“It’s very important that CTOs and administrators don’t ask researchers in the early stages of research for really compelling evidence that we can’t give,” Dede said. “And it’s important that researchers don’t ask for big implementation platforms that put [school leaders] at risk.”
“Research can be helpful, but it doesn’t determine policy practice and decisions,” Kleiman added. “The pace of research doesn’t meet the time requirements of policy practice and decisions. [The pace is] particularly different in the current world of technology, which changes so rapidly, and would provide data that’s completely out-of-date by the time the research is completed.”
He added: “I think research can help us answer important questions, but it’s very important that we ask the right questions and use appropriate research to inform the right questions. Can we just continue doing what we were doing without technology? I think the research … makes it very clear that we can’t just accept the status quo. I think research, locally done, can very much inform questions such as, ‘What should we focus most on improving?’ Research can help inform us about what approaches we might consider [and] what approach is the best fit for our context.”
“We need to create an ecosystem of learning in which students have freedom to navigate” which style of learning will work for them, Dede said. “We’re not going to come up with a one-size-fits-all educational model that’s effective. It’s got to be different strokes for different folks.”
For more news about ed-tech research, see:
New guidelines for ed-tech research could help educators, vendors
On ed tech, we’re asking the wrong question
Study reveals factors in ed-tech success
Report: Ed tech has proven effective … but hasn’t realized its full potential
- Children learn best on paper: We need better technology, not less of it - July 13, 2026
- In districts, reaching readiness, retention, and success - March 5, 2026
- AI use is on the rise, but is guidance keeping pace? - January 2, 2026
Comments are closed.