Michael Hicks: Thinking hard about the IREAD test


Gov. Eric Holcomb, along with leaders of the House and Senate, laid out a legislative plan to hold back students who fail to pass the third grade IREAD exam.

This is a significant policy issue because about one in five Hoosier third graders fail to pass the reading test.

This pass rate has worsened substantially since COVID.

My back-of-the-envelope analysis suggests that adding another year of school for 15,000 children would cost more than $125 million a year.

Other states have tackled this issue already, including West Virginia and Mississippi.

West Virginia will begin to hold back current first-graders who don’t pass their reading tests by the end of third grade.

For almost a decade, students in Mississippi who fail the test have been held back.

However, students in both states pass the tests at lower rate than do Hoosier students, though this may be due to different standards across states.

There are more than fiscal issues at play as well, and I certainly don’t know what the right policy decision might be.

Still, Hoosiers should welcome a full and open discussion of the matter.

Of course, this should begin with a full set of facts combined with rigorous analysis to share with the legislature.

Before taking up the question of whether to hold third-graders back, we’d want to know the costs, as well as the benefits of doing so.

So, I’d expect that we’ll soon hear not only a more rigorous fiscal analysis than I provided above, but much more details on the effect of the IREAD test on ultimate student success.

Knowing this requires more than just restating statistics.

After all, it should be very simple to report IREAD pass rates of high school graduates versus dropouts over several cohorts of students.

This could be done by demographics such as gender, race, ethnicity, poverty rates and school corporation size.

We’d also want to know the demographics of pass/fail for students in fourth grade this year.

Georgia has a “Governor’s Office of Student Achievement” that put together just such a study of their third-grade reading test in 2017.

These descriptive statistics data are fine, but they aren’t really sufficient for policymaking. The key question is what the effect failing the IREAD had on later student achievement, controlling for all these other factors.

Any competent researcher should be able to undertake this statistical test, so I would expect we will hear the answer in the coming weeks of testimony.

One recent study performed by the University of Chicago found that once individual student demographics and school effects were properly controlled for, very few educational outcome differences were predicted by third-grade reading scores.

This study is a perfect illustration of the issue.

The descriptive statistics of third-grade reading scores suggested major impacts on later reading scores, school attendance, course failure, GPA, graduation, school attendance and college attendance.

But, when individual demographics (race, gender, poverty rate and cognitive impairment), eighth grade reading scores and school characteristics were included in the statistical model, the third grade test scores lost all ability to predict student outcomes.

Let me state this more plainly.

In a study of Chicago’s schools, the research team found that when you control for individual student demographics, their eighth-grade reading scores and individual school characteristics, passing or failing third-grade reading tests had no effect on later educational outcomes (GPA, absences, high school graduation or college attendance).

It is worth noting that this comprehensive study was performed on Chicago’s schools in 2010.

This is a classic problem in evaluating school outcomes and interventions. Many different variables are correlated, such as race, poverty, cohort performance and test scores. Descriptive statistics alone cannot answer questions about the role of third-grade reading and later school performance.

To answer that question, you have to do analysis. Without it, you might be trying to fix a problem that doesn’t exist.

Doubtless such a study has been conducted on Indiana students, and we’ll soon learn the results of that study. But, there are other issues as well that the legislature should know as they approach this problem.

First among these is simply whether holding a student back improves later outcomes. A recent doctoral dissertation at Ball State reports that among the students held back for IREAD, outcomes improved in later reading tests. But, this study didn’t go beyond eighth grade.

Another study by a team of economists found that higher retention rates affected long-term average educational and economic outcomes.

Interestingly here, the benefits of this effect accrued to students who were promoted. Those who were held back faced a large economic cost. Because the number who were promoted was much greater than those held back, the net benefits were very positive.

As COVID demonstrated, not everyone is comfortable with this sort of benefit-cost analysis. But legislators and governors don’t have that luxury. They must consider such issues when evaluating policy alternatives.

Of course, this debate raises other questions. Does grade retention have other costly effects that might not have been considered? What is the optimal age to hold students back, and does that vary by gender? Are there lower-cost interventions that could improve reading outcomes? Are the reading standards appropriate? This debate raises even bigger questions.

Indiana is more than 20 years into expanding school choice. Those policies vastly expanded options for parents, and focused heavily on cutting costs for taxpayers. What has been the effect? In 2000, before the first school choice, 88% of Hoosier students attended local public schools. In 2022, that number was 89%.

Indiana has an almost fully operational school choice system, and traditional public schools have won the competition for students.

At the same time, we struggle on every important domain of educational attainment. In 2000 we were ranked 43rd among states in educational attainment. We rose to 38th by 2010, and are back down to 43rd today. The push for cost savings and choice has benefits, but higher levels of educational attainment are not among them. There are other concerns that warrant a focus on quality over cost savings.

Nationwide, the wage premium for completing high school is a roughly 35% pay boost above those who don’t graduate. Here in Indiana, it is 10.1%. That may be because we have an excess supply of high school graduates, but it may also be that employers don’t value our high school diplomas as strongly as those from other states. Either way, a focus on quality is in order, and third-grade reading is a fine place to start.

Michael J. Hicks is the director of the Center for Business and Economic Research and the George and Frances Ball Distinguished Professor of Economics in the Miller College of Business at Ball State University.

No posts to display