Non-Objective
Data drives our society. We are computer driven, and it’s impossible to “crunch” non-objective sources. If something can’t be boiled down to numbers, then it can’t be understood. So everything: children’s learning, Presidential candidates, calling plays in the National Football League; are forced through the “black-box” of data collection.
We saw it happen in public schools in the late 1980’s with the beginning of statewide public school testing. I remember committees of teachers taking days out of the classroom, away from their kids, and working on curriculum to “pass” the tests. Those scores became the primary source of data on educational progress; honestly, more important than student grades. Get a ‘C’ or ‘D’ in a class, the kid still passed. Fail the “Student Achievement Test”, and regardless of grades, the kid might be held back, or at the high school level, not even graduate.
And since there was all that “objective data”, the Achievement test results soon infiltrated into teacher evaluations. If kids did well on the tests, then the subject teachers were obviously “good teachers”. And if they did badly, regardless, then the teacher’s “need improvement”. Teachers wanted to keep their jobs, so not surprisingly, they taught kids to do well on the tests. It was in their best interest, and the kids. But was it the “best” way for students to learn; did it prepare them for their future in life work or higher education? It really didn’t matter; the “data” was there, regardless of whether that data was a valid measurement of education or not. And that “data” drove everything else.
Garbage In
It’s still the way public schools work. (Here is Ohio’s “interactive page” with my school district’s “report card”). In most schools, high stakes testing still drives education, even though the ability to pass a test doesn’t necessarily apply to success in future employment. There are a few courageous school districts that “opted out” of the tests, but they took their chances on losing state funding. Because the test results can go on a “spreadsheet”, because it’s easy to see, the “data” becomes the most important “evidence”. What really “good” teachers do, is get the kids through the tests, AND teach them what they need to know for the future. But it’s a lot, and forces them to teach in ways that aren’t necessarily in the kids “real” best interest.
Sabotage
We crave for a way to make decisions based on “data”, rather than experience, or non-objective factors. Football coaches use “data analytics” to decide whether to run or pass, go-for-it on fourth down or punt. Someone, even during the game, is “crunching numbers” in the stadium.
I’m not a “Luddite”. (Named for the bands of workers who broke the machinery that was taking their jobs in the early 1800s. Some workers in France threw their wooden shoes into the machines. The shoes were called “sabots”. Thus came the modern word, “sabotage”.)
And I’m sure that NFL coaches use any way they can to get one-up on their competitors. But it’s important that the “data” that goes into the crunching is valid. Back in 1974, I was taking computer programming, tapping out programs on green screens in “Basic”. We had an expression: GIGO. It meant that if your program was garbage, all you would get was garbage results – Garbage In, Garbage Out. So if the data isn’t valid, or really doesn’t measure what it’s supposed to measure, then it’s GIGO.
Garbage Out
So let’s look at a modern “polling” question, asked quite frequently to provide the “data” to drive who should be the Presidential candidate:
“On a scale of strongly agree to strongly disagree, respond to the following question – Joe Biden is just too old to be an effective President.” (NYT).
Joe Biden is eighty-one years old. He has all sorts of physical problems that old men have, including spinal arthritis, acid reflux, A-Fib, peripheral neuropathy, and seasonal allergies (WH). He looks, walks, and talks like an old man. Certainly all of those issues impact his ability to be President. But does that mean he is or isn’t “effective”? That’s a totally different question.
If I were taking that survey, I would “mildly agree” with the question. But the answer to that question doesn’t “drive” my answer to the decision all Americans have to make: Trump v Biden. Yes, Joe’s old, and he’s for sure going to get older. But it’s Biden versus Trump, and I’ll take Biden every time, even if he’s “too old”. There’s no other practical choice to stop Donald Trump; only a couple of years younger, and already exhibiting his own symptoms of old age.
Who’s Talking
And there’s a lot of other questions about the “polling” that’s driving the argument to “drop Biden from the ticket”. As Biden himself reiterates, the polling has been wrong, over and over again. In the past the polls under-estimated Trump support. But they also under-estimated strength of pro-choice voters. And since polling is not just counting numbers, but “crunching” them through a model of what the pollster “thinks” the electorate looks like, it’s easy to miss “closet” Biden or Trump voters.
In today’s highly polarized society, where even a casual overheard conversation or a bumper sticker can result in confrontation, how many people are willing to “tell where they stand” to the pollsters coming up on their cell-phones? How many are keeping their views on the “down-low”, ignoring the repeated text, email or phone messages to “participate in a poll”.
Sure polling data is “all we’ve got”. But garbage-in, garbage-out is still true. Just because we have “data”, doesn’t mean it’s accurate or meaningful, and doesn’t mean we should make decisions based on it.
This decision is too important. It has to be right.
