The advent of publicly available scientific polling in 1936 is one of the great accomplishments of modern social science. Through it, we can obtain good general information on the state of public opinion, as well as the views of subgroups through what we call crosstabs. These effectively break down larger samples into smaller parts; by looking only at, say, black voters, we can discern their views and, if significant, compare them to those of white voters and other subgroups.
But one problem with this approach is the lack of controls. At the end of the analysis, we can’t tell if black voters…
Originally found on Real Clear Politics Read More
