Professor of Education and Social Policy at Northwestern University C. Kirabo Jackson discussed the implications of his research on the impact of spending on K-12 education at a Harvard Graduate School of Education talk in Larsen Hall Tuesday afternoon.
The research sought to investigate how much marginal spending affects educational outcomes. To quantify educational outcomes, Jackson primarily measured two factors: one variable measuring test scores and another variable combining high school graduation, college attendance, and dropout rates.
Jackson introduced his lecture with a question: In education, “How much does money matter?”
To investigate the effects of spending on education, Jackson and co-author Claire L. Mackevicius conducted a meta-analysis on a comprehensive set of design-based studies, which are considered credibly causal, and found various factors affecting educational outcomes.
The meta-analysis consisted of “31 studies that estimate the effect on school spending on student outcomes that meet our inclusion criteria,” Jackson said.
Responding to a question from the audience asking whether the number of papers is enough, Jackson said he had seen meta-analysis studies with as few as five to six papers, but “ideally” these studies would have a sample size of 20 to 30.
Jackson emphasized the importance of using accurate and informative articles for their study and considering which papers to include, though he said studies tended to measure scores differently.
“Sometimes studies report things not in terms of raw test scores, but they report things in terms of proficiency rates or passing rates,” Jackson said. “In this case, we have to find a way to standardize this.”
Though dropout rates were the least desirable piece of data because its implications were so unpredictable, they were used as a backup form of study, Jackson said.
“To the question of how we combine these things, essentially, what we did is we said, ‘Look, if you look at the literature, I think there’s a consensus that measuring dropout is very, very notoriously difficult to do,’” he said. “So in most cases, if they report dropout and high school graduation or some other measure, well, we’re not going to use a dropout if we can avoid it.”
Jackson shared that on average, an increase of school spending by $1,000 leads to a marginal increase in test scores, an increase in graduation rate by 2.8 percentage points, and an increase in college matriculation by 2.8 percentage points. Still, he said increasing the budget does not necessarily guarantee greater performance.
“To be clear, it’s not obvious to me that even when you increase the budget by $1,000, all of that is going to the most productive use,” he said. “It’s possible that half of it is wasted, but half of it is still doing something and that’s what we’re finding.”
Jackson said school spending studies are more consistent than they might initially appear, but the question of using the money to realize “positive effects” is “tricky.”
“One takeaway, clearly, is that it seems to have a larger effect in areas where we have more disadvantaged populations,” he said.
Jackson concluded his talk by recommending following the practices laid out in education research.
“The best thing we can do, in my opinion, is to say, ‘Well, if you want to know how to spend your money, forget all this, forget about the school spending stuff, just look at the research — it talks about best practices, right?” he said. “There are interventions that improve outcomes, we know there are certain kinds of curricula that tend to improve outcomes.”
“All these things I’m describing take money,” he said.