More sordid details are leaking out about that New York Times poll released over the weekend, and these details are alarming, to say the least. I want to focus on one particular article written by nwprogressive.org. In it, they point to the problems with this particular poll — starting with the samples of respondents.
Republicans appear to have been oversampled. To quote from the article: “Republican voters were oversampled in these polls.” NW Progressive says the word “oversampled” is nowhere to be found in the WRITEUP of the poll, but “if you open the actual dataset, look through it and read the endnotes,” you will see for yourself they ADMIT to this Republican oversampling.
NW Progessive writes that it is “irresponsible” of the New York Times not to have TOLD readers this immediately. “This is an important design choice and ought to be explained,” they write. Now, readers if you do venture into the dataset, you will find it — hidden away — likely not seen by the majority of readers.
“The survey was weighted by the Times using the R survey package in multiple steps to account for the oversample of republican voters.” Straight from the horse’s mouth itself. Northwest Progressive looked for a REASON why Republicans were oversampled. They could not find one offered.
They guess that the NY Times and Sienna were afraid of “undersampling” Trump support. Then, there is the little problem of progressive respondents. To put it bluntly — there were hardly any.
Poll crosstabs — “75% of the respondents identify as very conservative, somewhat conservative or moderate (which isn’t an ideology),” says nwprogressive. According to this article, which is much better done than the junk poll of the New York Times, “36% identified as conservatives, nearly twice as many as those who said liberal.”
It kind of puts things in perspective, doesn’t it? Furthermore, this article discusses how the “New York Times’ newsroom and data team thought Republicans would be “very dominant in the midterms.” (last year.)
“But the red wave that the NYT’s people thought was coming did not materialize.”
They also say: “Our team doesn’t think it’s possible to make even an educated guess about who’ll show up at this juncture, not even with the best available software, sources of data, and poll design expertise.”
It is an excellent article that you should read. I also am disgusted by the blatant deceptiveness of the NY Times. Readers were not told the truth. The very first sentence should have been that Republicans were oversampled.
This article also rips apart the media, which they say is “failing Americans at an incredibly critical juncture for American democracy.” Every word of this article is the truth. I wish I could say the same about the NY Times.