Jacob Nestle
When Passing Students Becomes The Problem
Every year, schools move kids into classes they aren’t ready for. Everyone involved has a good reason for letting it happen. Holding students back is more work. Passing them is, in a sense, easier.
But now we have teachers like Darius Williams of Houston crying on camera because their students are struggling with basic tasks. The real problem isn’t that a student moving from the 3rd grade to the 4th grade is a little bit behind. It’s when that same student continues on, passed along with just enough improvement to make everyone feel like it’s fine, increasingly struggling to keep up with the next grade’s coursework.
Williams found this out when he gave seniors two paragraphs and a sentence to finish—the kind of assignment that should be easy for most students before the end of middle school, let alone high school. But his seniors could not read and write well enough to do it. It was just four words.
More Diplomas, Less Education
For a while, it was easy to assume that receiving a diploma and earning it went together. The national four-year public high school graduation rate rose and rose, remember? That sounds like straightforward progress.
But what happens if we look at what high school seniors actually learned over the same time period?
On the 2024 National Assessment of Educational Progress, the average 12th-grade reading score was lower than it had been in 2019. It was a full 10 points below the first comparable assessment in 1992.
Math was also down. The average 12th-grade score in 2024 was three points below both 2019 and 2005. Forty-five percent of high school seniors scored below the NAEP Basic level in math.
COVID obviously hurt. Phones and social media, too. There are a thousand arguments about curricula, parenting, school discipline, social media, funding, teacher shortages, and everything else. I’m not going to pretend one policy explains national test scores. But one thing is certain: We are getting better at handing students diplomas without seeing any evidence that students are leaving school knowing as much as they once did.
There is evidence of grade inflation, too. Not that the national NAEP test itself is secretly getting easier, but that the grades and credits students receive can become easier to obtain.
ACT looked at millions of students and found that, even after accounting for student and school characteristics, average high-school GPA among ACT-tested students rose from 3.17 in 2010 to 3.36 in 2021. Standardized achievement measures did not rise with the grades.
North Carolina accidentally created a particularly clean experiment. In 2014, the state changed its grading scale. An “A” moved from 93 to 90. More importantly, the minimum passing grade dropped from 70 to 60. Researchers found evidence that lower-performing students changed their behavior when the standards became more lenient:
“We find that increased grade leniency led students to obtain GPAs that were 0.27 points higher (11% increase), with no effects on student achievement. Furthermore, students that faced more lenient grading standards increased their school absences by 1.3 days per year (20% increase), which drove students to become more chronically absent.”
Absences increased. Students put in less effort when less effort was required to pass. In other words: Kids respond to incentives too.
It’s worth mentioning the concept of credit recovery: programs that allow students who fail courses to regain the credit, often through online coursework. Credit recovery can do something valuable. It can prevent one failed class from permanently derailing a student. But there is a tradeoff.
One study of North Carolina students found that students who went through credit recovery were about 7% more likely to graduate on time than comparable students who repeated the course traditionally. They also scored as much as two-tenths of a standard deviation lower on end-of-course tests. More graduates. Less mastery.
That does not mean we should abolish credit recovery. It means we need to pay close attention to which outcome we are optimizing: Educating students or giving them diplomas.
The seniors Darius Williams is trying desperately to help are not alone. And neither is he. Teachers across the country are struggling to connect with students who don’t have the skills they should.
Whatever went wrong with each of those students in Darius Williams’ class in Houston didn't happen their senior year. It happened long before that.
Our system failed them by passing those students to the next grade, in spite of their actual ability.
What Happens If Everyone Has to Graduate
It is worth remembering how radically our expectations for education have changed. In 1940, only 24.5% of Americans 25 and older had completed four years of high school. By 1970, it was 55.2%. By 2000, 84.1% had completed high school, and by 2023 the number was 91.4%.
College expanded even faster. The share with at least four years of college or, in later years, a bachelor’s degree rose from 4.6% in 1940 to 11% in 1970, 25.6% in 2000, and nearly 40% in 2023. That is, in most ways, an enormous success.
We should not go back to a world where finishing high school is unusual or where college is effectively closed to most people. Expanding access to education has opened opportunities to millions of Americans who would not have had them before. But it also changed what failure means.
When only a minority of Americans finished high school, leaving school early was relatively normal. Now a high school diploma is the baseline expectation. Not getting one can close doors before a student is even old enough to understand which doors are being closed.
