For decades academic researchers have dueled over how to distinguish good teachers from bad ones. Some have zeroed in on using student test scores as a way of measuring which teachers are most effective. Critics have shown this measuring stick, called value-added measurement, to be deeply flawed.
So researchers have been going back to the drawing board, trying to prove that, no matter which measuring stick you use, the worst teachers usually end up teaching the most disadvantaged kids. Last month, one of the top researchers in this field, labor economist Dan Goldhaber, published a new study with some of the most convincing evidence yet.
His study, written with two colleagues, one at the American Institutes for Research (AIR), where Goldhaber is a vice president, and one at Macalester College, measured teacher quality in three different ways for every teacher in the state of Washington. They looked not only at student test score gains, but also at years of teacher experience and teacher licensing exam scores. Their study, “Uneven Playing Field? Assessing the Teacher Quality Gap Between Advantaged and Disadvantaged Students,” was published online in the journal Educational Researcher on June 29, 2015.
And no matter which of these three measures of teacher quality they used, guess what? They got the same result. Disadvantaged students across the state’s elementary, middle and high schools ended up with the worst teachers — the ones who not only produced the smallest test score gains, but also had the fewest years of experience and the lowest licensure exam scores. The researchers also defined student disadvantage in three different ways, by income, by race and by whether the kids were struggling academically.
And again the Goldhaber team found the same results. Low income kids ended up with the worst teachers. Black, Hispanic and Native American kids ended up with the worst teachers. And low-achieving kids ended up with the worst teachers.
“For the most part, no matter how you split the data, I hesitate to use the word ‘worse,’ but teachers who are less qualified, or less effective, tend to be teaching more disadvantaged students,” said Goldhaber.
It was true between districts. For example, a district with more black and Hispanic students had lower-ranking teachers than a district with more white and Asian students.
And it was true within districts. That is, a school with more low-income students had lower-ranking teachers than a school with a wealthier student body in the same district.
Of course, not every disadvantaged student had an ineffective, novice teacher with weak qualifications. But when Goldhaber looked across the state, on average, that’s what he found.
Goldhaber also looked to see if this was happening within schools. But he didn’t find that classrooms with more minorities or low-income students had much worse teachers than other classrooms within the same school. The worst inequities were between schools within a district or between districts within the state.
Previous studies have found similar inequities in other states and school districts. But earlier research hadn’t looked at several measures of teacher quality at once over a large geographic area. Goldhaber looked only at Washington State, but believes the inequitable distribution of teachers is true across the nation. “I would be surprised if there were different results in different states,” he added.
To be sure, there will be critics who will dismiss the three measures in this study — student test scores, teacher experience and licensing scores — as imperfect measures of what makes a good teacher. And perhaps one could find an excellent novice, who scored poorly on the licensing exam, and whose students didn’t improve as much on tests as one might expect. But there are probably not many of them.
Goldhaber was among the researchers who testified in the Vergara case last year, currently on appeal, which would dismantle teacher tenure in California. He argued that the union’s seniority system for layoffs meant the neediest children got the worst teachers, as measured by his earlier “value-added” analysis of student test scores.
But this time, when I interviewed him, he didn’t call for the end of teacher tenure.
Instead, he says it’s a classic labor market problem. Teaching in poorer, minority schools is a tougher job, he says. “As teachers get more experience, they have more negotiating power, and they tend to go to more advantaged schools,” he explained.
His solution is simple: pay teachers more to teach in tougher schools, and you’ll lure better teachers to them.
[Cross-posted at The Hechinger Report]