(no subject)

[theweaselking]

British Office Of National Statistics: There are 750,000 GLBT people in all of the UK!
gaydar.com: Uh, we have 2.2 MILLION UK profiles - GLBT people in the UK actively, right now, looking for a date. With us alone, and that doesn't count any other dating sites.

Comments

[almostwitty]

I don't doubt that the ONS stats are somewhat flawed to say the least ("Why, yes, strange man on my doorstep asking if I'm gay or not, of course I'm gay. Why, I don't mind that my parents standing behind me didn't know up till now!"), I'm sure it's possible for one gay/lesbian person to register more than once.

[theweaselking.livejournal.com]

Oh, sure, it's possible.

But our requirement, here, is that on ONE single (albeit popular) dating site, EVERY SINGLE GLBT PERSON IN THE UK HAS CREATED THREE PROFILES.

Even assuming that a full third are fakes and trolls, we have "every GLBT person in the UK has created an average of two profiles".

On one, single, specific internet dating site.

[almostwitty]

My best friend (lesbian) makes a whole brand new profile on a certain dating site after each failed relationship, so that her now-ex won't find her. So by my reckoning, she has at least six profiles on said site by now.

[theweaselking.livejournal.com]

Unless this is a standard, universal thing, she's in the outlier group moving towards "unbelievable average of two" department.

By which I mean, even if every gaydar.com user is legit and had 6 profiles? I still doubt that accounts for every GLB (no T) person in the UK

[dscotton.livejournal.com]

I don't know a lot about dating sites but I do know a bit about spammers and account creation - and I wouldn't be at all surprised if 90% or more of their accounts were bad.

On the other hand, the official number is pretty clearly an underestimate just going by the old 10% guideline. But I'd consider the dating site's response a good publicity stunt more than anything else.

[snakey.livejournal.com]

The ONS are fucking idiots, and this is indeed the kind of crap that damages us. D:

On a related note...

*gentlecouch* The ONM statistic claims to speak to only LGB people, not trans* people.

Also, having just made an experimental profile there to find out, I can confirm that the Gaydar site (at least the men's version) makes no mention of trans people at all - so the 2.2 million statistic from gaydar.co.uk only includes GB trans men (and presumably LB trans women). The site you link to doesn't seem to have taken this into account at all, but assumes that "T" is an automatic extension of "LGB" and questions about "sexuality".

[ipslore.livejournal.com]

Well. Almost certainly, the BONS number is low, and the Gaydar.com number is high. And, given the ratios, the real number is probably closer to the 2.2 gigapeople. But, LGBTNetwork hasn't even taken the five minutes it would require to do even that extremely rudimentary analysis; they trumpet their figure as though it were a hard, irrefutable fact, and use it to laugh at those moronic 'number crunchers'. They mention the Daily Mail (lol) and politicians as though they were an intrinsic part of the study's findings. I know there's a whole bunch of ways to divide the geographical region of 'Britain and the surrounding areas'; are these studies even investigating the same population, or asking the same questions? I don't know, because neither one is linked!

So, in short: Gay Representation Fail for the BONS, sure, but also Statistics Fail for LGBTNetwork.

[theweaselking.livejournal.com]

Why would gaydar.com's number be high?

Shouldn't the number of troll profiles be reasonably expected to be *lower* than the number of gay people who don't use gaydar.com online dating? Especially since the site apparently does work? Which would, then, mean that the number of UK gaydar.com users is still lower than the total number of gay people in the UK?

they trumpet their figure as though it were a hard, irrefutable fact,

More, they say that the ONS number is transparently, obviously, trivially shown to be way, way, way too low, and they mock the Mail for taking it seriously.

[ipslore.livejournal.com]

Okay, yeah, brainfart there; by 'real number', I mean 'real number registered on gaydar.com', not 'real number of gay Britons'. And I'll agree that the latter number is probably larger than 2.2GP. What I'm saying is, there's enough unknown variables here-- differences in response rates between a phone poll and signing up for a website; possible regional variations; maybe the gaydar numbers include GLBTQ while the BONS was only looking for GL; etc; and then random noise on top of that-- that I don't think it's as simple as they're making it. I'll agree with you that the BONS number is way, way, low, but I won't add the third 'way', and I wouldn't call it transparent, obvious, or trivial.

That, and I'm a little pissed about the crack at 'number crunchers sticking their nose in trying to put a number on everything'.

[snakey.livejournal.com]

See my point above re ANY of the numbers taking into account trans people (ie, they don't).

[kjn]

If there were gay 2.2 gigapeople in Britain, I'd be impressed. After all, the UK has about 60 megapeople total.

[sparkindarkness.livejournal.com]

I am so going to rant on this one - but can there be anything more ridiculous than this survey?

