The San Francisco Unified School District spent millions overhauling the way its youngest students learn to read. Two years later, roughly half of third-graders still can’t.

The district has invested $5 million in a coaching initiative, pushed out a new standardized curriculum, and set ambitious proficiency targets. But a progress report (opens in new tab) presented to the Board of Education last month found those goals to be “significantly off track” — and pointed to a problem at the heart of the reform: Roughly a third of teachers aren’t using the new curriculum at all.

Third-grade reading proficiency slipped from 53.1% in the fall to 51.8% in the winter, against a year-end target of 62%. Among English learners, reading proficiency stands at just 12.4%. 

San Francisco’s elementary schools had taught reading in a way that asked children to guess at unfamiliar words using pictures and context clues. The new approach emphasizes phonics, teaching students the connection between groups of letters and the sounds used to form words.

Mississippi, once ranked near the bottom of the country for reading, redesigned its approach (opens in new tab) to literacy instruction a decade ago — putting phonics at the center of how its youngest students are taught — and saw fourth-grade proficiency surge from 49th in the nation in 2013 to ninth in 2024. San Francisco’s reformers had that example in mind. 

But to date, the district’s efforts have largely failed. Literacy rates actually slid backward from 2022, when the targets were set. 

To understand why SFUSD may be falling short, we spoke with Rachel Canter, who heads education policy at the Progressive Policy Institute and previously led Mississippi First, a nonprofit that advocated for changes to that state’s education policy. 

She argues that Mississippi’s success went far beyond curriculum changes — accountability for schools and districts was key. That largely hasn’t been the case in California, where a strict statewide mandate (opens in new tab) on phonics instruction was watered down after opposition from teachers unions. 

Canter, who recently published a study (opens in new tab) on why other states have yet to achieve Mississippi’s results, reviewed SFUSD’s progress report.

This interview has been edited for length and clarity.

The SFUSD report shows reading proficiency slipping and most key targets way off track. What does that tell you?

It’s not surprising that implementation takes time. But what I read in the report is a distinction between teachers using the curriculum materials and teachers actually changing their practice. That gap seems to be the central problem.

You can assign a passage from the curriculum and call it “using the curriculum.” But if the work students are doing isn’t rigorous, you’re not getting all the way there. What the report doesn’t say clearly enough is that this is not just on individual teachers. It’s on the structure the district has designed to support these outcomes. Are teachers being coached? How frequently, and by whom? How bought in are school administrators? Who is responsible for making sure teachers actually use the materials — and why haven’t they done that? 

What Mississippi did that was really important was the State Department of Education was actually in school districts helping facilitate those changes all the way down to classrooms — particularly in our lowest-performing schools. I’m not sure I see that same commitment in California.

Is the third of teachers not using the curriculum the main issue, or is the other data around classroom engagement and rigor play a larger role?

Both are true. It’s a red flag that a third of teachers haven’t used the materials at all — the question is why, and who is responsible for fixing that. But the deeper problem is in classrooms that are using the curriculum. Why haven’t we gotten to the change in student work we want to see? That’s where the rubber meets the road. 

What should the school board be doing in response to the report?

First question is: Are board members looking at achievement data at every single meeting? Because if they’re not, they’re missing their most important job. If I were on the San Francisco school board and handed this report, I would look at the student outcome data and say, “We have a problem.” I want to know immediately what the plan is. I want to know what coaching looks like, how administrators are holding teachers accountable, and what teachers say they need differently. And what is the consequence if we don’t meet these goals?

It struck me that for some of these targets, the district had higher expectations of students than of teachers. They were expecting students to grow 10 percentage points but only expecting teachers to change by five. You should be above where you are now after two years. The question is whether you’re being held accountable not just for purchasing the materials but for what’s actually happening when teachers open them.

I’m always the person who says faster, harder, longer — we’ve got to keep going. That was my role in Mississippi. People don’t always like me for it. But somebody has to be that voice. And it’s not clear to me who in San Francisco is playing it.

In Mississippi, how did you handle the tension between improving core instruction and making sure kids who were already behind still got support?

Mississippi didn’t just coach teachers. There was also a process around once you’ve identified children with reading difficulties, what do you do for those children? What are your interventions — reading specialists, tutoring — and how connected is that to what you’re doing in the classroom?

If you’re going to pull money from reading specialists to focus on general education curriculum, yes, that touches every student, and quality tier-one instruction could prevent a child from needing a specialist. But have you really put the structures in place? Because if you haven’t, you’ve made a bet you haven’t backed up.

Were there teachers who just refused, who said, “I’ve been teaching reading for 20 years, and I’m not changing?”

There’s always going to be somebody who says that. But that is a much smaller portion of teachers than the portion who simply don’t know and don’t have enough support. By and large, teachers were very receptive once the state sent trained coaches directly into their classrooms. Teachers want their kids to succeed. If they’re told this is a better way to teach reading, they want to know that. The harder problem is what a principal does when a teacher will not listen. Ultimately, that’s where consequences for adults come in. We’ve given you the training, the support, the time. If you still won’t do it, that has to be dealt with.

SFUSD often couches its policy changes as being in the name of equity, and reading proficiency as just one part of that. What do you think about balancing those interests? 

When people tell me they deeply care about equity, and then they’re not willing to hold adults accountable for solving the problem, they don’t deeply care about equity. I want to see the action. What are you willing to do? Who are you willing to hold accountable?

It is easier to put on a costume of equity than to go do the job. We allow people in our politics and in education to get away with wearing the costume without doing the work. The primary mission of schools is academic preparation. If you are successfully teaching kids to read and do math, nobody is going to care whether you also have an ethnic studies curriculum. The issue is whether you are serving your fundamental purpose. If your English language learners are at 12% proficiency, the belief structure you say you hold has a significant hole in it. Equity is not served by talking about identity if kids cannot read.

What really drove Mississippi was something more basic. We were at the bottom because we had not done enough for kids. The decision that we are going to do whatever it takes meant we instituted policies that said to districts: Like it or don’t, this is how we’re going to do it, and we’re going to hold you accountable. Leaders in California have not made that decision. If they had, it wouldn’t be talking points and press releases. It would be: What are we all going to do, and how are we going to hold ourselves accountable? Our politics are set up for quick wins. The easiest way not to make too many people mad is to say the right things and do nothing, but that’s why kids can’t read.

What did accountability look like in Mississippi? Some of the concern here is that nothing is compelling the district other than good intentions.

We started rating every school and district with an A through F letter grade — very publicly — based on proficiency and growth in reading and math. If your school was in the lowest 25%, you could become a literacy support school, and then the state would dictate your curriculum and how that process looked. You had to go to training, coaching, all of it. And if you were an F for two consecutive years, or a D for four, the state could take over your district — remove the school board, remove the superintendent, and run it directly.

Then there was the third-grade gate for children. A child would not advance to fourth grade if they could not read at a basic level. And because we required parental notification three times a year, parents received a letter with their child’s scores — in big, bright-red letters if their child was below grade level — from kindergarten through third grade. Imagine being a teacher or a principal who’s had to explain to a parent, after all of those warnings, why their child still can’t read. That creates a certain kind of urgency. We gave adults in the system real pressure and real fire to get the job done.