Making data make sense

Education is all about fads and, every few years, seems to lurch from one to another. Some are simply re-cycling, while others appear novel and even random. One which has been around for a few years now, but looks recently to have found a second wind, is that of "data-driven instruction". Based on the extremely reasonable premise of "what works", this mantra asserts that educators should gather data and then make decisions as a result of that (those?) data.

The two basic approaches are to record what is already happening and then to associate specific outputs with specific inputs. For example, classes on the sunny side of the school have more discipline incidents than classes on the shaded side. Then play with this input to see what happens to the output, so swap sunny and shaded classes and see what then happens to the new sunny and the new shaded classes. If the former goes up in the incident tables, voila, We now have data to inform our decisions.

The other is to record what is already happening, then to make a change and see what results. In this case, we might take several classes of 25 students and increase them to 30 and record any or specific changes.

Of course, a significant moral problem with any of these is that we are using our students as guinea-pigs and while successful interventions are to be applauded, what about the opposite cases?

I remember one example in my past where I was recruited to the school and was assured that it was excellent, highly-performing, academically focussed, happy, blah, blah, blah. None of this was true but that's another story.

I very quickly found the elementary grades were 1 - 3 grade-equivalents below my previous school so clearly there was something about the teaching and learning. Tests scores reflected the privileged background of the parents, and the teachers were complacent and happy with what they were (not) doing. I needed data.

Over the next few weeks, every time I was out and about, I looked through windows and recorded (a) what the teacher was doing and (b) what the students were doing, I had 100 observations at different times on different days so representative. I found that Teacher A spent 94 / 100 sitting at his desk with his back to the students while they were at their desks. Teacher B spent 65/100 on a chair, talking, at the front with her students seated on the mat in front of her. Teacher C was on a chair, front center, with students coming to her with books, work, tasks etc.

What followed was a series of "difficult conversations", not the least of which were disputes over the data. Again, another story.

I recently put in a proposal for program improvement to a local school who are loudly proclaiming "data-driven instruction". The academic director spends all day, every day collecting data and creating graphs and charts which she then proudly publishes. However ... The data reflects nothing but tests. Students are tested more than they do anything else. Teaching and learning does not change, apart from some "teaching to the test (Nickleby anayone?).

This AD is not alone. I have met several like here. Data is safe, change is scary.  Data is essential for decision-making, but if the emphasis is not on instruction, then the data and the whole exercise are useless.

**Please leave your comments or queries below.**