Firstly the report is very clear what its sample method is. From page 1:
"Sampling approach. Research was conducted with gay, lesbian, and bisexual supporters of LGB Alliance. LGB Alliance has a large supporter base with diverse political views, working in various sectors across the UK"
Not only did they emphasise the survey wasn't random, they actually vetted responses to ensure submissions met a preselection criteria.
The reason why this report is unique is that it's, in effect, impossible to collect information on gay men and lesbians because trans people identify as gay men and lesbians when in fact they are heterosexual. And they are often highly emphatic about that designation. It is less of a problem for gay men, because homosexual men still predominate and there are relatively few women who identify as gay men, but they would still distort the statistics.
With respect to lesbians, however, because if the high portion of transwomen who are heterosexual, potentially more than half the respondents of a survey of lesbians could be transwomen, many of whom insisting that their sex was female.
So, as an example in the unvetted world of MN, it could well be that you are in fact a heterosexual male identifying as lesbian, and the proffered anecdote that you "asked to join the group and I was told no, not because I’m a lesbian but because they already had a representative from my area of the business", would be totally irrelevant for the purposes of this survey, which was to gather the opinions of female homosexuals.