This misses a couple of important points about the survey. The first is that, as Island Global Research themselves report, the sample for BAME respondents is small, consisting of just 87 people. That means that, were this a random sample (which it isn't), there would be some pretty hefty confidence intervals around that figure. The second is that BAME typically excludes anyone who is white, but plenty of white people in the British Isles can consider themselves to be victims of racism, i.e. Eastern Europeans, Irish Travellers, etc. A further point is that there's no useful breakdown of how the two groups answered: it may be the case that a majority of the BAME respondents who reported a perception of racism did so for the two higher categories, while white respondents by and large reported lower categories, in which case the analysis would have to be a lot more nuanced than "Look! White people think it's a bigger problem than Black, Asian, etc., respondents do!"
That alone should act as a caution against leaping to (or insinuating) a particular conclusion about racism on the Island and how it is perceived.
Then there's the fact that accurately investigating the actual prevalence and degree of racism and racist attitudes in a society is notoriously difficult and probably requires much more subtle and sophisticated techniques than Island Global Research were able to employ (which, to be fair, they more or less admit in the introduction).
It's not really buried though. The result is right there on the page dealing with the Isle of Man, clear enough to see for anyone capable of adding three numbers together.
It's not like they consigned the figure to a passing mention within a dense wall of text in section 5.6 of Appendix F.