This post is a follow-up to The Leveling Down, which documented how Oak Park District 97’s 2017 elimination of elementary math acceleration led to declining scores for advanced students with no improvement for anyone else. This post examines why: the promise of differentiation was never realistic.

In 2017, Oak Park District 97 eliminated its elementary math acceleration pathway. The replacement: “enrichment for all” - a model where teachers would differentiate instruction within regular classrooms rather than providing separate advanced instruction.

The promise was beautiful: every student would receive instruction tailored to their level. No more pulling kids out. No more demographic imbalances in gifted programs. Teachers would meet each student where they are.

The reality is something else entirely.

The Promise

“Differentiated instruction” sounds like exactly what every parent wants. The theory: skilled teachers assess each student’s readiness, then provide appropriately challenging work - scaffolded support for those who struggle, acceleration for those who’ve mastered the content.

This was the explicit rationale for eliminating pull-out gifted programs. Instead of identifying some students as “gifted” and giving them separate instruction, all students would receive differentiated instruction in their regular classrooms.

As one systematic review summarized: “Differentiated instruction is described as an approach whereby teachers adjust their curriculum and instruction to maximize the learning of all students: average learners, English learners, struggling students, students with learning disabilities, and gifted and talented students.”

In theory, nobody loses. Everyone gets what they need.

What the Research Actually Shows

Finding #1: Teachers Don’t Actually Differentiate

In 2013, the Institute of Education Sciences - the research arm of the U.S. Department of Education - conducted a randomized controlled trial across 32 elementary schools in five Illinois districts. The study examined whether the NWEA MAP assessment system, combined with training on differentiated instruction, would lead to improved teaching practices and student outcomes.

The findings:

“MAP teachers were not more likely than control group teachers to have applied differentiated instructional practices in their classes.”

“The MAP program did not have a statistically significant impact on students’ reading achievement.”

Teachers received the data. They received the training. They didn’t change their instruction.

This wasn’t an isolated finding. In 2008, the Thomas B. Fordham Institute surveyed teachers nationwide: 83 percent said differentiation was “somewhat” or “very” difficult to implement.

A University of Virginia study provided teachers with extensive professional development and ongoing coaching. Three years later, researchers wanted to know if the program improved student learning. They couldn’t answer the question. Why?

“No one was actually differentiating.”

The researcher, Holly Hertberg-Davis, told Education Week: “There is no consistency in the few teachers who did make earnest attempts.”

Finding #2: Teacher Capacity Is the Bottleneck

Why don’t teachers differentiate? Because it’s extraordinarily difficult.

Research on teacher preparation confirms the problem:

“Beginning teachers indicate their implementation of DI significantly lower than teachers with more than three years of experience.”

“Teachers who do not feel competent to differentiate instruction will not implement DI in a meaningful way.”

“Teachers struggle to understand how differentiated instruction should be applied in their classrooms, despite the fact that the concept is widely recognized.”

This isn’t a failure of will. It’s a failure of design. A teacher with 25 students, 45-minute periods, and a mandated curriculum cannot simultaneously provide five different levels of instruction. The math doesn’t work.

Eye-tracking studies of teachers in actual classrooms reveal where attention goes: teachers focus disproportionately on students who demand attention - those who are disruptive or those who actively participate. Students who are quiet, passive, or “middle of the road” receive less visual attention.

As one study noted: “Passive students are frequently overlooked by preservice teachers.”

Finding #3: The Students Who Become Invisible

There’s a category of student that differentiation research describes with eerie precision:

“This learner will often be in the middle set for core subjects like Science, Mathematics and English, will achieve average test scores that parents and teachers don’t feel require any extra attention.”

Sound familiar?

Large-scale observational studies paint a consistent picture. Sirotnik’s examination of 1,000 classrooms found:

“The majority of class time was spent either with the teacher lecturing to the class or students working on written assignments. Students were typically involved in whole-class instruction and not interacting with either their teacher or other students. Students rarely selected their own instructional activities, and they were generally very passive in the classroom.”

Students at both ends of the spectrum become invisible for different reasons:

  • Advanced students: They’re not struggling, so they don’t need intervention. They’re not disruptive, so they don’t demand attention. They complete their work and wait.

  • Struggling students: When teachers are overwhelmed, the students who need the most support often get the least. Without systematic observation of student work, teachers don’t know which students are falling behind until it’s too late.

What Actually Improves Learning

John Hattie’s Visible Learning research synthesized over 50,000 studies to identify what actually impacts student achievement. The findings are striking:

FactorEffect Size
Teacher estimates of achievement1.29
Feedback0.70
Average effect0.40

“Teacher estimates of achievement” - the ability to accurately judge a student’s current level - has an effect size of 1.29, more than three times the average.

But here’s the catch: accurate teacher estimates require knowing the student. As researchers explain:

“These estimates are not based solely on tests or formal assessments — they rely on professional judgement built from observation, dialogue, informal checks, and knowledge of the learner.”

Feedback has an effect size of 0.70 - nearly double the average. But feedback requires knowing what the student got wrong and why. Research confirms:

“Task-specific feedback on a student’s cognitive process tends to be more impactful than general summary feedback or traditional academic feedback in the form of grades.”

In other words: the highest-impact teaching strategies require seeing each student individually. And that’s exactly what doesn’t happen in undifferentiated classrooms where teachers don’t check work, don’t observe student thinking, and don’t provide individualized feedback.

The Same Problem, Two Symptoms

I know two families whose children attend the same elementary school in Oak Park.

