Voices: Election signals trump noise

Alexander Ooms argues noise – rather than rigorous analysis – was given too much attention this election season by the media and elected officials. Cross-posted on Ooms With a View.

In my opinion, the biggest unheralded winner in Tuesday’s election was Thomas Bayes. Now Bayes died back in 1761, but his influence lives on in Bayes’ Theorem, which is one of the tent poles of modern statistics and probability. His modern incarnation is Nate Silver, proprietor of the 538 blog, which devised an aggregated polling and economic model to predict elections, and called 49 of 50 states correctly four years ago. Silver’s statistical approach has drawn many detractors, but to paraphrase MLK, the numeric arc of elections is long but it bends toward probability theory.

For the last few weeks, the noise around the election became louder and louder; at the same time, the signal on 538 became more and more clear. The long march of history is that data and signal triumph over noise, and by the early hours of the evening, it got pretty quiet. Silver pretty much nailed this one too.

There are few specific education items in this election worth much comment, but there is a lesson here and it is a simple one: Signal Beats Noise.  This is poignant for education, as many debates are still dominated by the noise of a vocal and shrill minority, a discordant clamor which captures the attention of far too many legislators. Education still, and almost uniquely among major topics, has fact and data buried by noise.

Denver 3A/3B prime examples

Nowhere was this more evident than in the false ballyhoo around 3A and 3B. Denver has a long and consistent history of supporting similar measures, and there was never any indication that the basic signal had changed. There was a discordant noise – formed by the asynchronous demographics of both anti-tax tea party Republicans and a sliver of the status quo Democratic base, and amplified by media. Loud and shrill, these voices were given far too much attention: their claims rarely passed any analytical review, and if anything, treating them as serious opponents gave them a credibility that they should have lacked. Any doubt about the ability of opponents to generate support might have been quickly discredited when campaign finance reports showed their impact, as the No on 3B campaign finished with all of $1,890 in contributions. This was noise.

But realize as well that any claims of credit for the victory on 3A and 3B by the politicians who supported them is likely an indication that the rest of their education agenda is strikingly shallow, and is roughly akin to celebrating the sun’s rise this morning over the eastern horizon. There should be little pride in beating back something that was never a threat. Delivering a bond and mill in Denver is simply not much of an accomplishment.

But the Denver school board has, in the past year, shown a proclivity to pay an overt amount of attention to noise instead of increasingly clear data about which schools and policies are having an impact. There are lots of education issues where ignoring the noise and focusing on the signal will be critical, including some that will need attention in the coming months. And if there is one thing increasingly clear about DPS and the Denver school board, it is that both sides now celebrate noise far more than they acknowledge signal.

There was not too much more of interest for education wonks. Bond measures across the Front Range did very well – this unanimity comes as somewhat of a surprise given the failure last year of Proposition 103. But Prop 103 was amorphous and unstructured, dollars thrown into a dark hole while bonds issues, by their nature, are specific to the point of dullness. So the hypothesis that can sit in the intersection of both defeat and victory – that voters will support additional dollars for education as long as they believe in what these dollars will buy – may still hold true.

Individual results were largely predictable. While it was no surprise, I was glad to see Brandon Shaffer soundly defeated, even though I am no fan of his opponent.  Shaffer’s consistent opposition to sensible education legislation – culminating in a cowardly move to send last year’s well-supported and bipartisan literacy bill to a kill committee – has been a prime example of noise over signal for years.  I doubt many tears hit the floor when his results came in.

One of the closest races of the night was Evie Hudak in the State Senate. A former member of the Colorado Board of Education, Hudak has usually taken fairly traditional stances on most issues, such as her opposition to SB-191, the bill that changed teacher evaluation. However, even Hudak has shown some flexibility recently, and she voted to support the early literacy bill last year.

If there is an unrecognized bright spot in the election for education reform, it may be the ascendency of Mark Ferrandino to speaker of the Colorado House. Ferrandino is the son of schoolteachers, and an ardent believer that our education systems need change. He has supported much of the legislation at the heart of education reform, and he will ensure that issues are debated on their merits. I don’t agree with him on everything, but his positions are well-considered – they are signal, not noise – I deeply respect his views, and think he will make Colorado proud.

About our First Person series:

First Person is where Chalkbeat features personal essays by educators, students, parents, and others trying to improve public education. Read our submission guidelines here.