A complete stranger knocks at your door or calls you up and asks your sexuality - and they expect everyone to just out themselves? Really?

Also they've assumed the 6% of people who declined to say or who said "other" are straight (or not likely to be GBLTQ) hmmmm

[le-trombone.livejournal.com]

Also they've assumed the 6% of people who declined to say or who said "other" are straight (or not likely to be GBLTQ)

Yeah... "what part of 'other' don't you understand."

[lederhosen.livejournal.com]

You might want to read what the ONS report actually said about that:

"The ‘Other’ option on the sexual identity question was included to address the fact that not all people will fall in the first three categories and that some people such as those that are asexual, may feel no sense of sexual identity at all. In addition, individuals who disagree with the simplistic male/female gender binary, or who were against categorisation based on the gender of people to whom they were attracted or with whom they had relations, could also prefer to identify as other. Previous ONS work also showed that a small number of heterosexual respondents may not understand the terminology used in the question so may select ‘Other’ instead (Betts, 2009). "

[sophia8.livejournal.com]

"A complete stranger knocks at your door or calls you up and asks your sexuality - and they expect everyone to just out themselves? Really?"
No. Look at the ONS's own report (http://www.statistics.gov.uk/articles/nojournal/measuring-sexual-identity-report.pdf) and they tell you the measures they took to preserve respondents' privacy. (Basically, they had people replying with the number of the response that fitted them.) And of course the survey was totally voluntary, and agreed to in advance.
But, yes, they did leave out the T in LGBT - the only answers were Gay, Straight, Bi, Other. So a big boo there.

[sparkindarkness.livejournal.com]

Aye voluntary and agreed to - but half the country wouldn't even know who the ONS is, let alone enough to trust them with this information

There is also guarantees of the information being kept secret from household members, but the siumple fact a stranger is at your door, show cards or not, is intimidating. I don't think either measure is likely to reassure that much.

[sophia8.livejournal.com]

As I said, these types of surveys are agreed to in advance.
That means the household, - selected at random - receives a letter laying out the survey aims, and is then INVITED to take part. Nobody turns up on the doorstep, or telephones, unannounced.

[sparkindarkness.livejournal.com]

Which in turn means the information is not going to be accurate. Since you have the "I know what's being asked, not comfortable answering it so I'll not take part" or the "I don't know what's being asked, know that they're going to ask questions - not comfortable with a stranger asking these questions so lying about it"

In both cases, it doesn't lead to an accurate accounting

[snakey.livejournal.com]

Interesting critique.

[snakey.livejournal.com]

But, yes, they did leave out the T in LGBT - the only answers were Gay, Straight, Bi, Other. So a big boo there.

No. They were seeking statistics on sexuality. Trans people can be "Gay, Straight, Bi, Other" just like cis people.

[lederhosen.livejournal.com]

Also they've assumed the 6% of people who declined to say or who said "other" are straight (or not likely to be GBLTQ)

No, they haven't. The ONS report (pp. 8-9) states that 95% identified as het/straight, 1% as gay, 0.5% as bi, 0.5% as 'other', and 3% 'don't know' or refused. (Where did that "6%" come from? I can't see it anywhere in the report.)

It then goes on to discuss some reasons why people may have answered 'other' and 'don't know'/refusal. It specifically notes that non-response was not assumed to be heterosexual.

[sparkindarkness.livejournal.com]

Sorry, that they was a reference to the media reporting, not the ONS

[lederhosen.livejournal.com]

The ONS don't have a good reputation these days, but FWIW, 'identity' issues are notoriously difficult to measure. I worked on an identity-based survey a while back, and the rate at which people do/don't identify fluctuates tremendously depending on just about any factor you can think of.

In particular, it's known to be an issue (http://en.wikipedia.org/wiki/Men_who_have_sex_with_men) with sexuality: there are a lot of men who have sex with men, but identify as heterosexual. (It's not just a Baptist preacher/Republican congressman thing.) Which is why a lot of sexual health educators/researchers prefer to refer to "men who have sex with men" rather than "gay/bi men".

I suspect ONS are undercounting, but I would guess that a fair chunk of those 2.2 million account-holders on gaydar.com (or however many after eliminating duplicates) don't identify as 'gay/bi'.

ONS Measuring Sexual Identity

[psycentral.livejournal.com]

The ONS survey is so deeply flawed. The authors freely adnit that one question is not sufficient as a measure, but went ahead and did it anyway. They also admitted that had they gone for more appropriate, more complex measure then the figure would have been higher.

So at best it's a pilot study showing what not to do next time, although they will probably not learn from the feedback and produce lots more worthless stats.

http://psycentral.wordpress.com/2010/09/24/sexual-identity-figures-flawed-psychology/