This is an interesting question (and I'm not sure that the poll will really help us answer it).
My guess would be that there is a large gap between people without a HS degree and people with a HS degree (for many reason, including the fact that most actual illiterates would be in the non-HS degree group). There may be a small gap between HS degree holders and bachelor's degree holders. I suspect that there is little - if any - difference between people with a bachelor's degree and a more advanced degree.
As other posters have suggested, at least for people with a bachelor's degree and beyond, what the person studied is probably more relevant to their reading than the actual degree. I.e., the reading habits of a person with a Ph.D in English lit are likely to be more similar to the reading habits of a person with an BA in English lit than they are to the reading habits of a person with a Ph.D in Finance. (And the person with the Ph.D in Finance will probably have similar reading habits to the person with the BA in Finance).
I would be interested in seeing data that tracked the educational level of various readers based on specific books, too.
Re: most literate cities - I like the idea, and there may be a way of making a decent ranking like this...but I don't think that this ranking succeeds in doing so. The study is based on six factors: newspaper circulation, number of bookstores, library resources, periodical publishing resources, educational attainment, and Internet resources. But I'm not sure that many of these are good proxies for literacy.
1. Newspaper circulation - in 2011, I don't think that this number is meaningful at all. None of the literate people that I know subscribe to a newspaper.
2. Number of bookstores. This probably has some validity...Amazon notwithstanding, bookstores won't survive without literacy.
3. Library resources. There is probably some correlation between library resources and literacy, although "resources" is a kind of weasel word and may obscure more than it reveals.
4. Periodical publishing resources. I'm not exactly sure what this means - I suspect it has to do with magazines being published in that city. I don't think that this is a proxy for literacy at all.
5. Educational attainment. This is probably roughly correlated with literacy, although perhaps not above the bachelor's level. But a city with 15% college grads is probably less literate overall than a city with 45% college grads.
6. Internet resources. Again, I'm not sure what they mean by "resources," but probably an area with only dial-up would be less literate than an area with broadband. However, I don't think that having 25/5 mbit service would make a place more literate than a place with 10/2 service.
Another statistical problem with this type of survey is that the variables are not independent - locations with higher educational attainment will also be areas with higher income and thus greater internet and library "resources." In fact, I would be curious to see how the rankings would change if they only included educational attainment. I suspect not much.
But there is an easier way to find this kind of information out - just conduct a statistically valid poll of a representative sample of the population, and then rank the results.