A new study from the University of Arkansas Department of Education Reform indicates that the achievement gap between higher and lower performing students has been widening significantly over the past two decades. Misty Gallo and Danny Shakeel were two of the lead researchers on this report.
Shakeel says the idea of looking at the achievement gap came from previous research he had been working on. Prior to this, he was studying data dating back to the 1970s to determine how test scores have changed in the last half century.
Shakeel: And there we dispel the myth that test scores have actually grown. You hear a lot that things have worsened. But what we found was that there were huge gains, especially at the young age in math, and for the low SES, the socioeconomic status, students and ethnic minorities. That was shocking in some ways. We also wanted to look at how things fared in the 21st century. So Patrick Wolf and I teamed together and used that same data set, but for different years, and as I said, 21st century. And there we focused, instead of the test scores at the average point, we looked at the top and bottom points.
Moore: There's two things that really stand out to me about this study. The first is the approach to this achievement gap, which is kind of the main throw of this study. The thing that really kind of stands out is this achievement gap data, you compare the 75th percentile versus the 25th percentile gap, as opposed to like the 90th and the 10th.
Gallo: So I'll jump on that first. I was a school teacher for a long time, for about 20 years. And as a teacher, you have a knowledge that it's very easy to change the average if you just focus on the outliers, the very top and the very lowest performing students. It's much more difficult as a teacher to address that middle portion. And so from a pragmatic standpoint and from a practitioner standpoint, that made sense to me.
So Danny, as he said, when we first came together, we did that. We did what you're describing. We looked at the 90th and the 10th percentiles of the students using the NAEP data. But then, wanting to be a bit more nuanced, we looked at that 75th and 25th percentile. That's called the interquartile range, to really look at those students who aren't the outlier students, but really are the top-middle and the bottom-middle students, to see what's really happening to the students who cluster around that average point. That is the first reason. The second reason is that's much less commonly studied and assessed. And so, wanting to add something to the conversation about the achievement that American students are experiencing, that was the secondary reason that we looked at it.
Moore: I think most people associate the achievement gap, especially as we sit here in 2026, as a result of COVID-19, and a lot of, you know, online education was happening in a way that schools weren't prepared for. But your data shows that this happened at a much different time than COVID-19. It happened before that, before the pandemic.
Gallo: That's right. Part of the analysis that we looked at is we actually were able to separate out the pre-COVID trend and then the full-term trend, to see what was happening with the student-level data prior to COVID, and then over the full span of the almost 20 — I guess 17 years that we looked at. And so when we were able to do that, we were able to see that the trends, for the most part, were declining prior to COVID-19. COVID-19 may have exacerbated or accelerated the trend lines, but they didn't start them. COVID didn't start that trend. And so we were really interested in what was happening with the overall trend, but then also what was happening with that 75th percentile group, and then the 25th percentile, prior to COVID, to see what was causing the overall trend to either, as Danny said, get better or worsen.
Moore: I'm someone who's got two little kids. Every time I go into the doctor's office, you know, they take the measurements, and then they talk about where they fall in the percentile, as far as head circumference, height, weight, those sorts of things. And the thing that always sticks out to me about that is, let's say you've got a kid who's in the 10th percentile for weight. That's pretty small. But the thing that the pediatrician always tells us is, as long as they're growing, as long as they stay in that 10th percentile and up, then that means things are okay. You don't want to see them going down a percentile. Is that more or less true when we're thinking about percentiles with education as well? That even if a child is in the 25th percentile, that doesn't necessarily mean they're a, quote, bad student. That just means this is kind of where they fall on that trend, and as long as they are 25th and above, they're still doing okay. Is that oversimplifying it?
