Can We Help Students Become Better Readers by Buying Them Bigger Shoes?
In this video, Dr. Justin Baeder uses a vivid analogy to explain why education policymakers often confuse correlation with causation, leading to ineffective interventions.
Key Takeaways
- Correlation isn't causation - Just because two things happen together doesn't mean one causes the other
- Bad reasoning leads to bad policy - Confusing correlated factors with causes results in interventions that don't address the real problem
- Focus on direct instruction - The most effective way to improve reading is to teach reading better, not to address loosely related factors
Transcript
It is incredibly important that all students be able to read well for their success in life.
You might have heard that they plan prison capacity based on third grade reading scores.
So it is absolutely urgent that we help all students read on grade level, become proficient lifelong readers.
And one way that we can do that is by buying them bigger shoes.
What do I mean by that?
Well, in this book, The Book of Why by Judea Pearl, which I've mentioned in some of my other videos, this is backwards, but here's what it looks like.
Fantastic book on causality.
Pearl gives this example of the correlation between shoe size and reading ability.
At the elementary level, shoe size is highly correlated with reading ability.
And I'll give you a second to think about why that might be and why that effect might taper off.
in the later grades obviously shoe size is correlated with reading ability because when kids start school they have smaller feet and lower reading ability and as they get bigger they also tend to get better at reading so there's a very strong correlation between shoe size and reading ability But as we know, correlation is not causation.
And if we get confused about that, we can end up with policy choices that are counterproductive and just nonsensical.
If someone said to you, hey, it's so important that we raise reading scores.
Therefore, we're going to buy all the kids bigger shoes.
You would probably say, hey, that doesn't make any sense.
And B, it's not going to work, and C, it's going to have lots of unintended consequences.
And you might be called names.
You might be accused of not caring about kids or not believing in kids if you raise those objections.
And you might be forced to assist in this process of buying all the kids bigger shoes in order to raise their reading scores.
And you would probably notice in the ensuing months and years that kids were tripping more.
They were getting hurt.
They were unable to run in PE class.
And all of this did nothing for their reading scores.
And again, this is because reading scores and shoe size are only correlated.
There's not a causal relationship between them.
They are both correlated with each other because they're correlated with age.
And when it comes to breaking the school to prison pipeline, we're attempting something similar.
We're saying we're going to improve students' life outcomes and cut down on the chance that they get arrested as adults by not suspending them when they cause huge disruptions or are violent in school because we know there's such a strong correlation between school suspension and later incarceration.
And that correlation is spurious.
We can't manipulate one of those to improve the other.
because they're both rooted in something common.
Just like reading scores depend on age and shoe size depends on age, what we're seeing with both suspensions and the cause of the correlation between suspensions and incarceration is behavior.
So we have to do something about behavior if we want to improve students' life outcomes.
We can't just eliminate the consequence and hope it will work.
So if you see articles like this that say proving the school-to-prison pipeline and they're all about just correlations, look deeper and question those studies.
Let me know what you think.