"And then a miracle occurs..."
Without uncovering assumptions, most plans are likely to produce less-than-expected results. For example, your professional development activities may have little impact if you simply assume that staff are already willing to participate or that high-quality training is available. As you develop a logic model, surface the assumptions behind your program’s theory of action and find ways to support them or test them.
Resources
VIGNETTE
Why Assumptions Matter: Getting on the Same Page (.pdf 115 KB)
Consider this scenario to understand why key partners should discuss beliefs and assumptions before implementing a program.
VIGNETTE
If You Build It, Will Students Come? (.pdf 177.3 KB)
Reflect on real-life assumptions related to recruitment activities of magnet programs across the country.
TOOL
Checking Assumptions About Magnet Programs (.doc 72.5 KB)
Generate discussion about your magnet program’s underlying assumptions and what it takes to meet its objectives.
TOOL
Testing Assumptions: Do Our Activities Lead to Desired Outcomes? (.doc 1.6 MB)
Use if-then statements to check the rationale for your planned activities, as well as test and surface key assumptions.
Extra Resources for MSAP Rigorous Evaluation
-
Expect probing questions from your rigorous evaluator. To design an effective evaluation that measures a program’s impact, your evaluator must first understand the assumptions behind your program and how it is supposed to work.
-
Don’t assume that magnet program implementation looks the same across grade levels and themes. A common pitfall is assuming homogeneity of magnet activities to determine a districtwide effect of the program without accounting for major differences between, for example, STEM and art themes, or elementary and high schools.
-
A good timeline for conducting an experimental or quasi-experimental evaluation reflects the reality that full implementation of program activities often doesn’t occur until year 2 of the grant cycle. Build in checkpoints for testing your assumptions related to implementation at the end of year 1 and adjust your logic model accordingly.
