Democrats
Severe Undersampling of Quinnipiac University Poll That Shows Trump Down by 15 Points.
| July 15, 2020 – Biden Widens Lead Over Trump To 15 Points In Presidential Race, Quinnipiac University National Poll Finds; Trump Job Approval Rating Drops To 36 Percent |
Quinnipiac University Poll July 15 PDF
Quinnipiac University Poll Sampling Data
Skewed polling was a significant problem in the 2016 election, and it looks like the pollsters still haven’t learned their lesson in 2020. Just a brief look at the sampling data from a July 15 poll from Quinnipiac University bears this out.
In this poll, only 24% of the sampled Registered Voters were Republican. This means a full 76% of those polled were not evenly matched with the populous. How can this poll be considered non-biased?
However, this still didn’t stop the University of Quinnipiac from stating, “Registered voters back Biden over Trump 52 – 37 percent, according to a Quinnipiac University national poll released today.”
Here’s the data the pollsters used to come up the “data” that Biden is ahead:
PARTY IDENTIFICATION QUESTION WORDING – Generally speaking, do you consider yourself a Republican, a Democrat, an Independent, or what?
REGISTERED VOTERS PARTY IDENTIFICATION:
Republican 24%
Democrat 34%
Independent 34%
Other/DK/NA 8%
According to a sampling of voters from Gallop just over a month ago:
Republican 25%
Democrat 31%
Independent 40%
Other 4%
As you can see, Republicans were under sampled. Independents, who usually trend conservative were also under sampled. The voters in the “Other” category were also skewed by 35. Finally, the Democrat sampling showed an oversampling.
The polled data showed Republicans and Independents sampled at 58% while that nationwide total as indicated by Gallop showed these two groups at 65%. A total of 7%. When combined with the “other” category there is upwards of a 10% shift in total voters that should have been correctly vetted in order to present a fairly managed poll.
Here at Citizen Media News we can’t speculate as to why Quinnipiac would publish a poll with gaping non-representative samples, but it should serve as a caution that in this election year, yet again, the pollsters haven’t learned a thing.
