> 2016 US voting age population was approx. 250 million, so it’s actually 25%. Why do you exclude people who are not of voting age? > Statistically, unless yo

> 2016 US voting age population was approx. 250 million, so it’s actually 25%. Why do you exclude people who are not of voting age? > Statistically, unless you have specific data that shows otherwise, we should expect non-voters to have (mostly) voted the same way was actual voters You can't be serious. This is flatly false, and has been throughout the entirety of the period of modern polling methodology. Non-voters are dramatically less likely to support the government at all, let alone a particular leader. Consider also that many of the populations who are unable to vote (felons, incarcerated people, undocumented immigrants) are dramatically underserved by government institutions and are typically not polled in any of the presidential approval polls. Why in the world do you suggest, in the absence of any evidence, that they are likely to have a similar degree of likelihood of supporting Trump?