Response to the argument that "Sanders voters" put Trump in office. [View all]

[Ken Burch]

In statistics, there are two crucial terms-causation, and correlation.

They do not mean the same thing.

Correlation is when two facts simply occur at the same time, or in sequence, without being related.

Causation is when one fact is actually the result of another fact.

The assertion that "Sanders voters elected Trump", in an online article released Wednesday is correlation without causation.

In three swing states, a certain number of voters happened to cast their votes for Bernie Sanders in the primaries, then happened to vote for Donald Trump in November. This is an empirically factual statement. However, there is no actual demonstrated relationship between those two facts.

All we can say is that the survey identified a group of people who happened to pull the lever for two different presidential candidates at two different times in one election yeat.

We have no evidence for any conclusions about these people beyond that.

There's no evidence that these voters were politically active, or were even in any way involved in the Sanders campaign-or, for that matter, the Trump campaign. Sanders voters and active Sanders supporters were entirely different groups of people.

There's no evidence, and there is no reason to surmise, that the Sanders campaign did anything to cause these voters to support Trump in the fall, or that that campaign wished for those voters to do so.

And there is nothing in these two electoral occurrences that offers any evidence for the argument that these same people would not nave shown up in November in equal numbers to vote for Trump, throwing the Electoral College his way, if only the Sanders campaign had not been allowed to enter the Democratic primaries.

The likeliest surmise is that this group voted for the candidates they preferred, in the spring and in the fall, as an expression of alienation, not on ideological grounds, and not out of willful group intent.

It's comparable to the well-documented fact that a large number of those who voted for George Wallace as a third-party candidate in the 1968 presidential election had been supporting Robert Kennedy for the Democratic nomination and would have voted for him in the fall if he had lived and been nominated.