Often in education research, the answer is embedded in how you ask the question.
The statistical news blog FiveThirtyEight (named for the number of electors in the U.S. electoral college) launched a newsgame allowing readers to use the same data set to “prove” that either Republicans or Democrats are better for the economy, by adjusting categories included in the analysis.
Source: FiveThirtyEight
While the accompanying story focuses on political research, it’s an interesting read for anyone interested in why education findings can be so difficult to replicate, and why studies of the same data set often conflict. (It’s also a pretty fun tool to play with.)
And for more on persnickety education research results, check out our look at why so many education results fade out over time.