More on Anglican Mainstream’s Sexual Exclusivity Post

You might have noticed, if you read Anglican Mainstream, that Lisa has posted two responses to my critique of her piece a few days ago. I am of course disappointed that she couldn’t engage in open discussion on the comment thread on this site. Anglican Mainstream don’t allow any comments or interaction on their own site, so I couldn’t post any response there.

Anyway, you know where this is going don’t you?

Here’s the first piece that Lisa wrote.

Please note, first, that the survey it quotes (1) relates to men who identify with the gay community. Gays who do not identify with the gay community will not necessarily conform to the patterns which emerged.

Second, the roughly two-thirds preference for open partnerships (more accurately, 65%) in this survey was based upon these statistics:  44% of the respondents had had both a main partner and other casual partners while 24% had had only one sexual partner.o do not identify with the gay community will not necessarily conform to the patterns which emerged.

Finally, the partnered gays in the survey are self-identified as partnered and this will have various meanings. Nevertheless the tendency towards sexually open partnerships that emerges finds support in the survey of the first cohort of  Vermont civil unions (2000-2001), where 50% of gay men in civil unions held to sexually open patterns for their civil unions  (this compares to 79% of married male heterosexuals who believed in sexual exclusivity).(2) There is little reason to think that marriage in place of civil unions/partnership will have a significant influence on this tendency.  

(1) Rosenberg, E. S., Sullivan, P. S., Dinenno, E. A., Salazar, L. F., & Sanchez, T. H. (2011). Number of casual male sexual partners and associated factors among men who have sex with men: results from the National HIV Behavioral Surveillance system. BMC Public Health, 11, 189.

(2) Esther Rothblum and Sondra Solomon, Civil Unions in the State of Vermont: A Report on the First Year.  University of Vermont Department of Psychology, 2003.

Right, where to start?

Please note, first, that the survey it quotes (1) relates to men who identify with the gay community. Gays who do not identify with the gay community will not necessarily conform to the patterns which emerged.

This is the first error. Rosenberg et al describes the sample as follows,

We used data from the first MSM cycle of the National HIV Behavioral Surveillance system (NHBS-MSM1), collected from MSM in 15 metropolitan statistical areas (MSAs) from November 2003 to April 2005; participating cities have been previously reported [14]. Men were considered eligible if they were male, at least 18 years of age, current residents of participating MSAs and able to provide informed consent. Men who were determined to be eligible were invited to participate in a face-to-face interview. The NHBS-MSM1 sampling strategy and rationale have been described previously [9,15,16]. Briefly, venuetime-space sampling was used to systematically recruit participants in venues, such as bars, dance clubs, and social organizations , frequented by MSM.

Read the bit in bold carefully. The sample was lifted off the larger database, specifically from MSM (men who have sex with men). Because this was particularly a survey about sexual practices, the sample was specifically geared towards men who have sex with men, regardless of their orientation – that’s why a significant sub-population identified as “heterosexual” (contrary to what Lisa wrote when she claimed this was a group that identified as “gay”).  In brief, the sample was specifically designed to locate promiscuous men in order to find out how promiscuous they were. Other men who would confidently identify as “gay” wouldn’t be chosen for this sample because they wouldn’t be attending on a regular basis institutions which were essentially sex cattle markets.

This means that this survey cannot be used to extrapolate any behaviour patterns beyond this sub-population (promiscuous MSM). It certainly can’t be used to infer any behaviour patterns on gay men who enter civil partnerships or who would want to enter gender-neutral marriage if that was available to them. That’s not to say that it isn’t useful research on the promiscuity of a particular sub-section of the gay community, just that it is only applicable to that sub-population. When Lisa writes, “Gays who do not identify with the gay community will not necessarily conform to the patterns which emerged”, we should add “and plenty of Gays who do identify with the gay community will not conform to the patterns which emerged either”.

Next section.

Second, the roughly two-thirds preference for open partnerships (more accurately, 65%) in this survey was based upon these statistics:  44% of the respondents had had both a main partner and other casual partners while 24% had had only one sexual partner.

Here’s the relevant section from the paper.

Among respondents, 76% reported having had a casual male partner; 32% had only male casual partners and 44% had main and casual partners; 24% had main male partners exclusively . Participants had a median of 3 casual male partners (first quartile: 1; third quartile: 9) and a median of 1 (first quartile: 0; third quartile: 1) main male partner. Those who had no main male partners during the previous year had a median of 5 casual male partners , where as those with a main male partner had a median of 2 casual male partners (Wilcoxon p < .0001)

OK, this statement from Lisa is accurate if one assumes that the 44% who had main and casual partners were having concurrent sex with both the main partner and the casual partner(s). This however isn’t so, as the paper describes,

Participants were asked about the total number of men and women with whom they had sex (men: anal or oral sex; women: vaginal, anal or oral sex) in the 12 months before the interview. These total numbers of sex partners were  classified, by sex of the partners, as either main sex partners ( “someone you feel committed to above all others”), or casual sex partners. Exchange sex partners were considered to be a subset of casual sex partners . HIV status was determined by self-report . Other behaviors, such as use of internet chat rooms and drug use , were ascertained by self-report for the 12 months before interview.

