Test Scores Are Rebounding. Back to 2019 Still Isn’t Good Enough.

Scores are rising and the headlines are delighted. But the sentence "average scores are up" is the least useful thing a parent can read, because no family lives in an average.

A rebound can be real and still be the wrong thing to celebrate. Recovery means getting back to where you were. It does not mean getting somewhere good.

A note on sourcing before we start: I have no fresh figures to hand you, and I will not invent any. Where I describe patterns, I am giving you my judgement and labeling it as such. Where I describe how the tests work, that is how score reports are built, and you can check it against your own student's report. Wherever you see a number from a news story, go find the assessment name, the test date, and the sample size. If a story will not tell you all three, treat the number as a rumor.

Scoreboard 1: How much is back?

My read: the encouraging movement is concentrated in the youngest grades, the students who spent the fewest years in disrupted classrooms. That fits the logic. A second grader today has spent nearly their whole school life in post-disruption schooling, with more stable attendance and more deliberate early reading instruction.

That is judgement, not a finding I can cite to you.

What you can do is demand the specifics from any story claiming a rebound:

• Which assessment? National and state tests measure different things and report on different scales.
• Which test dates? A fall-to-fall comparison and a spring-to-spring comparison are not the same animal.
• What sample? A national sample and a single district are not interchangeable.
• What is the gap to 2019, grade by grade? "Up" is not "back." A score can climb for two years and still sit below where it started.

If the coverage gives you a rebound number without the gap remaining, it is selling you the first half of the sentence.

Eagle’s pro tip

Policy changes before anyone announces it

The rule that affects your kid is usually settled in a board meeting months before a letter goes home. I read those. Get it while it still matters.

Subscribe today

Scoreboard 2: Who is still behind?

Here is what I think, and I am flagging it as opinion: the students who lost the most ground are the ones least likely to be sprinting back. Older cohorts hit the disruption during years when skills build on skills, and middle and high school math is a tower. Miss a floor and everything above it wobbles.

Reading recovers more easily than math, in my judgement, because students keep reading outside of class and no one practices algebra at the dinner table. Math is where I would look hardest for lingering damage.

Now the arithmetic problem with averages. An average can rise while the bottom stays put. If strong students climb and struggling students flatline, the mean goes up and the headline reads "recovery." The students who needed the recovery most are invisible in that sentence.

This is why the single most important question in the whole story is not "did the average move?" It is "did the bottom move?" Averages are a statement about a population. Your student is not a population.

Students who are behind in the upper grades often need a different structure, not just more of the same pace. That is a big part of why some families look at a flexible format like Virginia Online High School, where a student can spend as long as it takes on a skill before moving up the tower.

Scoreboard 3: Back to what?

This is the part the victory laps skip. Brookings put the U.S. next to international peers, and the uncomfortable takeaway is the one in the title of this piece: 2019 was not a peak. In international comparisons like PISA and TIMSS, the U.S. in 2019 was a middle-of-the-pack performer in math, not a leader that got knocked down.

I am not going to quote you rank positions or point totals from memory, because you deserve a number you can verify. Pull the actual tables from Brookings and the assessment sponsors, and note the testing year on each, because international results arrive on their own schedules and you will often be reading a year or two old.

The logic holds without any figure. If full recovery returns the U.S. to 2019, and 2019 was mid-pack in math, then full recovery still leaves the U.S. mid-pack in math. That is not doom. It is just the honest ceiling of the current goal.

A country can be proud of a comeback and still ask why the baseline was low. Those are separate conversations, and the first one keeps drowning out the second.

Reading your own child's report

Here is the tool. Pull your student's most recent score report and work through this list.

1. Find the scale score, the percentile, and the proficiency label. They are three different things. The scale score is the raw measure on that test's own ruler. The percentile tells you how your student compares to others, not how much they know. The proficiency label (basic, proficient, advanced) is a cutoff someone drew on the ruler. Only the scale score tells you how far your student actually moved.

2. Distrust the proficiency label as a progress meter. A cutoff turns a smooth line into a staircase. A student can lose real ground and stay "proficient," or gain real ground and stay "below proficient." Always look underneath the label at the number.

3. Compare your student to their own prior years, not the state average. Pull last year's report and the year before. Growth against yourself is the only comparison that is about your student. The state average is a different population with a different history.

4. Check percentile movement carefully. A flat percentile in a rising-average year can mean your student is growing but only keeping pace, or it can mean they are slipping relative to a recovering field. Percentile rank without the scale score cannot tell you which.

5. Look for subscores. Most reports break a subject into strands: fluency, comprehension, number sense, fractions, geometry. A decent total can hide a weak strand that will cause trouble two grades from now.

6. Ask the teacher these questions when the score is flat while the average climbs:

   • Is my student growing at the pace you expected this year, or only holding steady?
   • Which specific strand is weakest, and what is the plan for it?
   • What does your own classroom evidence say that this test does not?
   • What would you want to see by the next test, and what would we do if we do not see it?

7. Write down the date the test was taken. A spring score and a fall score are not comparable without adjusting for summer, and a late-year report can be months stale by the time you read it.

If the answers are vague, that is information too. Parents of students in the upper grades should push hardest on the weakest strand, because gaps compound. A student working in English can use a course like English 3 as a concrete benchmark of what grade-level work actually looks like, rather than guessing from a percentile.

Eagle's take

Take the good news. Early-grade students getting more stable footing is worth being glad about, and I mean that.

Then distrust the average, because it describes a crowd your student may not be standing in, and because "back to 2019" was never the finish line. Measure your own student against their own history, find the weakest strand, and act on it before the tower gets another floor taller.