Topic: Newsmax
In his Jan. 6 Newsmax column, John Tantillo complains about "Democratic party bias in polls of how Americans view the impeachment issue, and Donald Trump" in an October Fox News poll showing a 52 percent of Americans favor President Trump's impeachment:
The poll cited that 48% of the sample were Democrats. But the electorate, according to Gallup, consists of 31% Democrats, 29% Republicans and 38% Independents.
[...]
It appears, that all — that’s right, all — of the polls cited in the venerable Real Clear Politics Poll (RCP) average have a bias of at least six points toward Democrats with Independents being underrepresented by at least two percentage points and always leaning more Democratic.
How can this be?
Well, it’s rather simple:
It appears that these esteemed pollsters may be basing their stratified poll sample on this Dec. 4, 2017, Gallup column "Democratic Party Maintains Edge in Party affiliation."
The only problem is that this is no longer the case.
In December’s Gallup Political Party Affiliation poll 28 percent are Democrats; 28 percent Republicans and 41 percent Independents.
TRantillo invoked a New York Post "analysis" of the Fox News poll that Newsmax itself devoted a "news" article to at the same time. But Tantillo linked to Gallup's affiliation poll history, which hasn't shown much change between 2017 and now. And contrary to Tantillo's claim and the Gallup affiliation polling, registered Democrats consistently outnumber registered Republicans, as the Pew Research Center points out:
Gold-standard, nonpartisan surveys have found for decades that more U.S. adults identify with or lean toward the Democratic Party than the Republican Party – whether these surveys take place under GOP or Democratic presidential administrations. That is the finding of two of the highest-quality surveys that use nationally representative data collected through in-person interviews: the General Social Survey and the American National Election Studies. It’s also the result obtained by numerous other reputable surveys that poll Americans by telephone or online using randomly selected samples of adults, including those done by us here at Pew Research Center, as well as those done by Gallup, Fox News, Kaiser Family Foundation and The Associated Press-NORC.
Because Tantillo doesn't understand how polls work, he decides to try and "unskew" the Fox poll results:
For those of us more conservatively analytical, a better metric to use describing public sentiment regarding impeachment is 46% (Take the poll’s 52% response and subtract from it the 6% Democratic oversampling error into account.) A more aggressive investigator would also consider the independent bias as well and then use a 40 percent benchmark — (46 percent minus the 6 percent oversampling against independents.)
In either case, both the 46 and 40 percentages are significantly different from the poll’s major headline of 52% favoring the removal of the president from office.
It seems that "all" those pollsters are correct and not their politically motivated critics.
