The consensus in education research circles has settled into a comfortable position: we need better replication studies, stronger statistical standards, and more rigorous methodology. Universities are hiring reproducibility officers. Journals are tightening protocols. Grant agencies are demanding pre-registration. All sensible. All insufficient.
But here's the question nobody wants to ask loudly: What if our reproducibility problem isn't actually a methodology problem? What if it's a permission problem?
Consider what happens when we focus narrowly on replication and rigor. We're essentially asking: "Can we run this experiment again and get the same results?" That's a good question. It's just not the only question. And by making it the dominant question, we may be inadvertently protecting something far more broken: the original decision about which questions deserve research attention in the first place.
Education research doesn't suffer from a shortage of published studies that fail to replicate. It suffers from a shortage of published studies about things that actually matter to most schools and students. The reproducibility crisis gives us something to fix that feels technical and solvable. The real crisis—that research agendas are often set by institutional inertia, funding availability, and academic career incentives rather than by evidence of what's broken—feels messier and harder to solve.
Think about recent research trends worth celebrating: work on early literacy instruction, decision-making skills as critical competencies, how to help small agricultural producers improve yield. These studies exist because someone decided these questions were worth answering. Someone secured funding. Someone had institutional backing. But in education, that "someone" is often a particular type of researcher at a particular type of institution, asking questions that fit particular funding streams.
Meanwhile, thousands of practitioners face problems daily that have never been rigorously studied. Teachers know which classroom management approaches work in their schools. Counselors know which student interventions stick. Principals know which scheduling models actually improve teacher collaboration. Yet the research literature on these localized, practical problems remains sparse. Not because the problems are unimportant. But because they're not fundable in the traditional sense. They don't fit neatly into controlled trials. They don't generate publications that move careers forward.
The reproducibility movement, for all its virtues, tends to reinforce this structure. It says: "The research that exists should be done better." It doesn't challenge: "Should this research exist at all? And what research is missing?"
Here's what concerns me: we may be creating a two-tier system where highly rigorous, perfectly replicable research answers questions nobody needed answered, while the messy, contextual, localized questions that would actually improve schools remain unstudied and undervalued.
This isn't an argument against rigor. Rigor matters enormously. It's an argument that rigor alone, pursued without interrogating the questions being asked, can become performative. It can become a way of saying "this institution is serious about knowledge" while leaving the fundamental question—whose problems get solved?—untouched.
The better question isn't whether our next study will replicate. It's whether our system for deciding which studies to fund in the first place is actually serving education. What would change if we asked researchers and funders: "What problems matter most to schools, and are we studying those?"
That question breaks the comfortable consensus. It suggests that better methodology isn't enough. We need better democratization of what counts as a worthy research question.