College increasingly sits on top of that expectation. In 1960, 45.1% of recent high school completers enrolled in college. In 2022, over 60% did. Darius Williams, in his video, directly assumed that many of those seniors in his class—the ones who could not read and write—were not only going to graduate, but go on to college.
Everyone involved has a very good reason to keep a student moving. The parents do not want their child to fall behind. The administrator does not want another dropout. The teacher does not want to destroy a kid’s college prospects over one class. The student does not want to repeat a course or a grade—in addition to the extra work, it can be personally and socially embarrassing.
Critically, the government pressures schools to graduate students, too.
Beginning with No Child Left Behind (2001), federal education policy put much more emphasis on measurable school accountability, including annual reading and math tests in grades 3 through 8 and consequences for schools that failed to make adequate progress. Its successor, the Every Student Succeeds Act (2015), changed the system substantially but retained statewide testing and explicitly incorporated graduation rates and other measurable outcomes into high-school accountability.
The intentions behind all of this are easy to understand. We wanted schools to educate more kids, especially kids whom the system had historically failed. We wanted to know when schools were doing a bad job. But incentives have consequences.
Once graduation rates, test scores, passing rates, college readiness, and similar numbers become markers of whether a school is succeeding, everyone has a reason to make those numbers look better. The question is whether the education underneath the numbers is getting better too.
More Diplomas, Less Education
For a while, it was easy to assume that receiving a diploma and earning it went together. The national four-year public high school graduation rate rose and rose, remember? That sounds like straightforward progress.
But what happens if we look at what high school seniors actually learned over the same time period?
On the 2024 National Assessment of Educational Progress, the average 12th-grade reading score was lower than it had been in 2019. It was a full 10 points below the first comparable assessment in 1992.
Math was also down. The average 12th-grade score in 2024 was three points below both 2019 and 2005. Forty-five percent of high school seniors scored below the NAEP Basic level in math.
COVID obviously hurt. Phones and social media, too. There are a thousand arguments about curricula, parenting, school discipline, social media, funding, teacher shortages, and everything else. I’m not going to pretend one policy explains national test scores. But one thing is certain: We are getting better at handing students diplomas without seeing any evidence that students are leaving school knowing as much as they once did.
There is evidence of grade inflation, too. Not that the national NAEP test itself is secretly getting easier, but that the grades and credits students receive can become easier to obtain.
ACT looked at millions of students and found that, even after accounting for student and school characteristics, average high-school GPA among ACT-tested students rose from 3.17 in 2010 to 3.36 in 2021. Standardized achievement measures did not rise with the grades.
North Carolina accidentally created a particularly clean experiment. In 2014, the state changed its grading scale. An “A” moved from 93 to 90. More importantly, the minimum passing grade dropped from 70 to 60. Researchers found evidence that lower-performing students changed their behavior when the standards became more lenient:
“We find that increased grade leniency led students to obtain GPAs that were 0.27 points higher (11% increase), with no effects on student achievement. Furthermore, students that faced more lenient grading standards increased their school absences by 1.3 days per year (20% increase), which drove students to become more chronically absent.”
Absences increased. Students put in less effort when less effort was required to pass. In other words: Kids respond to incentives too.
It’s worth mentioning the concept of credit recovery: programs that allow students who fail courses to regain the credit, often through online coursework. Credit recovery can do something valuable. It can prevent one failed class from permanently derailing a student. But there is a tradeoff.
One study of North Carolina students found that students who went through credit recovery were about 7% more likely to graduate on time than comparable students who repeated the course traditionally. They also scored as much as two-tenths of a standard deviation lower on end-of-course tests. More graduates. Less mastery.
That does not mean we should abolish credit recovery. It means we need to pay close attention to which outcome we are optimizing: Educating students or giving them diplomas.
Should We Actually Fail More Kids?
The research does not say that simply holding more kids back will fix education. A systematic review on grade retention found, on average, no significant effect. The results varied considerably depending on the age of the student, the way retention was implemented, and what happened during the extra year.
There is also real evidence that retaining older students can backfire. Research has found that holding back eighth-graders increased their likelihood of dropping out. Younger students, though, saw no significant similar effect.
So “just make them repeat the grade” is not a silver bullet for all students at all grade levels. But neither is “just pass them anyway,” which is what schools across the country are currently incentivized to do. The more promising results come when failing a student is paired with an actual plan to fix the failure.
