There’s a lot of turmoil now in the world of education policy and in research and development, generated by massive changes in Washington, DC. This is leading to confusion and fear among those who work in schools, education R&D, and more, but on the other side of this may be a chance to rebuild better.
In talking about how to rebuild better, one of the terms that gets used a lot (including within the Alliance for Learning Innovation) is “evidence-based.” We want more education decision-making to be evidence-based, and we want education R&D to generate the kind of evidence that allows for evidence-based decision-making.
In other fields, like health care, the term “evidence-based” is well-understood. Because of how those fields have grown up, many participants and decision-makers already have enough “evidence-based” underpinnings that you hardly need to say more about this.
I worry that the education and training ecosystem, as a whole, has had limited experience using the evidence and science that exists about human learning, motivation, identity, belonging, reactions to stress, etc. to make better decisions using evidence at different scales: individual students, classrooms, schools, districts, products, investments, and more. This means there is a real chance for confusion around what “evidence-based” decision-making needs to look like in education to be most useful going forward.
“Evidence-based” Does Not Equal “Number-Using”
One incomplete vision of “evidence-based” is “number-using.” The tools we use more and more generate numbers of various kinds that then “get used” in some fashion (possibly by other tools) to make decisions about learning, process, and outcomes. So if we’re using numbers, aren’t we by definition “evidence-based”?
If we look at other fields, like healthcare, it is the case that “evidence-based” decision-making involves using numbers to make decisions. However, there’s a whole infrastructure of validity and reliability, including demonstrating these numbers represent real biological situations, that underpins these numbers, and is assumed to be in place. Indeed, diagnostic tests that generate numbers tie in with known and accepted broader principles from research on how bodies work, how disease works, and how the body responds to disease.
These diagnostic tests link to a principled decision-making system. Typically in health care, when you are problem-solving, you are not expected to start from a flurry of data without any underlying rhyme or reason. Rather, the tests are ordered because of some initial diagnostic work-up, some connection to our current understanding of the principles of biological function and disease, which then makes some additional specific evidence valuable for the practitioner to make decisions (also grounded in evidence). This also helps with efficiency. We have a huge array of diagnostic tests we could be running on a person’s body, some of which cost thousands of dollars. We need a science-grounded, evidence-based case for when we call for those deeper tests – and we do call for those tests when needed. (And develop them for just those situations.)
Grounding the Numbers in Principles of Human Learning
By analogy, when we talk about wanting education to be more “evidence-based,” what we should mean is to go beyond merely performatively “using data” to actually doing principle-based diagnoses, development, and treatments, where the principles themselves are evidence-grounded, and the starting point is how humans learn and are motivated.
What does this look like in practice? Let’s say Rory, a fourth-grade math student, seems disengaged with his math work. He was enthusiastic about math last year, based on the evidence provided by his third-grade teacher. What could be going on?
We might first distinguish between a cognitive problem and a motivational problem. Does the fourth-grade math work draw on knowledge and skills Rory never mastered last year?
Or is there evidence that Rory has already mastered the prerequisites? If so, that shifts the diagnosis more towards a motivation question. Why is he not starting, persisting, and putting in mental effort? Based on research about motivation blockers, we might investigate different causes:
- Does Rory value the math he’s trying to learn and how he’s learning it?
- Does he think he can master this math?
- Does he think something is in his way to being able to do the practice and feedback work needed (e.g., he doesn’t have enough time to complete math homework)?
- Is Rory in a negative emotional state this year?
With a few questions, we can narrow in on a potential evidence-grounded cause for his motivation issue, and then engage in the right kind of intervention based on what might be going on.
Concluding Thoughts
To ground education R&D in principles of human learning, the broader education community must more deeply understand the sciences of learning, motivation, identity, belonging, and how humans respond to stress. Education stakeholders also need more practice using these sciences to propose what evidence to gather about an issue, to use that to do some diagnoses about what is going on, grounded in principles of human learning, then propose solutions consistent with those principles, to be evaluated with valid and reliable evidence-gathering in the field.
For this to work, an evidence-based ecosystem would likely have more agreement on some fundamental principles of how learning works and how it doesn’t: things like the definition and importance of cognitive load, how much expertise is (and has to become) tacit/non-conscious, the importance of the right kinds of practice and feedback, the critical role and reasons for failure of motivation, impacts and mitigation of stress, and specific techniques for the hardest and most important skill outcomes. We don’t yet have that across the board. I would argue we won’t be “evidence-based” if we don’t get these mastered in some fashion, including by families and learners themselves.
Therefore, in the various places where we want to describe how education and training can improve, we may need to be clearer about the ambition to have that ecosystem be “evidence-based,” not just “number-using.” We should clarify that we need stakeholders to understand what the science says about learning, motivation, and more in order to get the right kind of evidence and make the most impactful decisions based on it on behalf of students, teachers, and families.
