The last few lessons discussed the evidence that an awful lot of California's students are learning too little in school, especially relative to those in other states and nations.

Despite the scale of the challenges, there is also good news. Two decades ago, it was difficult to credibly identify high-achievement schools in high-poverty and high-minority settings. Today, it is easy to find them. Schools like those in Compton, once seen as a metaphor for disadvantage, prove that children’s destinies are not coldly and totally predetermined by poverty and race or ethnicity.

Yes, school results have improved, but…

Graduation rates have risen. Test scores have kinda flattened.

Over the long run, there is strong evidence that educational achievement has slowly improved all over the world, as both an effect of economic growth and a cause of it. In America, including California, long-term measurable learning results have generally improved for students in all subgroups. Graduation rates are improving. College-going rates have risen. Disciplinary cases like expulsions and suspensions are becoming more rare.

Unfortunately, test scores aren’t necessarily tracking with these improvements. After decades of slow, unequal progress, scores plummeted in the COVID-19 pandemic. The decline reversed years of gains.

As described in Ed100 Lesson 1.1, America's best tool for measuring educational progress is a national program called the National Assessment of Educational Progress (NAEP, pronounced nape). It enables meaningful statistical comparisons over time and place.

Over decades, NAEP scores generally rose, with big gaps among racial and ethnic groups. In recent years, however, scores tipped broadly downward, leaving experts scratching their heads, frowning in concern and sifting the data for exceptions.

Poverty plays a huge role in the measurable gaps in student learning. Over the long term, scores rose for students from both richer and poorer families, according to the NAEP results. This graph uses eligibility for the National School Lunch Program (NSLP) to represent income status:

Scores on the NAEP exams matter. They help describe long-term changes in the academic capabilities of America's student body. They quantify differences in academic readiness between groups of students. As discussed in Lesson 1.3, scores like these may even predict the future of the economy.

Squaring the Boredom Triangle

To many people, numerical scores feel abstract and hard to interpret. Reactions to the raw scores generally fall somewhere in the boredom triangle between "uh-huh," "so what?" and "I'm sorry, were you talking to me?"

What are Cut Scores and Scaled Scores?

To give their scores some meaning, the NAEP governing board defines a cut score that it labels as proficient for each test. Students that score above the cut score are proficient. Those below it aren't.

The proficiency cut score for each grade level happens to have been set about ten scale points above the grade beneath it. This is almost certainly the source of a widely-used rule of thumb: a ten-point difference in a NAEP score is said to be equivalent to a year of learning. This is a myth. The scale was never designed to be continuous.

In the Ed100 blog
How can test scores be going up and down?

Cut scores with labels can help make a conversation about test results meaningful, but they deserve skepticism. Students know about cut scores from grades: a score of 79 isn't so different from a score of 80, right? But on a report card one is summarized as a B and the other as a C.

Cut scores that describe proficiency rates for groups of students work in a similar way, exaggerating the significance of tiny differences. As we describe in Lesson 9.6, when a normal distribution (a bell curve) shifts and a cut score stays put, tiny changes can seem like a big deal. A lot depends on where the cut scores have been placed, and how they are described. In 2025, the California State Board of Education agonized over the meaning of words like basic, minimal, and foundational.

Educational improvement is normal. And difficult.

Statistical wobbles and wording changes notwithstanding, the evidence suggests that each successive generation of California's kids has been learning a bit more than those before them… just like kids all over America and all over the world. Slow, widespread improvement in education results is normal. For California's students to thrive, they have to do more than keep up.

How can California improve education systematically? Changing one thing at a time isn't enough. In 2013, California began implementing a set of policies to evolve the system in a coordinated way:

• Standards: California's academic expectations were overhauled with the adoption of the Common Core State Standards, discussed in Lesson 6.1.
• Assessments: To support these new standards, the state rolled out new state tests, discussed in Lesson 6.5.
• Accountability System: The state dumped its old academic accountability system, the Academic Performance Index (API), in favor of a more sophisticated performance dashboard approach. As Lesson 9.7 explains, it took years to make the California School Dashboard useful, and the COVID-19 pandemic made the transition even harder.
• Resources: The state replaced a highly inequitable system of school funding with a fairer one by adopting the Local Control Funding Formula (LCFF), explained in Lesson 8.5.
• Federal Law: At the end of the Obama administration, in 2015, the US Congress replaced the No Child Left Behind Act (NCLB) with the Every Child Succeeds Act (ESSA). A muscular law, NCLB had required school districts to show continuous improvement in student test scores, or take politically difficult actions. ESSA substantially removed this federal pressure. Learn more in Lesson 7.2.

In 2018, a project known as Getting Down to Facts II coordinated research into the effects of all of this interrelated change. Based on a national comparison titled A Portrait of Educational Outcomes in California, students’ academic achievement improved a bit faster in California than in other states, but with big gaps.

What is a student subgroup, and why does it matter?

A key function of school accountability systems like the California School Dashboard has been to highlight differences in results among student subgroups. For example, if there are a significant number of students of a specific ethnicity in a school or district, the California School Dashboard can focus on the results for just that subgroup. Differences in subgroup scores can suggest where a school or district ought to focus its attention. Like it or not, these differences also deliver a bit of "shame" motivation for schools to remember all of their kids when evaluating their success.

Schools that "beat the odds" demonstrate that all kids can learn, with the right support

But shaming isn't the only reason to look at subgroup scores: they can also show reasons for pride, and for learning, because some schools beat the odds.

Schools that beat the odds are important because they show that the kids are not the problem. They prove that, given the right support, every child can learn. They provide evidence that investing in kids really can make a difference, even where it seems hard. They transform the suspicion that "those kids can't learn" to "those kids won't learn unless..."

It's an important difference.

What are the ideas, approaches, programs, interventions, investments, and inspirations that can lift California’s student achievement, boost economic growth and make a crucial difference in children’s lives and America’s future?

Answering that question is the focus of the lessons ahead. But first, let's back up a little. Education isn't all about scores, or even about economic competition. What is it for? And has our view of education's purpose changed over time?

Updated June 2025