One family’s daughter scores in the 99th percentile on standardized tests. Three teachers confirmed she was working two grade levels ahead. She was denied math acceleration because she scored 63% on a rubric that measures first-grade performance, not third-grade readiness. The system couldn’t see her.

Another family’s daughter struggles with basic math - adding 7 and 9, for example. Her mother describes the problem: teachers don’t check students’ work, don’t observe their strengths and weaknesses, don’t provide meaningful feedback on their math development.

Same school. Same classroom model. Same fundamental failure: the system doesn’t see individual students.

One child is invisible because she doesn’t struggle. One child is invisible because her struggles aren’t noticed. Both are failed by a system that promises personalized instruction but delivers none.

The Dark Side of Detracking

Here’s what makes this worse: research suggests that mixed-ability classrooms may actually harm the students they’re supposed to help.

A 2023 study published in Learning and Instruction - titled “The dark side of detracking” - found:

“Detracking negatively affected self-concept of students with low achievement. In both studies, students with low academic achievement had a lower self-concept in untracked cohorts than in the tracked ones.”

“This study highlights potential side effects of detracking school reforms that might result from students with low academic achievement being exposed to unfavorable social comparison processes.”

The self-concept of high-achieving students wasn’t affected either way. But struggling students did worse - not academically, but emotionally - when placed in mixed-ability classrooms.

The rationale for detracking was that ability grouping harmed struggling students by signaling low expectations. The research suggests the opposite may be true: mixed-ability grouping can harm struggling students by forcing constant comparison with higher-achieving peers.

The Fordham Conclusion

The Thomas B. Fordham Institute - a center-right education policy organization - summarized the research bluntly:

“In the case of differentiation, the evidence is lacking. There is plenty of academic literature on the topic—professors talking about the theory and why it ought to work—but comparative studies are few and far between.”

“No high-quality research shows that this sort of heterogeneous differentiation can work at scale for the full range of student readiness levels that are typically present in American classrooms. And several large-scale meta-analyses say it doesn’t work.”

Robert Pondiscio of Fordham described his “downright hostility to another fashionable educational idea—the myth of differentiated instruction”:

“In practice, it tends to complicate a teacher’s work and dumb down instruction. The excruciating difficulty of doing it effectively means that many teachers don’t do it at all.”

The Core Problem

Here’s the fundamental issue: differentiation requires teachers to see each student individually.

That means:

  • Checking student work (not just completion, but understanding)
  • Observing how students approach problems
  • Providing specific, timely feedback
  • Adjusting instruction based on what you observe

But when you have 25 students, a mandated curriculum, limited planning time, and inadequate training, you can’t do any of that at scale.

So teachers lecture. Students listen passively. Work goes unchecked. Struggles go unnoticed. Advanced students wait. Struggling students fall behind. Everyone is in the same classroom, but nobody is getting differentiated instruction.

The promise was personalized learning for all. The reality is invisible students - children whose actual abilities, whether advanced or struggling, go unseen.

What This Means for Oak Park

When Oak Park District 97 eliminated elementary math acceleration in 2017, the stated alternative was differentiated instruction in regular classrooms.

As I documented in The Leveling Down, the E3 Group analyzed the district’s own data two years later:

“Third and fourth grade math GTD students saw clear declines in RIT math scores when switched to the enrichment for all program.”

“Those not in GTD math produced RIT scores similar to performance in previous years.”

Advanced students got worse. Everyone else stayed the same. Nobody benefited.

The research explains why: the differentiation that was supposed to replace pull-out instruction never happened. It couldn’t happen. The conditions don’t exist for it to happen.

The students who were previously identified and served became invisible - their needs unmet, their abilities unchallenged. And the students who struggled continued to struggle, because the same system that couldn’t differentiate upward couldn’t differentiate downward either.

The Questions

To Oak Park District 97:

  1. When elementary acceleration was eliminated in 2017, differentiation was the stated alternative. What evidence shows that differentiation is actually occurring in classrooms?

  2. The IES study found that MAP data plus training didn’t lead to differentiation. What makes the district’s approach different?

  3. John Hattie’s research shows that teacher knowledge of individual students has an effect size of 1.29. How are teachers systematically observing and assessing individual student understanding?

  4. Research shows students who are neither disruptive nor struggling tend to receive less teacher attention. What systems exist to ensure advanced students don’t become invisible?

  5. The 2023 “dark side of detracking” study found negative effects on struggling students’ self-concept. Has the district considered whether its approach may harm the students it’s designed to help?

The Bottom Line

Differentiation sounds beautiful. In theory, every student gets what they need. In practice, the research is clear:

  • 83% of teachers find differentiation difficult to implement
  • Three years of professional development didn’t lead to differentiation
  • IES found no impact on instruction or achievement from MAP data plus training
  • Students at both ends of the spectrum become invisible

The promise of “differentiated instruction for all” was used to justify eliminating programs that actually served advanced students. But the replacement was never realistic.

You can’t differentiate instruction when:

  • Teachers don’t check individual student work
  • No one observes student thinking
  • Feedback is grades, not specific guidance
  • Class sizes make individual attention impossible
  • Training is inadequate
  • Time is insufficient

When Oak Park eliminated acceleration and promised differentiation, it didn’t create a better system. It created a system where all students - advanced and struggling alike - became invisible.

That’s not differentiated instruction. That’s undifferentiated neglect.

Related Posts:

Research Sources:

This is part of an ongoing series documenting one family’s experience with gifted education acceleration in Oak Park Elementary School District 97. All facts are based on district data, published research, and official communications.