California May 2009 Special Election Results – What do they imply about voter education?
This is not on topic to my blog as a whole, but I’d like to put it out into cyberspace and see what happens. As I don’t currently have another forum for this, here goes!
I was looking over the election results for yesterday’s special election in California (you can see them by Clicking here) and noticed something that disturbed me.
Short version is that I think roughly 3% of the Californians who voted yesterday made their decision on at least 2 of the measures solely on the basis of the title of the measure! Long version follows…
Take a look at the number of “Yes” votes for Propositions 1A and 1B. You’ll see that 1B had significantly more votes than 1A – and if you go to the county by county results, you’ll see the pattern occurred statewide. Why is this weird? Proposition 1B was written to be contingent on the passing of 1A. If 1A doesn’t pass, then 1B will have no effect, even if it passes!
When we consider the possible permutations of votes for 1A/1B – We have: N/N, Y/N, N/Y and Y/Y. (Yes, I know I’m excluding the cases where a voter cast a vote in only 1 of the races – more on that later.) For each combination, what can we discern about the will of the voter?
N/N and Y/Y — Honestly, we can’t discern much. Either pattern can result from a multitude of factors and it’s impossible (without significant exit polling) to guess at the will of the voter.
Y/N and N/Y — Voters with these patterns have, at the very least, paid a little bit of attention to the specific issues in each proposition, either by recommendation of someone (or some organization) they trust, or by their own reading/research – which could even be as simple as their gut reaction to the title of the proposition.
Since Y/Y and N/N result in the same number of votes cast for each measure, let’s look at the difference:
First significant datum: Proposition 1B, even it it passed, would have no effect if Proposition 1A failed. This was clear in the actual text of the propositions, as well as the ballot summary – not to metion many analysis available through the web, or published as newspaper recommedations.
As of the time I composed this post, 1A had 1,327,400 Yes votes and 1B had 1,452,535 Yes votes, there were 125,135 more votes for 1B than there were for 1A. If we assume that nobody voted Y/N – than there were 125,135 voters who voted against the proposition (1A) that was necessary for the proposition they voted for (1B) to have any effect!
If we remove the assumption in the above paragraph, then each Y/N vote increases (by 1) the number of N/Y votes necessary to achieve the published result, so we can reasonably say that at least 125,135 voters voted in a nonsensical manner!
While I can imagine a voter deciding to vote N/Y after significant research who thinks to herself: “I know that 1B will have no effect if 1A doesn’t pass, but I’m going to vote N/Y to simultaneously: A) Expresses my displeasure that our legislature didn’t do the job (adopting a budget) that we pay them to do and B) Signify my desire that education be funded.” I have grave doubts that there were even 10,000 who researched deeply enough and thought hard enough about the meaning of their vote to come to this decision, much less 125,135.
However – I think the the voter who decided solely on the basis of the titles of the propositions could easily have come to a Y/N decision as follows:
1A “Rainy Day” Budget Stabilization Fund – I’m voting against this – we need a budget now – why are we talking about rainy days?
1B Education Funding, Payment Plan – I think we should fund education, so I’m voting for this
Thus the only way I can imagine the bulk of the N/Y votes happening is through voters who decide on a proposition base only on the title!
What does this say about how California voters (and by extension all voters) decide how they’re going to vote? That at least 3% of the voters made their decision on these 2 propositions based only on the title! I want people making decisions based on more substantial things than the title!
I mentioned above that I’m discounting the people who might have voted on only 1, but not both of the propositions – the undervotes. The difference between the number of votes cast for each proposition (3,882,919 – 3,874,441 = 8,478) was less than 0.3% of the votes cast, as a factor it pales in comparison to the 3% difference in votes observed, so I don’t think speculation about undervotes is worth the time.