1.   Research is best when there is

•   rich description of an intervention,

•   alignment of study design and outcome measures with the goals of an intervention, and

•   the use of control groups.

2.   The field—educators, program developers, researchers—could benefit greatly from a common framework for both description of an intervention and, when appropriate, for the research strategy.


3.   Delineate the impacts on achievement, interest, identity, and persistence. Avoid the “integrated STEM is good for everything” strategy.

4.   Examine the long-term impacts on interest and identify among diverse audiences.

Design and Implementation

5.   Delineate a logic model, including goals, necessary supports, and outcome measures.

6.   Be explicit about teaching and learning goals.

7.   Use the rapidly developing cognitive and learning literature to understand learning goals and learning progressions.


8.   Rethink assessment to enable the development of high-quality assessment tools.

9.   Embrace continuous improvement.

The STEM education community, both inside and outside schools, has an opportunity to build a virtuous cycle of continuous improvement, said Honey. She said a particularly promising option would be to take the framework developed by the committee and operationalize it so that different programs can be mapped onto the framework. The result could be a tool that, as Honey said, could provide much greater coherence and discipline in describing “what it is we want, how it’s going to happen, what it’s going to take to get there, and how we are going to know if we are successful.”

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement