The path of least resistance and least trouble is a mental rut already made. It requires troublesome work to undertake the alternation of old beliefs. Self-conceit often regards it as a sign of weakness to admit that a belief to which we have once committed ourselves is wrong. We get so identified with an idea that it is literally a “pet” notion and we rise to its defense and stop our eyes and ears to anything different.
John Dewey*

Let me start by being really clear: I am very much in favour of conducting research into the merits of educational claims. While I do not believe that such investigations can ever provide definitive evidence on what teachers must, or must not, do at any given point, I do think that we can and should seek to learn from the evidence unearthed by research.

And that’s the point: how do we learn from evidence? It’s possible to find evidence to support any idea and prove pretty much anything. ED Hirsch Jr says this:

Almost every educational practice that has ever been pursued has been supported with data by somebody. I don’t know a single failed policy, ranging from the naturalistic teaching of reading, to the open classroom, to the teaching of abstract set-theory in third-grade math class that hasn’t been research-based. Experts have advocated almost every conceivable practice short of inflicting permanent bodily harm.

It’s all too easy to reframe our beliefs to incorporate evidence and then continue to maintain that we were right in the first place. This is exactly how prosecutors accept DNA evidence may be correct, but continue to argue that a defendant is guilty. It’s how doctors accept the evidence that a patient died but continue to maintain that it wasn’t their fault and was ‘just one of those things’. And it’s how teachers can accept the evidence on phonics but continue to maintain that there’s more than one way to skin a cat.

If it’s not possible for you to admit you have made a mistake, or that you are wrong, you are unlikely to ever learn or improve your ideas. 
The trick is, I think, to set out when and how you might be wrong and then, if reasoned evidence contrary to our beliefs is produced, instead of trying to reframe the debate to show how we might still be right, we should try to learn and adjust our beliefs in line with the new evidence.
For instance, I wrote some months back about how I thought we might get assessment right in a post-levels world. I shared a system I had developed which some schools have adopted and of which I felt rather proud. Then Daisy Christodoulou pointed out why I was wrong:

David Didau responded to this argument accepting the major points, but also putting forward a replacement for levels that was based on performance descriptors. This was a pattern I was starting to see again and again – people would accept my criticisms of NC levels, but then produce a replacement which was essentially a rehashed version of levels. In this post, I tried to tackle this head on, by saying that Assessment is difficult, but it is not mysterious.

My instinct was to defend my beliefs, reframe the debate to show the circumstances in which I might be right, and protect my vested interests. In the aftermath I spent a fair bit of time following the background reading suggested by Daisy (Koretz’s Measuring Up and Wiliam’s Principled Assessment Design are both excellent) and after striving mightily to wrestle my ideas so that they fit the facts, I’m finally forced to accept that I was wrong.

Even though I’ve generated a bit of reputation for changing my mind (and have even been accused of lacking the courage of my convictions) this is still painful for me to write and some people may find themselves annoyed with me. But, if I’m going to ask others to attempt to falsify rather than simply confirm their ideas, then it seemed only fair that I should model the process myself and show that although it feels risky, this is actually the best and wisest course of action. Being nothing if not persistent, I’m now back at the drawing board trying to salvage some of this work and give Daisy another opportunity to point out my mistakes.

This is how we make progress.

*Readers should not let the irony of using this quote from Dewey escape them.

If you want to read more about the very predictable ways in which we all make mistakes all the time and what we might do instead, you might find my book, What If Everything You Knew About Education Was Wrong? interesting.