Achievement gaps: why they’re so stubborn
School teams have worked for decades to close achievement gaps in reading and math—especially gaps connected to family income and, in some settings, race and ethnicity. Yet many gaps show up early (even by kindergarten entry) and don’t change much as students move through elementary school.
A research team from The Meadows Center for Preventing Educational Risk at The University of Texas at Austin (Scammacca, Fall, Capin, Roberts, & Swanson) took a close look at what “typical” instruction produces in real schools. Their goal wasn’t to test a new intervention. Instead, they wanted to understand how students grow in reading and math across Grades 1–5 when schools do what they normally do—universal screening, progress monitoring, core instruction, and the interventions districts already have in place.
What the study looked like (in plain language)
The researchers analyzed data from about 5,900 students across 18 elementary schools in a large, diverse Texas district. Students were assessed in reading and math six times across two school years (fall, winter, spring each year). That matters because many studies only have one test per year, which can hide important within-year patterns.
They used Renaissance STAR Reading and STAR Math, which are computer-adaptive, vertically scaled assessments. In simple terms:
Computer-adaptive means the test adjusts difficulty based on student responses, reducing “too easy” or “too hard” items.
Vertically scaled means scores are on a growth scale that can track progress across grades on one continuous metric.
The research design was “cohort-sequential,” meaning they followed multiple overlapping groups for two years each (for example, Grade 1 to Grade 2, Grade 2 to Grade 3, etc.) and linked them to describe growth across Grades 1–5 without waiting five full years for one cohort to move through all grades.
Big finding #1: students who start low often grow faster
One of the most important results is also one of the easiest to misunderstand.
The study found that students with low initial proficiency (those starting in the bottom quartile) often showed faster rates of growth than students who started higher. This pattern was especially strong in math.
This matters because many educators have heard of the “Matthew effect” (the idea that students who start ahead grow faster and the gap widens). In this dataset, the pattern looked different: lower-performing students frequently grew faster.
So, problem solved? Not quite.
Big finding #2: faster growth didn’t mean catching up
Even though students who started behind often improved at a faster pace, the growth was usually not enough to close the gap. After two years, many students who began far below peers were still below average and still well behind higher starters.
In other words, the study suggests a “stable differences” pattern: gaps persist even when struggling students make meaningful progress.
From a school leadership lens, this is a key takeaway: growth and gap-closing are not the same outcome. A student can show strong growth and still remain significantly behind grade-level expectations.
Big finding #3: growth slows down as students get older
The researchers also calculated effect sizes for growth within each grade. The pattern was consistent:
Growth was largest in earlier grades and declined steadily through Grade 5.
Reading growth slowed more than math growth as grades increased.
For example, within-year reading growth (beginning to end of year) dropped from a large effect in Grade 1 to a much smaller effect by Grade 5. Math showed the same downward trend, though growth remained comparatively stronger than reading across grades.
Practical implication: it is typically easier to accelerate skills in Grades 1–2 than in Grades 4–5. That does not mean older students can’t improve—it means the system must be even more intentional and intensive to create the same “distance traveled.”
Big finding #4: the first half of the year matters
Across grades, more growth tended to occur from beginning-of-year to mid-year than from mid-year to end-of-year, particularly in math.
That pattern supports what many school teams observe: fall-to-winter is a high-leverage instructional window. It’s also the window where timely identification and service delivery can prevent “waiting until spring” to respond.
What demographics added (and what they didn’t)
The study examined demographic predictors including free/reduced-price lunch status (as a proxy for socioeconomic status), ethnicity (primarily White vs. non-White due to sample size constraints), and gender.
Key points:
Demographic effects were small but significant for both initial scores and growth rates.
Socioeconomic status was the most consistent predictor of slower growth in reading and math.
Ethnicity showed small associations with growth when SES was included in the model.
Gender effects were minimal, though there were small differences in math growth in some cohorts.
In day-to-day school terms, this reinforces a familiar reality: demographic variables do not “explain away” low performance, but they do correlate with risk and opportunity. That’s exactly why strong universal screening, data review cycles, and equitable access to intervention are non-negotiable.
What this means for school teams using screening and progress monitoring
This study is especially relevant for districts using universal screeners and progress monitoring tools (including vertically scaled measures). It suggests that typical instruction plus typical intervention supports may help students grow faster when they start behind—but not enough to erase the gap.
So what should schools do with that information?
Start early and act early. The strongest growth happens in early grades, and gaps are easier to shrink before patterns solidify.
Use growth data alongside proficiency data. Celebrate growth, but keep asking: “Is the student on track to meet grade-level expectations within a realistic timeline?”
Plan for intensity, not just placement. Being in an intervention group is not the same as receiving an intervention that is intensive enough in time, frequency, group size, and instructional precision.
Watch for cross-domain needs. The study found connections between initial reading and later math growth (and vice versa). Many students need coordinated supports, not isolated “reading only” or “math only” plans.
Where online therapy services can fit in (a practical district perspective)
As a Special Education Director, I see the same operational challenge across districts: we can identify needs faster than we can staff them. Therapist shortages (SLPs, OTs, school psychologists in some regions) create delays that directly affect students—especially those already starting behind.
Online therapy services, like what TinyEYE provides, can support districts by:
Reducing service gaps caused by vacancies so IEP minutes and intervention supports are delivered consistently.
Supporting early intervention efforts when schools need timely communication, language, and learning-related services that influence reading development.
Improving scheduling flexibility during high-leverage windows (for example, fall-to-winter) when growth potential is often greatest.
Helping teams sustain progress monitoring cycles by keeping related services stable while instructional teams adjust interventions based on data.
Online services don’t replace strong core instruction or evidence-based intervention systems. But they can help districts execute those systems more reliably—especially when staffing is the barrier.
Bottom line
This research offers a clear, data-based message: students who start behind can and do grow—often faster than peers—but typical growth is rarely enough to close achievement gaps by itself. If districts want gaps to shrink meaningfully, we need earlier action, more intensive supports, and fewer interruptions in service delivery.
For more information, please follow this link.
