Who’s Afraid of Data in Schools?

It’s hard to have a school-wide faculty meeting, or open an education magazine, without encountering the word “data.”  Once highly idiosyncratic, teaching has rightly become more and more a skill to be learned with outcomes to be measured.  I could not be happier about this move from teaching-as-witchcraft (or worse, merely bad theatre) to teaching-as-skill.

Those opposed to the use of data in schools are legion.  Some teachers say gathering data gets in the way of instruction.  Lots of stakeholders argue that it couldn’t possibly capture the complexity of good schools and teaching.  The most troubling objection is that data isn’t necessary to improvement; it’s somehow just a sideshow for accounting fetishists.

A recent post by Robert Wachter in New York Times puts it like this: “the objections [to using data] became harder to dismiss as evidence mounted that even superb and motivated professionals had come to believe that the boatloads of measures, and the incentives to ‘look good,’ had led them to turn away from the essence of their work…. The focus on numbers has gone too far. We’re hitting the targets, but missing the point.”

Then, quoting a “hard-nosed scientist” and measurement expert about how to improve the helping professions, Avedis Donabedian, Wachter claims he said on his deathbed: “The secret of quality is love.”

There you have it: love.  Retool, everyone!  Make loving schools and the rest falls into place.

Hardly.  Under the rubric of “love,” lots of pretty awful things happened: low expectations, damning “ability” streaming, corporal punishment, whole-language instruction, and on and on. Yes, teachers should be loving: but we know that because ample meta-analyses suggest that teacher-student relationships have a huge impact on student outcomes.  That’s data.

Why We Should Embrace Data

“Count something.” That’s the advice of Atul Gawande, surgeon, writer, health care reformer.  I am fond of borrowing his ideas, and this is no exception.  We ought to be counting more, not less; we ought to be asking research questions in our own schools and classrooms; we ought to be finding the best evidence about the effectiveness of our individual schools and making decisions based on it.

Ask Yourself: Without “Data” or “Information” or “Evidence”…

… How would you know if your classroom or school is functioning at its best?  Rather than gathering data, would it be better to merely guess?  Or perhaps listen to the loudest teacher complaining in the faculty lounge?

… How can you find improvements?  How do you know if your students are being well-served? How do you know if program changes are aiding student learning or the opposite?

Models of Using Data That Are Likely to Work:

– Validity: Measure what is valued instead of valuing only what can easily be measured

– Balance: Create a balanced scorecard

– Insist on credible, high quality data that are stable and accurate.

– Design and select data that are usable in real time.

– Develop shared decision-making and responsibility for data analysis and improvement.

(This list is taken from Andy Hargreaves, sometimes quoted as being opposed to the use of data in schools.)

Imagine This

Using the above criteria, I suggest that each school (or department within) embrace as their operating principles – something they just do as a matter of course – a set of measures.

ImagineColleagues getting together to decide, informed by stable accepted research (yes, it does exist) and professional judgment, what things really matter.  A few possibilities to illustrate: student engagement; preparation for the next step; effective instruction; and so on.

Then, consciously asking: how can we collect data – qualitative and quantitative, in valid and practical ways – to figure out how we are collectively doing across these domains?

Then, embracing an ongoing effort to track and discuss the data to improve programming or approaches.

The Scientist in the Room

I’m fond of saying that the best teachers are scientists in the room, looking for data about how the students are doing. That data could come from a lot of different sources: it could come from examining exams after students have written them; it could come from moderated or parity marking with colleagues; it could come from student surveys; it could come from student exit interviews; it could be routine (brief “stop/start/continue” surveys for students at the end of units). The more conscious we are of the information we base decisions on the better our decisions are likely to be.

The important thing is that we count, measure, and gather information, from a variety of sources to paint the most accurate picture of what we’re doing and how well it’s working.  We need to start now.  And we need to make it part of our regular practice.

We need more data, not less.  It needs to be collected efficiently and routinely, used consciously and conscientiously, for the perpetual betterment of our schools.  People have said the use of data harms children; I argue the careful use of data is the best defence against a complacent or dysfunctional approach to schooling.

If we won’t tolerate the random best guesses of heart surgeons, why would we demand any less from those that teach?

(Photo: JD Hancock)