Acadamics say polls underestimate both the size of gay population and anti-gay sentiment

Illustrated rainbow pride flag on a white background.

Researchers have called into question the accuracy of surveys which detail the size of the gay population and people’s views on sexuality, suggesting they should be done anonymously in order to be more accurate.

A working paper, published this month by researchers from Ohio State and Boston Universities, showed non-heterosexuality and anti-gay sentiment are substantially underestimated in existing surveys.

The researchers from Boston say: “It is perceived as socially undesirable both to be open about being gay, and to be unaccepting of gay individuals”.

By making the process anonymous, they believe the data collected will be more accurate.

They called this technique the “Veiled Report” method and asked questions about sexuality and LGBT issues to over 2,500 anonymous participants.

The experimental method increased the rates of anti-gay sentiment. When asked if they disapprove of having an openly gay manager at work, those in the Veiled Report group were more likely to answer ‘Yes’ than if they were asked directly.

It also increased the likelihood to be honest about non-heterosexual experiences. The share who have had a same-sex experience increased from 17% when asked directly to 27% when asked through the Veiled Report method.

The Pew Research Centre, an American think-tank, says the study does not “attempt to draw its own conclusions about size of the LGBT population or public attitudes about it since the participants were not a random or representative sample of all adults 18 and older”.

Earlier this month, the Office for National Statistics found only 1.5% of adults in the UK identify as lesbian, gay or bisexual.

The ONS’ experimental data was criticised by the data editors at the Guardian and Twitter because of its social desirability bias: the tendency for people to not to reveal behaviours or attitudes that they fear may be viewed as outside the mainstream.