Basically, participants were asked how many sexual partners they had had over the past 12 months and to describe these partners as either a “main” partner or a casual partner. However, no questions were asked about concurrence of sexual partners and no assumptions can be made that two (or more) sexual partners were being engaged at around the same time, or that whilst a participant had a main sexual partner they also had a casual partner. We have a reference to “exchange partners” but there is not further data to quantify the numbers involved. Neither is there anything in the data to determine anything about “open partnerships” (as Lisa describes them) and all we have is the number of sexual partners over a 12 month period. We can make statements on the basis of those figures about promiscuity rates, but we cannot make statements about “open relationships” because no such data exists in this survey to determine that.

Lisa continues,

Finally, the partnered gays in the survey are self-identified as partnered and this will have various meanings. Nevertheless the tendency towards sexually open partnerships that emerges finds support in the survey of the first cohort of  Vermont civil unions (2000-2001), where 50% of gay men in civil unions held to sexually open patterns for their civil unions  (this compares to 79% of married male heterosexuals who believed in sexual exclusivity).(2) There is little reason to think that marriage in place of civil unions/partnership will have a significant influence on this tendency.

(2) Esther Rothblum and Sondra Solomon, Civil Unions in the State of Vermont: A Report on the First Year.  University of Vermont Department of Psychology, 2003.

This is perhaps the most disappointing paragraph of everything that Lisa has written over the past few days. Let me explain. Lisa refers to a specific paper, namely Esther Rothblum and Sondra Solomon, Civil Unions in the State of Vermont: A Report on the First Year.  University of Vermont Department of Psychology, 2003. This paper does not exist.

Let me clarify that statement. There are a number of papers that Rothblum and Solomon published last decade as part of a ground-breaking longitudinal study looking at the first ever cohort of men and women and we’ll have a look at them shortly, but this paper does not exist. Not convinced? Click here to look at Sondra Solomon’s full publication list on her CV. It’s not there.

The first paper Solomon and Rothblum published on this study was Pioneers in Partnership in 2004. It is a summary of the 1 year point (i.e. talking to the relevant couples after they’ve been in a Civil Union for a year) and it contains no reported information on sexual behaviour. There is only one measure that looks at issues around fidelity (“Seriously Considered Ending Relationship”) and on this question women in Civil Unions scored lower then their married sisters and men in Civil Unions scored a statistically similar score to their married brothers.

In 2008 Solomon and Rothblum published the study of the three year point in Developmental Psychology and this study did show the results of the question “Understanding of sex outside of relationship”. At the three year point the figures were 0.4 for gay men in a Civil Union, 0.5 for gay men in a permanent relationship but not in a Civil Union and 0.1 for heterosexual men. For women the figures were 0.1, 0.1 and 0.1 respectively (women in Civil Union, women in lesbian relationships outside of Civil Unions and heterosexual women).

What does this tell us? Solomon and Rothblum found that three years after entering a Civil Union, gay men were four times as likely to consider having sex outside the relationship (which of course is not the same as actually having sex outside the relationship) then their  heterosexual married brothers. At the same time, lesbians in a relationship, whether in a Civil Union or not, had the same propensity to consider sex outside the relationship as their married sisters.

However, the claim that Lisa makes, that “50% of gay men in civil unions held to sexually open patterns for their civil unions  (this compares to 79% of married male heterosexuals who believed in sexual exclusivity)” is completely unsupported by either of these two papers. If anyone thinks otherwise, I would welcome being pointed to the relevant page and paragraph or line on a table.

It seems clear to me what has happened – Lisa has copied a secondary source (I tracked it down to the Family Research Council website) without checking that the claim being made is correct. Indeed, the actual citation is completely incorrect – no paper on this subject was published in 2003. It makes no difference that the actual research shows a result that would support Lisa’s case – what is more important from an integrity point of view is that claim Lisa made was not substantiated in the research AND the citation of the source was blatantly incorrect.

Now, as Lisa points out, there is a press release that reports the figures Lisa has quoted. But this is a secondary source and it is not the original University of Vermont press release which you can find here. It reads,

Regarding monogamy, about 82 percent of women in all three types of couples did not approve of sex outside their relationship. Men’s responses were more varied: 79 percent of married heterosexual men and 50 percent of gay men in civil unions considered non-monogamy to be inappropriate, a higher percentage than gay men who were not in civil unions.

You might think this supports Lisa, and you would be correct in as far as she is simply copying what someone else has written, but for me the issue is not that the original press release is in line with Lisa’s assertions, but that the source she quoted from was clearly secondary (an aggregator of journalist’s pieces). The wording in the Newswire piece is different to the original press release and is written in a provocative manner to create a controversy. Compare “straying from the conjugal bed” to “considered non-monogamy to be inappropriate”.

So let’s summarise. The first post written in response to my critique contains one assertion used to imply behaviours across a much wider demographic then the research covered, conflating a sub-population specifically designed to find promiscuous men with “the gay community”. The second assertion interprets a question designed to determine consecutive sexual partnerships as one uniquely interested in concurrent sexual partnerships. The third assertion cites a reference that doesn’t exist and quotes from a secondary reworking of a press release at the start of a series of papers (none of which are actually referenced) in which the data to support the assertion is never actually presented.

I’ve said it before and I’ll say it again. Conservatives have a duty of care to ensure that we handle research in this area fairly and accurately. If we do not then we will be exposed for our laxity and it will undermine the case we have to make.