Saturday, February 24, 2018

Why do we listen to John Hattie?

Recently I received a memo to remind me of John Hattie’s research on collective teacher efficacy, the number one influence, according to his mega-analysis, on student achievement:
Collective teacher efficacy (d = 1.57). This is a factor that can be manipulated at a whole school level. It involves helping all teachers on the staff to understand that the way they go about their work has a significant impact on student results – for better or worse. Simultaneously, it involves stopping them from using other factors (e.g. home life, socio-economic status, motivation) as an excuse for poor progress. Yes, these factors hinder learning, but a great teacher will always try to make a difference despite this, and they often succeed.

There are so many questions no one seems to ask. First, how do we know this is true? So often we are handed this nugget of information as though it were fact. Hattie’s famous list of influences is a meta-analysis of 800 meta-analyses of more than 50,000 quantitative studies of variables affecting the achievement of students. But what does that mean? Despite the large numbers, it means that Hattie selected only those studies that focused on factors that he argues can be influenced in schools, so topics like class, poverty, health in families, and nutrition are excluded. Thus, although he says teachers should not use those factors as an “excuse,” he does not calculate their impact on learning. He includes only quantitative studies, that is, studies that measure what can be measured, test results, for example. Anything that documents the experiences of children and teachers, their joy, their anguish, their hopes and fears is excluded. He includes or excludes further studies based on his beliefs about the quality and motives of the research. These 50,000 chosen studies, then, conducted in the past, are stirred in the pot of statistical analysis, place, gender, culture, age, ability all blended, amalgamated, averaged and then summarized to give us the list of influences.

I keep wondering why we accept this list as though numbers so derived matter more than common sense. Children who are hungry, who don’t know where they are going to sleep at night, whose parents are suffering in the grip of generational poverty and trauma are not going to magically “improve” when we use concept mapping (effect size 0.64) in our classrooms. Although schools can’t “influence” poverty, our time and resources are used to heal, at least in small ways, the effects: we have food programs, extra clothing, washing machines and showers, counseling for children who are angry, hurt, grieving; we send home grocery cards, Christmas hampers, used laptops; we have after-school programs, organize funding for running shoes, and to pay costs so all the children can attend field trips or get extra music lessons. When we shush a child who is talking during the announcements, and he tells us to fuck off, pushes over his chair, and storms out of the room, we understand that it’s very possible he didn’t sleep last night, because he shares his room with two brothers and his aunt’s baby who often cries most of the night; we know he came late to school and so missed the breakfast program and likely hasn’t eaten since we fed him lunch yesterday; we understand that anger is a response to frustration, hurt, pain. We feed him; we love him. But it doesn’t help us teach him math. This is not an excuse. A teacher cannot be responsible for the success of a child. Even a great teacher.

But then, what is a greatness in a teacher? With the Olympics on right now, we know what great athletes are. Still, even a great snowboarder like Mark McMorris is excused when gusts of wind shift his balance. Even a great skater like Nathan Chen is excused when the complex, impossible to measure factors of heart, nerve and confidence destroy his poise.

Most of us are just ordinary. Still, I’ve never met a teacher who didn’t try. I see day-to-day miracles from ordinary teachers who try, try, try every day, without fan support, without a coach helping from the sidelines, without the resources they need, despite ludicrous lists that tell them to simply try harder, despite gusting winds and broken hearts.

I wonder why we continue to listen to the statistical machinations of an Australian academic rather than the teachers in our own community who say, please help.


  1. here here Shelley - well said. I've largely rejected 'hattian' voodoo since I first encountered it and realized in about 3 minutes just how many important 'effects' important to learning he was wilfully excluding. I wish that were the end of it but of course it's not - education loves simple stories it seems - however I'll formed they may be - and Hattian myopia has come to wield much more influence than it deserves. I think the best course of action is to continue to chip away and expose the fallacies of VL and share your insights with admins who've drank the kook-aid (and lots have).
    - michael maser (

    1. Thank you, Michael. The influence of the influences. I wonder what the measurement of that would be! I wonder, too, what would happen if we spent fewer resources chasing “simple stories” and more simply accepting and supporting the complexity that is.