​OpenAI co-founder Greg Brockman shared this tweet at a conference earlier this year, and it's a great example of how artificial intelligence is going to change the way we teach and learn. Here's another example, in a headline from Education Week:

*Note: ChatGPT is only one manifestation of AI. I highlight it here because it's been in the news and because it's easily accessible for those who want a test drive.

**I am grateful to TED, my former employer, for creating opportunities to learn about AI via TED Talks. I relied heavily on the knowledge shared by TED speakers in creating this piece. I promise I did not ask ChatGPT "Hey, write an overly-long article about how AI will change education using only information in TED Talks" -- though I could have done so. :) ​

Jamestown, Randolph, Carlin Springs, and Claremont. At these four APS elementary schools, something really interesting is happening: for two years in a row, these schools have produced double-digit growth in reading.*

APS uses a tool called DIBELS to measure the development of early literacy and early reading skills at the beginning, middle and end of the school year. I got curious about this question: what percentage of K-5 students are in the DIBELS “green zone” of proficiency in September, and how many more (or fewer) are testing at that level by the end of the year?

In eight of our 25 elementary schools, the number of students at or above proficiency increased by at least 10%. And at Jamestown, Randolph, Carlin Springs and Claremont, that happened two years in a row.

These schools had quite different journeys: one of them started with just 28% of its students in the green zone in September 2021, while another was at 73%. One of them is a Spanish immersion school while another is a neighborhood school. One is part of the  International Baccalaureate program and another operates with a community school model. At one school 72% of the students are Latino, and at another they comprise only 8% of the student population.

So what do these schools have in common—is there a “secret sauce” that’s driving this growth? In APS as in other districts, school and division leaders look at a number of contributing factors, including what kinds of curricula and instructional practices teachers are using; having a highly skilled and high-performing school staff; access to resources that meet the unique needs of that school’s population; and more. It’s not always be easy to identify with absolute certainty what’s working, but it’s really important to try to do so, so that those factors can be sustained and potentially spread to other schools.

Some other things I noticed in the DIBELS reading data:

1.  We have a few schools that experienced significantly more growth in 2022-23 than they did the previous year. What changed? (These schools are Ashlawn, Drew, and Escuela Key. Nice work!)

2.  We have a few schools that showed impressive gains for students with disabilities, while at other schools the percentage of that population in the green zone stayed flat or even decreased.**  This makes me wonder three things:

• whether there are staffing challenges at certain schools (nationwide, school districts are having trouble recruiting and retaining special education teachers)
• how to navigate the nuance in special education reading data (for example, some students receive special education services and are also English learners. One student could have an IEP, be an English learner and be identified as gifted.) and
• how to factor in the fact that some families can afford supplemental private services.

3.  Ten schools showed double-digit growth in the number of Latino students reading proficiently last year, led by impressive 15% growth at Drew. The year before, the growth in proficiency for Latino students at Drew was just 4%. IMO, we should be congratulating the Drew community and asking,”How did you do that?”

4.  Ditto for English learners at Drew: an impressive 16% growth in proficiency over last school year.

5.  Ditto for the 12% growth in the number of Black students proficient in reading at Drew. Oakridge was a close second with 11% growth.


Some of you will take issue with my focus on growth measures. A hypothetical example: 20% growth in a single year may mean that just 10% of students in School XYZ were reading proficiently in September, and by June only 30% are doing so. 30% is still not where we want to be.

I appreciate this concern and wholeheartedly agree that our work isn’t done until 100% of our students at every school are reading proficiently. But for a variety of reasons, some of them not directly under a school’s control, that journey is longer for some school communities than others. (Consider, for example, that at some sites about 30% of kindergarteners start school with entry-level proficiency on DIBELS, while at other schools more than 70% do so.)

This is why school- and student-level growth measures are so important. They honor the hard work that is happening in certain schools even if they’re further from the finish line. Our commitment should be to support and accelerate that growth as much as possible.



