A bad recipe

Well, the local elections are over in Montana, and I’m sure there will be some controversial results, even if there are not any. I mean, it just seems that somebody’s got to find fault with something even if it’s imaginary. I’ve been thinking about this column for over a year, ever since the Montana Election Integrity Project (MEIP) came out with their “canvass” of 4347 Montana voters to show how screwed up our elections are. (Go to mtelectionintegrity.org/canvas) It’s taken me that long to write this because I just couldn’t figure out where to begin because there was so much wrong with it from a scientific standpoint.

Based on their canvass of those 4347 Montana voters, MEIP figures there were 120,261 what they call “invalid” votes in the 2020 Montana election and that’s not to mention some 196,370 “invalid” voter registration files. They talked to those 4347 voters about 13 months after the 2020 election and based on what the voters told them got the facts, or at least the facts as the voters could remember them more than a year after they voted. MEIP compared the voters’ answers about when they sent their ballots in with the database that the Secretary of State keeps on ballots received and if the answer was different from what the SOS had, then it was proof positive that there was dirty work afoot. 

MEIP volunteers went to doors in selected areas of six counties. They don’t say how the areas were picked, much less how they picked the counties. In fact, their lack of detail in such things as that is what got me in the first place because—and don’t hold this against me— I was trained as a scientist and am familiar with what’s known as the “scientific method”. It’s a fancy name, but it’s got a lot in common with a cake recipe. The scientific method lays out the guidelines for conducting an experiment or survey. That’s because one of the rules, if you’re going to be a scientist, is that what you say you find has to be able to be verified by other scientists using the exact same procedure. In the write-up of an experiment this is the section called “Methodology”. In baking a cake it’s called a recipe, and if you follow the recipe accurately you should come up with a cake that is just like—or almost, anyway—the cake the recipe writer made. Likewise in science. Leave something out or put something in and you get different results.

Well, MEIP used words like methodology, but they were pretty vague on details, sort of like my mother’s recipes that called for a dollop of this and a dash of that. So, my questions are why and how did they select where they canvassed. Could it be because they had a need to show the canvass would come up with the results they wanted? In their Methods paragraph they say: “The Canvass consisted of knocking on doors of both voters who we considered to have at least one suspected issue associated with their voter data, as recorded by the Secretary of State’s office, and voters for whom we did not show any suspected issue.” Whatever those “issues” are goes unsaid, but this is looking to manipulate the outcome by screening the participants beforehand. They find that 1138 voters out of the 4347 had what MEIP calls “invalid” votes. By extending those results to all Montana voters they come up with their conclusion that there were the above listed number of “invalid” votes. 

The problem is there is no way of telling if those 4347 folks were representative of the rest of the 2020 voters. If they are, well and good, if they weren’t, you can’t extend the conclusion from the small group to a larger one.

Why am I concerned about this? The vote is the most important right in our Republic and spreading baloney about “invalid” elections with inaccurate science is purposely undermining voter confidence in our elections. 

People forget, people mis-remember, people lie. As a losing candidate once told me, “If all the people who said they voted for me actually had, I would have won.”

Jim Elliott served 16 years in the Montana Legislature as a state representative and state senator. He lives on his ranch in Trout Creek.