Florida, for example, has required some 3rd graders who cannot demonstrate basic reading proficiency to repeat the grade. But those students do not simply get the same year twice. The policy has included summer reading programs, individual improvement plans, intensive reading instruction, and other support systems.
Research on that policy found substantial short-term achievement gains. The initial test-score advantage faded when retained children were compared with peers of the same age, but retained students later had higher high-school GPAs, took fewer remedial courses, and weren’t any more likely to drop out or fail to graduate.
The principle is simple: if a student has not learned something, tell the truth about it—and then give him another opportunity to learn it. Failure can be more than a speed bump. Failure should be a signal.
How I Learned the Work Isn’t Optional
I know what that can look like because, in a small way, it happened to me. I was homeschooled K-12. At the grand old age of 10, I thought that I might be able to skate by on that. For all of 5th grade, my focus was on three things: football, Transformers, and Star Wars. So I didn’t do my math.
The classes that came easy to me—the history and English—were fine. But math made me think, it made me work, and I did not want to make sure I had division down. I wanted to envision my personal Transformers design and imagine myself as a clone trooper during the Outer Rim Sieges.
We moved halfway across the country before I started 6th grade. During the move, my mom told me I’d failed 5th-grade math. I would be part of a homeschool tutorial, and I’d be in classes with other kids regularly for the first time. But not the math class.
Math would stay 100% at home, and I would go back and finish 5th-grade math before I started 6th-grade math.
I was furious.
Why couldn’t I just get away with it? Why did I have to do the stupid worksheets? It all seemed pointless. I probably told my mom a thousand times that math was unnecessary and I would never use it when I grew up.
She stuck to her guns. Her pride didn’t matter as much as making sure that I actually learned what I needed to learn. It was important that I knew how decimals worked. It was absolutely critical that I learned how to sit down and do the homework even when the answers weren’t automatic.
What my mom did was not complicated educational theory. I had not learned the material. She did not pretend that I had. So I had to go back and learn it.
And, yes: some homeschooled kids are passed because their moms don’t want to admit that their kid had to repeat a class. I’m not here to pretend that homeschooling is the silver bullet solution to the problem, either. But it worked for me, because my mom cared more about my education than about her pride.
I did not fail because I was hopeless. I failed because, given the incentives in front of me, I had decided that maybe I didn’t actually have to do the work. My mom changed the incentives for me.
Make Failure Useful Again
We aren’t trying to reclaim some romanticized era when lots of Americans dropped out of school. Nor a system where we flunk children, shrug, and tell them to try harder next year. We need a system where failure is possible without being final.
If a student cannot read at the end of a foundational grade, there should be a point where the system can say: You are not ready to move on yet. Then we take action. Maybe that is another year. Maybe it’s summer school. Maybe it’s intensive tutoring. Maybe it’s repeating only the course the student failed instead of an entire grade. Maybe schools need more flexible ways for students to catch up without making one failure a permanent mark against them.
There’s not just one silver-bullet policy answer. State by state, we need to do our best to find the best policies for our school system. What we cannot do is continue down the course we’re on—the one that has made Darius Williams, and many more caring teachers like him, cry about how much their students do not know.
The current incentive structure asks teachers to accomplish two contradictory things: maintain meaningful academic standards and keep as many students as possible moving toward graduation.
Teachers suffer in the short term. Students suffer in the long term. Teachers suffer because they can’t do their jobs properly. They live under pressure to make judgment calls they may not agree with and to produce numbers that show progress. Students suffer because eventually somebody stops rounding up.
The college professor does not have to pretend you can read. The employer does not have to pretend you can write. The bank does not care that you never really learned percentages and decimals. At some point, the real world starts grading.
We should also be willing to reconsider how much weight we put on the numbers that create these incentives in the first place. Graduation rates matter. Test scores matter. College enrollment matters. But none of them should replace the underlying question: Is the student learning?
If a policy produces higher graduation rates but weaker mastery, we should be willing to change it. If a grading rule causes students to put in less effort, change the rule. If a credit-recovery program gives students a real second chance, great. If it only makes the credit easier to collect, make it better.
And if holding a student back merely makes him more likely to quit school, then holding him back by itself is not enough either. This is a policy failure, but that does not mean there is one policy solution. It means we have to stop treating every failure as a number that needs to disappear.
Darius Williams should not have to discover in a senior classroom that students have spent years advancing without the skills they need.