*More precisely: APS uses a research-based tool called DIBELS to measure the development of early literacy and early reading skills. DIBELS yields really important data about individual students’ skills, but it can also show us what’s happening at the school level: for example, in this dashboard showing the percentage of students performing at different levels. For simplicity’s sake, the stoplight colors are useful: green generally equates to “proficient” and “above proficient.” Yellow and red are more concerning.


**Note: I am not a statistician but I do understand the general idea of statistical validity. My nod to this concept is: I did not analyze subpopulations that comprised less than 10% of the overall school population.

In America, nearly one in five graduating seniors (19%) leave school with only marginal reading ability. Despite decades of investment in reading research, curricula and teacher education catalyzed by the No Child Left Behind Act in 2001, we haven’t made much progress.

​Here’s what test data from the National Assessment of Education Progress (NAEP) show:

In Arlington, there are two ways we measure high school reading ability. First, juniors take a state-required Reading SOL test at the end of 11th grade. Second, APS just began using the  HMH Growth Measure, which students in grades 3-12 will take three times each year.

In APS, 97% of white students passed the 11th grade Reading SOL test last year. The pass rate for Black and Hispanic students was 20% lower, and nearly 30 percentage points lower for students with disabilities (69% pass rate). Less than half of our English learners (45%) passed this test.

​The HMH Growth Measure categorizes performance as Far Below Level, Below Level, Approaching Level, On Level and Above Level. Looking just at “Far Below Level” and “Below Level,” here’s what the most recent testing reveals:

What does research tell us about the problem?

I spent some time earlier this year reading the research and talking with two local experts: Dr. Olivia Williams, a postdoctoral researcher at the University of Maryland and adjunct professor with the Goucher Prison Education Partnership Program; and Dr. Carrie Simkin, a UVA professor and the director of AdLit.org. Here’s what I’m learning from my reading and my conversations:

1. There’s not a lot of research on high school students who struggle with reading.

In the decades since the National Reading Panel released its report, researchers have published thousands of studies on reading. Yet when I looked online, I could find very few that looked specifically at students in grades 9-12.

I found only one literature review focused on high school reading. Olivia Williams, the author of the review, searched for studies that a) examined interventions conducted on or after 2002; b) measured reading performance both before and after the intervention; and c) studied native English-speaking, general education high school struggling readers. Only 26 studies met her criteria.

Williams notes that even this small group of studies lacked consistent terminology. What does “comprehension” mean? How are we defining “struggling reader”? In the studies she reviewed, “struggling reader” meant everything from being at least one grade level behind to failing an 8th grade state assessment to being at least five years below grade level. “There’s a difference between kids who are significantly behind and those who are just a couple of years behind,” comments Carrie Simkin. “The approaches have to be different.”


2. There’s a disconnect around phonics.
It’s commonly believed that students have mastered phonics by the time they get to high school unless they have specific diagnosed learning disabilities. There’s some research, however, that suggests this might not be true.

As Williams recounts in her review, a 2015 study of reading comprehension among 9th grade struggling readers showed no effect until the researchers looked separately at students with high- and low-level decoding skills. Doing so revealed that the students with higher-level decoding abilities did in fact make statistically significant gains in comprehension (Solis et al., 2015). This is complicated, however: Williams comments that because publishers usually design phonics materials for younger students, their use with teens can be “stigmatizing.”

I asked Carrie Simkin whether high school struggling readers are really students with learning disabilities that have gone undiagnosed. Simkin concedes this is a reasonable explanation for some students, but not all. “Maybe,” she counters, “they have an instructional disability. Our first impulse is always to look at the student, when maybe that student just didn’t get great instruction.”


3. Sustained support may be needed.
Some research suggests that short-term interventions may not be particularly effective. For example, in one study, researchers evaluated the effects of two different reading programs during an intervention year and the year that followed. There were gains in GPAs, grades and performance on state exams during the intervention year—but the benefits disappeared the year following (Somers et al., 2010). Olivia Williams notes that we can’t be sure whether this points to flaws in the interventions themselves, or whether it says something about the need to work with struggling readers over multiple years; more research is needed.


