Growing race, class gaps in college graduation rates 

Study finds MCAS scores largely explain long-term income trends for college grads

ACHIEVEMENT SCORES AND educational attainment levels for Massachusetts students of all backgrounds have improved since the early 2000s, but large gaps remain and those for college graduation rates have grown wider, a troubling trend in a state already plagued by significant income inequality, according to a new study.

At a time of heightened focus on college completion being crucial to securing a place in the middle class, huge demographic disparities in higher education graduation rates exist even among students who appear similarly well prepared academically for college studies.

The report by researchers at Brown University, which examined 10th grade MCAS scores and later student outcomes, also found that scores are highly correlated with high school and college completion and later earnings in the labor market, but the strength of this association varied widely depending on students’ demographic background.

“I think it’s both good news and bad news,” said John Papay, an associate professor of education and economics at Brown and the lead author of the report.  “We see that educational attainment has been increasing for all groups of students, but we see that gaps in four-year college completion are also increasing.”

The report comes nearly three decades after the state’s 1993 education reform law paired a big infusion of new state spending on schools, especially in lower-income communities, with rigorous new academic standards and assessments. The researchers sought to understand a critical question that underpins the reform effort: Do achievement results in high school have long-term implications for educational and earnings outcomes? 

The researchers found that 10th grade MCAS scores were powerful predictors of college enrollment and completion as well as later earnings. For students taking the 10th grade math MCAS in 2003-2005, those scoring at the 75th percentile earned $22,342 more than students scoring at the 25th  percentile in 2019 — when they were roughly 30 years old, the report found. 

The researchers say the relationship of MCAS scores to earnings is hardly surprising, given the known correlation of scores to family income and other factors likely to give students advantages as they enter the workforce. That makes it hard to know on that basis alone whether the math and English skills captured by the tests themselves “pay off” later in the labor market or are just markers for other factors at play. 

But additional analyses at least suggest that they do, as higher scores were correlated with higher subsequent earnings when comparing students from similar demographic backgrounds. The connection between MCAS scores and later earnings held true even among those with similar high school GPAs and college graduate rates. What’s more, when the researchers compared students of similar backgrounds who had similar 8th grade MCAS scores, they found that those who had big test scores gains by 10th grade had 22 percent higher earnings at age 30 ($10,903, on average) than those whose scores had not improved. That suggests that gains in skills during high school, not just background factors that students arrive with in 9th grade, may translate to greater earnings in adulthood. 

While Horace Mann, the state’s first secretary of education, famously declared that education was “the great equalizer of the conditions of men,” the study found that it has yet to live up to that vaunted promise. Even when students had comparable 10th grade MCAS scores, their college competition rates differed substantially based on family income. 

Among those scoring at the state MCAS average in 2011, for example, about half of students from higher-income households had a four-year college degree within seven years compared with only about 25 percent of those from low-income homes. Stated another way, the researchers said, “low-income students scoring at the 90th percentile on the MCAS test graduate from a four-year college at the same rate as higher-income students who score at the 57th percentile.” 

“The equity implications of this finding are particularly troubling: students with similar academic skills in high school, as measured by the MCAS, have quite different educational attainments depending on their family income,” write the researchers.

Overall college graduation rates have improved in the state, with 42 percent of 10th grade MCAS takers in 2011 earning a four-year degree within seven years compared with 32 percent of their peers in 2003.

Gains occurred among all demographic groups but were much larger among students from higher-income homes, where college completion rates increased from 38 percent in 2003 to 52 percent in 2011. For lower-income students, the rate went from 10 percent to 18 percent, so the gap in college graduation rates actually grew from 28 to 34 percentage points. The gap also widened between white students and their black and Hispanic counterparts, with the white/black graduation gap increasing from 23 percentage points in 2003 to 26 in 2011. The white/Hispanic college graduation gap increased over that period from 28 percentage points to 33 points. 

Educators have increasingly turned their attention to not just equalizing college enrollment rates among students of diverse backgrounds, but reaching parity in completion rates. On that front, the report delivers sobering news. Researchers found a 20 percentage point graduation rate gap between lower-income students and higher income students who enroll in four-year colleges with similar MCAS scores. The researchers say this likely reflects a mix of factors, including high school quality, family financial resources, social supports while in college, and disparities in financial and academic support available at the higher education institutions they attend.

Among all students taking the 10th grade MCAS in 2003-2005, those from higher-income backgrounds earned, on average, 30 percent more than their peers from low-income households by the time they reached their early 30s. Among those with similar MCAS scores, that gap shrank to 16 percent, and among students with similar scores and college completion rates, the gap narrowed further to 10 percent. Thus, about two-thirds of the overall earnings gap could be explained by differences in MCAS scores and college graduation rates.

“Closing the existing gaps in high-school performance and postsecondary educational attainments could dramatically reduce current levels of income inequality,” the report says. 

For racial subgroups, the earnings gap essentially disappeared for Hispanic and Asian students who matched their white counterparts on 10th grade scores and college graduation rates. However, a gap of 12 percent persisted for black students, a finding that comes amidst national protests spurred by police brutality against blacks that are also spotlighting broader racial disparities faced by blacks. 

While state leaders regularly boast about top rankings in student achievement, the researchers say learning gains appear to have largely plateaued since 2008, with any increases since then  largely due to test score inflation or other factors not reflecting true learning gains. 

The changes have taken place during a time when the share of the state’s public school population with higher needs has grown substantially. The percentage of students from low-income households rose from 23 percent in 2003 to 37 percent in 2014, while the percentage of English language learners more than doubled from 6 percent in 2003 to 14 percent in 2018.