Thanks @ellaanais and @josh_nz for discussing how the data on sexual orientation might be used. As Josh points out, how someone identifies might not be reflected in their behaviour. Research around sexual orientation often uses 'sexual orientation' as an umbrella term, under which are: sexual attraction, sexual identity, and sexual behaviour - that is, three things make up a person's sexual orientation. Is this framing useful?

I would say @LisaAtStats that it is useful in general yes, but if there was a way that respondents could almost tick boxes as i'd mentioned and/or indicate (similar to a 1-10 scale) where they may fall I think that would be a better way StatsNZ would be able to get a thorough and well-dimensioned output. One that not only New Zealand can utilise, but would be extremely beneficial to countries and organisations worldwide. Not since Kinsey has there been a serious attempt at gauging sexuality, and that was male-focused, and I believe there is plenty of time between now and 2018 that StatsNZ would be able to word the question and answer so succinctly that responder error/omission (whether deliberate or accidental) is minimised.

If a digital platform is available by 2018 either wholly or partially, there is further opportunity to assist in anonymising data whereby a respondent would complete an exercise online for a particular question or set of questions, and is given an alphanumeric code to put as a response on the census form, giving further assurance that any member of the public (or any family member) would not be able to extract or assume the content of information/answers given.

Kinsey's work on men's sexuality was released in 1948. He followed than up with another, just as important report on women's sexuality in 1953 @Josh_NZ . 1 to 10 is probably too wide a scale - 0 to 6 like Kinsey should be enough. Maybe with the addition of other variables for asexual people.

The Ministry of Health in their New Zealand Health Survey is collecting data on the three aspects of sexual orientation. Data on Sexual Identity will be available mid December, with data on Sexual Attraction and Sexual Behaviour presumably some time next year.

that sounds interesting, lisa! can you talk a bit more about that, or point me to some links?

Kia ora, Bridget from Ministry of Health here.

As part of the 2014/15 NZ Health Survey we are running a Sexual and Reproductive Health Module, similar to the UK's National Survey of Sexual Attitudes and Lifestyles (NATSAL). The Sexual Orientation section covers: sexual attraction: ever, sexual attraction: in the past year; sexual experience and sexual identity.
So you may like to look at the NATSAL infographics downloadable here http://www.natsal.ac.uk/home.aspx, or at the UK's Integrated Household Survey's Sexual Identity infographic: http://www.ons.gov.uk/ons/rel/integrated-household-survey/integrated-household-survey/january-to-december-2012/info-sexual-identity.html .
Please note we've not committed to any publication time-frames for the Sexual & Reproductive Health data.

From 2015/16 onwards Sexual Identity will be part of the core NZ Health Survey - that is, the question will be asked every year.
Finally here's a link to the Health Survey publications (nothing on sexual orientation yet): http://www.health.govt.nz/nz-health-statistics/national-collections-and-surveys/surveys/current-recent-surveys/new-zealand-health-survey?mega=Health%20statistics&title=NZ%20Health%20Survey

Regarding the points made in the discussion document: there seem to be two major points: [1] it hasn't been done before and [2] the data may not be reliable.

The fact that no other statistical agency hasn't managed to get this question onto their censuses doesn't mean NZ shouldn't be first. It just means it's difficult to do.

Ethnicity, descent and religion are all difficult too - and the data from those questions are notoriously unreliable for certain purposes as well. Those questions are retained because, imperfect though the data may be, they give us all an indication of make up of our country.

With questions on sexual orientation already included in the NZ Health Survey, the difficulty with creating a classification standard suitable for the census seems overstated to me. Given the amount of work already done here and elsewhere it should be possible to create a question and a statistical standard classification.

Regarding acceptability to people answering the question - a "refuse" box would cover that concern. In any case the UK Integrated Household Survey report on Measuring Sexual Identity (2010) found (1) no reduction in overall response rates to the survey when the question was or wasn't included and (2) a greater than 96% response rate to the question where it was asked. That's a better response rate than the income question gets.

New Zealanders will get used to a question of this sort the more often they are asked it, and the more normal and everyday it becomes. The sooner we start asking, the sooner we'll get to that state.

• Richard Arnold

@richardarnold that's a great response! Couldn't agree more!

Just because its hard doesn't mean we should do it and we do for other questions so it is possible.

This is the classification of Sexual Identity that's used in the UK and in the NZ Health Survey:
1 Heterosexual or straight
2 Gay or lesbian
3 Bisexual
4 Other
.K Don’t know
.R Choose not to answer

The Ministry of Health is collecting all three aspects of sexual orientation (sexual identity, sexual attraction, and sexual behaviour) in a Sexual and Reproductive Health Module, and then going to collect Sexual Identity every year as part of the core Health Survey.

People have discussed many ways that sexual orientation data might be used. Is it necessary to collect all three aspects of sexual orientation, or can one (or two) aspects provide the information needed?

Thanks @bridgetmurphy and @richardarnold for your great contributions to this discussion, and welcome also to @karinfromuranus.