No sex in the census

ranging from straight to gay to bisexual to lesbian to other, will be in the 2018 General Social Survey that gathers data from 8000 inperson interviews.

There is a belief that interviewers add context to questions, which makes data more reliable. The downside is that 8000 is a much smaller sample size in which to capture the nuances of sexual orientation and identification.

Concern about erroneous or inaccurate data is valid although we already see confused or inexact answers in areas like religion and ethnicity, which are also considered private or politically contentious. In the 2001 census, 53,715 New Zealanders gave ‘‘Jedi’’ as their religion although only a handful are likely to have wielded a light sabre.

There has been a parallel disagreement about whether the word Pa¯keha or the phrase ‘‘New Zealand European’’ best describes white New Zealanders. The ethnicity category has aspects of subjectivity and self-identification that are very similar to questions of gender and sexuality.

The closest the census has come so far to the sexuality question is finding out, in the 2013 census, that 16,660 people lived in same-sex couples. Of that group, women outnumbered men.

LGBTI activists are correct in saying that without census data, there is no way of knowing exactly how numerous their community is or where it is located. Can policy and resourcing decisions be made without a precise sense of social need?

The New Zealand Attitudes and Values Survey, run by the University of Auckland in 2013-14, asked 18,000 respondents for their orientations and learned that among men, 3.5 per cent are gay, 1.5 per cent are bisexual, 0.4 per cent are ‘‘bicurious’’ and fewer than 0.1 per cent called themselves ‘‘asexual’’. For women, 1.8 per cent are lesbians, 2.1 per cent are bisexual, 0.7 per cent are ‘‘bicurious’’ and 0.4 per cent are ‘‘asexual’’. That makes around 95 per cent of us strictly heterosexual.

The university allowed respondents to choose their own label rather than respond to a list, as in a census.

Would the numbers be similar or different in a full, national survey?

We won’t find out in 2018.