Only because they're such a small group relative to the other categories reported! The wording of the article is poor; it gives the results in a way that represents population size rather than the likelihood that any given persoon will cheat on their spouse. Also, the survey's methodology is poor (a self-selected sample is not a random sample, and therefore is unlikely to provide accurate results).
98.76543210% of all statistics cited in internet articles are meaningless, made-up BS (including this one).