The argument states that the Leevile citizens prefer literary classics as reading material.
No, the argument does not state this. One study found that Leeville citizens reported this. In light of their actual behavior, the argument is that Leeville citizens misrepresent their reading habits, claiming to prefer literary classics while actually preferring mystery novels.
Thereby, it is completely illogical to conclude that the respondents in the first study have misrepresented their reading habits.
It's you who are leaping to wild conclusions. What evidence do you have that university researchers surveyed only older citizens and that, furthermore, those older citizens are not users of the public libraries? In fact, I believe that research indicates that older adults are more likely than younger adults to use public libraries.
Here's what I'm trying to say: The findings of the study depend on whether the researchers used standard research procedures. Standard research procedures would be, when studying the reading preferences of a population, to ensure that a random sample of the population is surveyed, with care taken to make sure that there is not a skew in terms of age, income, etc. The chances are high that a university research team took such precautions. Therefore, one may raise the question of whether such precautions were taken, but relying for one's argument on the slim possibility that university researchers did not follow even the most basic research guidelines seems ill-advised. Certainly, your adamant tone is ill-considered.
What, then, might be the flaw in the argument? Assuming that the first study was well designed and therefore did accurately reflect what Leeville citizens say they prefer to read, the question becomes whether public library borrowing habits reflect the preferences of the entire population. Probably not. Public library users tend to be less affluent, and also heavily weighted age-wise, with older adults and children/adolescents using the libraries more than young and middle aged adults.
Therefore, to get a true sense of what citizens really read, bookstore sales would also have to be taken into account. But here there is another problem, as more and more citizens -- particularly young, tech-savvy citizens -- buy online rather than shopping locally.
Nonetheless, the conclusion is probably true, unless Leeville is very different from the rest of the country, housing a concentration of literature lovers. Nationally, mystery novels out-sell literary classics many times over, and the likelihood is that a study of Leeville buying habits would not differ from that.