Gallo: I think you're exactly right. You want to see growth. There are always going to be — I also have two kids, mine are not little, mine are grownups now, but I remember those days — there's always going to be someone bigger, stronger, faster, higher-scoring than you, and there may be someone slower, less fast and lower-scoring than you, but you want to focus on your own growth. And the same thing for the 25th percentile and the 75th percentile students. We want to see incremental growth. We want to see the trend line of the 25th percentile scores actually improving over time. And, you know, unfortunately, that's not the case for many of the areas that we looked at.
Moore: When we think about this widening gap. From what I can tell, there doesn't seem to be any sort of assertion of a specific thing that is causing this gap. I think it's really just an examination of that — this gap is widening as time goes on, and that it started widening before the pandemic. Really, I think that was the two main takeaways I have from this. Am I understanding that right? That it's not necessarily that we're pointing a finger at any specific thing that is causing this gap, but what we're noticing is that this gap is growing, and it started growing before 2019, 2020?
Gallo: That's right. And, you know, it would be so nice if there were a thing, right? If we could say, if only we could do this one thing, then our lowest-performing students would begin growing in the way that we hope to see for those students. Unfortunately, it's not as easy to attribute the cause. And so, therefore, it's less easy to attribute a solution. I think that it's important, though, to understand that the gap is not widening because primarily the 75th percentile students are just growing leaps and bounds. The gap really is widening because our lowest-performing students are floundering. And that's something that, both as a mom and as a former teacher, is really distressing — to think that we have these students who are struggling now, and are struggling more now than they were, and they are struggling more in recent history than they were in the further recent history. So I think that even though we don't have a solution or a cause, it's important to pay attention to it so that we can begin to look for those solutions.
Moore: When we look at a state like Arkansas, and, you know, Governor Sarah Huckabee Sanders has made education a primary element of her time here in office — education is something that she's prioritized time and time again — when we're looking at a state like this, and we're thinking about this achievement gap that is growing, we can't really give a lot of pinpoints as to what specifically is making this grow. What sort of recommendations do you have to policymakers, to lawmakers, to say this is something we need to take seriously? Good luck. What's the answer there?
Shakeel: So, I would say two things. First thing is, I'll point out the value of reducing achievement gaps. So, if you can convert these numbers into years of learning, and when you do that, the gap comes out to be quite large in terms of the months of learning loss, the years of learning loss. So that's something that the policymakers should definitely think about. Nobody wants a child, on average, to lose a month of learning, or a half a year of learning. That's a huge loss. That's the first thing. And the second thing is, there are some interventions out there which have been proven to be effective at increasing learning. For example, high-dose tutoring, or differentiated instruction, things like that. If those kinds of interventions could be more available to the kids at the bottom-performing levels, or in the schools which are having those kids, that would be something that policymakers should seriously consider.
Gallo: And the other thing that I would add is, I think that it's really important also to look at different student population groups, and how those groups performed within the different sectors, because we do have bright spots for different groups. Not every student is the same. And so if one particular population of students is growing within, for instance, the charter sector, or the Department of Defense, for reading or for math and eighth grade — let's look at what those groups are able to do and able to experience within those sectors, and see if maybe we can, as Danny said, look for the things that they're doing differently and see if we can scale them. So things like high-dose tutoring — and I know that that's been a hallmark of Governor Sanders' policy implementation through LEARNS — looking to see how we can scale those kinds of interventions in ways that are really accessible for parents, so that there's not a lot of red tape that parents have to navigate in order to access those interventions, could be something that would really help our lowest-performing students. So now, instead of looking at that six-million figure, all of the students, if we can look at individual populations of students and how they're performing within different sectors, to address those gaps within each of the sectors.
Misty Gallo and Danny Shakeel are the authors of a new study conducted by the University of Arkansas Department of Education Reform. We spoke over Zoom last week.
Ozarks at Large transcripts are created on a rush deadline and edited for length and clarity. Copy editors utilize AI tools to review work. KUAF does not publish content created by AI. Please reach out to kuafinfo@uark.edu to report an issue. The audio version is the authoritative record of KUAF programming.