4. Executive functioning plays a role.

Recent research (not specific to high school students) demonstrates how executive functioning skills contribute to success in reading. These skills include cognitive flexibility (shifting), maintaining attention, using working memory, planning ahead, controlling impulses, and more (for a recent literature review, see Duke & Cartwright, 2021).

In one study, researchers looked at students who had poor reading comprehension despite adequate word recognition ability. The study revealed that a third of the students (36.8%) showed weaknesses in executive function but not in their component reading skills, like receptive vocabulary. In other words: for a third of the students in the study, weaknesses in executive functioning appeared to be the primary cause for their reading difficulty (Cutting & Scarborough, 2012).  

This suggests that what some students may need are interventions focused on executive functioning and the root causes of executive functioning delays or impairment, which include things like trauma, autism spectrum disorder, ADHD and depression.


5. Identity is important.

Only one of the 26 studies Olivia Williams examined factored in the social and emotional needs of high school students who struggle with reading (Frankel et al., 2015).  Williams writes: “The repeated experience of failure [by the time students reach high school] takes an emotional toll… Noncognitive aspects of academic development are important at all ages, but especially so in high school, where students with a history of “failure” may struggle with self-efficacy, motivation and engagement.”

What’s more, only one study among the 26 presented findings within the context of race, gender and class (Vaughn et al., 2015).  Williams writes: “Since we know that struggling readers are disproportionately minority, male and poor, it is worth exploring whether different reading interventions are more or less effective with these groups and whether the origins of their struggles demand different remedial attention.”

This is not just about the need to develop culturally-responsive curricula and interventions: it’s about the student’s overall experience of school. Students who are on the receiving end of any kind of “ism” in their school environment (racism, ableism, misogyny, homophobia, religious intolerance, and/or others) are less likely to be comfortable taking risks academically and more likely to be focused on shielding themselves from bias and aggression.



What can we do?

Based on my understanding of the research, here are some of my takeaways:

1. Avoid “one-size-fits-all” fixes. Carrie Simkin counsels, “Struggling readers shouldn’t be lumped together in a single, catch-all remedial class. Instead, through assessment, we can discern what kinds of support students need and, to the extent possible, treat them as individuals.” This might involve a combination of approaches, including special reading classes, tutoring services, virtual programs that students could take advantage of at home or in advisory periods, after school programs and more. Olivia Williams adds, “You have to know what’s available and have the time to plan and differentiate across reading levels.”


2. Look for authentic opportunities in core content classes. With proper support, every high school teacher can help strengthen students’ reading skills. This is particularly true when we think about “disciplinary literacy” and “academic literacy.” Disciplinary literacy involves developing the ways of thinking and communicating that are specific to a particular discipline. Academic literacy involves acquiring the skills needed to read, comprehend and learn from texts dealing with particular subjects (e.g., medical information; financial analyses).

Beyond academic and disciplinary literacy, core content teachers can help strengthen students’ general reading skills, but it has to be done with care. Carrie Simkin shares that we need to focus on “authentic opportunities” to do this—for example, decoding unfamiliar words in a biology class and understanding them by breaking them down into component parts (e.g., omnivore, ectotherm). Simkin suggests, “A literacy coach in a school can help a math or science teacher do this.”

In core classes, we can also apply universal design principles, normalizing audio access to content (and required subject-area assessments) for all students. This accommodation would ensure that students who are struggling readers can access the core content they need to know in an age-appropriate way while they are working in other settings to build their reading proficiency.


3. Build a bigger library. In reading intervention classes, teachers should use a wide range of texts that reflect student interests: it’s “literally anything they care about,” says Olivia Williams. Carrie Simkin adds, “Trust teachers’ professional judgment to curate resources. Give them time to know students and make personal connections.”

