Fake News: AP Touts Anti-Trump Poll with Warped Demographics Conducted by Soros-Backed Org

Unfortunately, the author doesn't seem to understand much about survey methods. Probably because that would blow his "story" out of the water.

Just because a particular group is oversampled doesn't mean that its responses skew the final results. If, for example, African Americans make up a disproportionate share of the sample, those results are adjusted to reflect the ratio of the actual population. Thus, even though participants from one demographic category may outnumber their actual proportion of the population, their results are weighted according to the actual proportion.

"A poststratification process is used to adjust for any survey nonresponse as well as any non-coverage or under- and oversampling resulting from the study-specific sample design. The poststratification process was done separately for each racial/ethnic group and involved the following variables: age, gender, education, and census region. The weighted data, which reflect the U.S. population of adults ages 18-30, and the 18-30 year-old populations for African Americans, Latino/as, Asian Americans, and non-Latino/a whites, were used for all analyses."

The results of the survey seem pretty reasonable from what I've seen from that age cohort.

 

Apparently you didn't read the notes on methodology. Feel free to make a case based upon the methodology. Even though you oversample doesn't mean that the oversampling shows up in the final results. Really, this is not rocket science. Except for folks who don't like the results.

I'm not sure what "a particular ground" is.

 

Yes, I learned a lot of things. It is possible to lose horribly in the popular vote and still win the presidency. You can alienate large sections of the voters and still win. You can boast of groping women and still win. You can tell blatant lies and still win.

As to polling, late-breaking votes in a handful of states make a mess of state-by-state surveys. Yep, that happens. DId they grossly missample? Probably not. The results were paper thin in some states.

Which has nothing at all to do with this particular survey. Again, the results correspond roughly to what I see among people of that age group.

 

What, exactly, does any of this have to do with the OP? Many, if not most, young people consider Trump an illegitimate president. Yeah, there are all kinds of problems with that view. He won the race as it was designed. And way too many of them are brought up to believe that anything they don't like is illegitimate. That's another story. But the fact is that that many, if not most, of them think Trump is not legitimate. Cavil all you will, this is a troubling fact.that bodes ill for the future of the republic.

Besides, the state polls were not "wildly wrong." And apparently winning the popular vote was a big deal for Trump since he claimed that he would have won the popular vote if it hadn't been for so many illegal votes.