In a press release yesterday, our campaign said that previous debate commission polls were conducted with a sample size of 400. I received an email from Christie Taylor, the debate commission’s social media director, who corrected the record and pointed out that the 2016 poll included a sample size of 500, not 400. Our point was that the sample size between polls had changed, and that a smaller sample would have allowed me to qualify for the debate. In any case, I apologize for the mistake.
My website also received a comment from Rusty Shackleford, who raised issues with the polling methodology. Here’s what he had to say:
Interestingly, it is impossible for John Curtis to get 50.17% in this poll if it had 607 registered voters. In order for that to happen, Curtis would need to have received 304.55 votes. How can anyone receive .55 votes in a poll? Same with Allen. In order for her to receive 19.82% of the vote, she needed to get around 120.28 votes. How does someone get .28 votes?! It is even impossible for Jim to get 5.59% without having a fraction of a vote!
Your campaign should ask for the data of this poll to be released to the public so that it can be determined how you, Curtis and Allen received partial votes in a poll that should only produce whole numbers? Either they got their percentages wrong, which means you can get into the debate, or they messed with the numbers as it is impossible for people in a poll to get non-whole numbers. Nothing in the poll says that it is being weighed in any matter, which it should not be in the first place if it was a pure RDD poll.
Seriously, ask for the data to be published and available to the public, or at least to all campaigns! If the 607 number is correct (which it is mathematically impossible to be), then Jim missed out on the debate by three votes! We need transparency in Utah!
Basically, it is impossible for the poll to be correct.
While I appreciate the support, I don’t think this is necessarily evidence of anything nefarious. From my limited understanding of statistics, small sample sizes require some kind of weighting to ensure that the sample is representative of the population at large. It would be helpful, however, for the debate commission to release its methodology and all the poll internals, so everyone can have complete confidence in its findings.
I also want to publicly thank Dr. Kathie Allen, who has written a letter asking for my inclusion in the debate commission’s event.
While I doubt that the debate commission is likely to change its mind, I do want to note that we have four other joint appearances/debates scheduled with John Curtis and Kathie Allen, the first being this Tuesday at Joe Vera’s Mexican Restaurant in Provo at 12:00 PM.
I hope to see you there.