3 ms·
Although I broadly agree with this, I think you're missing the point that statistical characteristics of particular demographics can be very important. For exa
by PaulDavisThe1st 4y ago
Although I broadly agree with this, I think you're missing the point that statistical characteristics of particular demographics can be very important.
For example, let's say that in fact roughly 30% of women are support a particular candidate, and that this number will not vary by much more than 5% in either direction. Given that women are more or less 50% of the population, that means that this candidate will need to make up a fairly significant vote deficit by over-appealing to some male demographic.
So yes, not all women or Latinos or African-Americans vote the same way, but if, statistically, enough of them do, then you can make some broad observations about candidate strategy that are likely consequential (and true).