please explain what, exactly, is the problem with ref 39? We say "Bailey et al. report that there is a preference by women for circumcised men, mentioning that the circumcised penis enters a woman more easily and is less likely to cause injury to the vagina.", and that is what Bailey et al. report. Jakew (talk) 09:31, 27 May 2010 (UTC)

Unindent. Sorry, I only had access to the PubMed summary, which lacks the last part of your sentence. I realize I'm not qualified to speak about statistics, and we have to use what's availaible though low quality. I need to read all those studies. Probably not.Zinbarg (talk) 17:32, 27 May 2010 (UTC)

Okay. Would you prefer to continue this discussion once you've had a chance to read the studies? Jakew (talk) 18:35, 27 May 2010 (UTC)

Yes, I've got work to do first. Already, I find reduced sensitivity is common to thestudies. Note, that may be a good thing if you object to masturbation, or premature ejac. I really can't see why you think Sorrell's is undue wt, if most studies find same.Zinbarg (talk) 02:07, 4 June 2010 (UTC)

Most studies do not find the same, Zinbarg. Jakew (talk) 08:21, 4 June 2010 (UTC)

the real funny thing is that gay men prefer being circumcised much more than heterosexual ones. Because they think it decrases HIV. —Preceding unsigned comment added by 188.174.35.55 (talk) 09:20, 18 January 2011 (UTC)

Dalton addition

81.253.49.189 (talk · contribs) recently added the following to the "Summary of research findings" section:

• An English doctor published on the BMJ's site a short meta-analysis concluding that circumcision practiced upon adults for medical reasons has the following effects: 1/3 of dissatified, 1/3 indifferent and 1/3 satisfied (Dalton J. Effect of circumcision on intromission and sexual satisfaction. 7.01.2008. http://www.bmj.com/content/335/7631/1180/reply#183746, find his e-letter to the Editor around the first fifth of the page).

There are several problems with this addition. In no particular order:

1. Factual error: John Dalton is not a doctor.
2. Factual error: Dalton's summary is not a meta-analysis. It's a simple pooling of data instead.
3. Poor choice of section: the "Summary of research findings" section presents a table showing all findings. A better place, if this were suitable for inclusion, would be the "Satisfaction" section.
4. Reference formatting: The citation is formatted improperly.
5. Reliability: Letters that are printed in a journal are less reliable than peer-reviewed articles. Electronic letters, such as Dalton's, are less reliable still. I would not regard this source as adequately reliable.

The first four of these problems are correctable, but the last is not. Jakew (talk) 18:23, 26 November 2010 (UTC)