Facebook

My study is interested in the life experiences of gay and bisexual men in the UK. In order to get as many respondents as possible, it is important to advertise.

One part of my campaign to attract people to my research questionnaire has been to place an advertisement on Facebook. It is aimed at those who identify as gay, are UK residents.

The potential reach of such a campaign is 191,500 men. In order to keep costs down, on a pay-per-click basis, I’m allowing a maximum of £10 per day and paying up to 40p per click. That translates to 25 people per day clicking through.

Online campaigns like this, I am told, get an average of <0.3% of people clicking on an ad they have been exposed to. As such I’m anticipating about 500 or so responses via Facebook.

Something I had not expected is that those who do ”click through”  do not automatically go on to complete the survey. This means my real cost per respondent is about £1. That in itself is not an issue.

What is an issue is that most of the respondents are under 25. It is of course only to be expected that Facebook is the domain of the young, and if need be my dissertation can comfortably be about the experiences of young gay people. But… that does make things a bit more difficult. By virtue of their youth, they will not have had the life experiences that my hypothesis needs to discriminate respondents by.

Under 25s are less likely to have postgraduate qualifications; biological children, or be in a civil partnership. As I want to explore how life experiences affect behaviour it is something of a handicap.

Still, I can always turn it to my advantage; compare under-25s with their older peers, for example; or focus on the main issues that they raise such as homophobic bullying in schools.

I suppose the non-question I’m posing is how to rebalance responses so that the younger gay male isn’t disproportionately represented.

I have had a number of flyers printed up and I’ve made some progress handing them out in London and Cardiff. I think I need to do more of the same. I need to think about where, of course, although anything that generates responses as an absolute good.

I suppose the real question is does the age profile of respondents matter? Can my research work around a younger demographic than I was expecting? Right now I don’t know. But it’s definitely something to think about.