Agree. Moreover, surveys are not to be trusted for such small numbers. Plus, how can be statistically significant a sample drawn from a single subset of population observations (Chicago) and being considered representative of the US population? That is odd at best