This speaks to the essential ingredient of student motivation. Hailey Love, a professor at the University of Wisconsin-Madison, says, “Often when children are perceived as being behind, they’re subject to practices that are actually found to decrease motivation.”


4. Set goals with students. In reading classes, teachers should establish daily purposes for instruction that connect to week-long, month-long and year-long goals created collaboratively with students. Carrie Simkin advises, “We have to ask kids, ‘What’s your goal?’ You have to give a purpose to everything, and kids have to buy into that.”


5. Let students lead. Add in opportunities for peer-mediated group discussions of texts, invite students to generate their own questions, and create other opportunities for students to play meaningful roles in classroom activities. Research supports the positive impact these practices can have (Vaughn et al., 2015; Balfanz et al., 2004; Frey & Fisher, 2014).


6. Recruit and retain exceptional reading teachers. In one study, comprehension gains from the same intervention were twice as high in classes taught by the most effective teachers (Balfanz et al., 2004). Olivia Williams writes, “The success of a reading intervention may not lie exclusively in the strength of the intervention materials or process alone, but may also depend upon a number of outside, less-tangible factors like a teacher’s ability to maintain engagement.”


7. Design special programs that offer struggling readers unique opportunities.  “If placement in remedial reading classes is a tangible reminder of the label of deficiency and serves as an affront to identity,” Olivia Williams observes, “then students may understandably choose to disengage with remedial strategies.”

When we spoke, Williams described with enthusiasm one program in which ninth grade struggling readers tutored second- and third-grade readers who were experiencing their own reading challenges. The ninth graders, who initially reacted with “anger and outrage at being categories as remedial,” grew to view the experience as a privilege. They practiced with the children’s books they used in the program—thus bolstering their own reading skills—so that they would be prepared to work well with their tutorees. The ninth graders scored an average of two grade levels higher on the program post-test and reported higher levels of motivation and attachment to reading (Paterson & Elliott, 2006).

This example reminded me of my own million-years-ago time as a high school English teacher, working with one class of students who were reading below grade level (in some cases, far below grade level). At the time, I wondered what experience I could create for them that would really engage them and reinforce their sense of themselves as worthy and capable. I ended up asking them if they’d like to create an online magazine full of their own writing and artwork—something no other students in the school were doing. They wrote, read each other’s work, peer reviewed, rewrote, created companion artwork, and then hand coded HTML to create the magazine’s website pages. We invited friends and family members to an after school launch party (because it was 1998 and not many of them had computers and internet access at home). The experience was painful for me as a teacher (dial-up modems, only sporadic access to the school’s computer lab…) but highly motivating for my reluctant readers and writers.
Fast-forward 25 years and I’m now a school board member in Arlington. I wonder what we could design, today, that would feel like a privilege and not a punishment for our high school students who are struggling readers.

For example, could we pair reading instruction with corporate/community job shadowing and paid internship experiences? Students would be incentivized to improve their general, academic and disciplinary literacy if they had the opportunity to spend part of their time in “real world” settings where the relevance of their reading skills was immediately evident. Are there summer learning experiences we could offer that look radically different from traditional summer school, combining environmental study (involving reading) with outdoor activities? How could we engage students in the social justice work that so many of them care about and layer in reading instruction?

I’m interested in this subject because it’s creative work—but also because it’s a moral imperative. In one of the articles I read, former NPR reporter Claudio Sanchez recalled visiting a public high school in Tennessee. There, a vice principal told him, “Having a high school diploma does not mean that you can read and write.”

​The United Nations, together with most of the humans on the planet, considers literacy to be a fundamental human right. It’s the very heart of public education. We must do more—and do better.

When we left off in Part One, Shelley Moore’s student had just pointed out that this illustration doesn’t really represent inclusion.  Can you see why?

Shelley’s student pointed out that this illustration is really about assimilation, not inclusion. It subtly suggests that green is the majority and the norm to which we should aspire.

Wouldn’t a more realistic rendering look something like this?

​Another student then chimed in: doesn’t each of us have multiple identities that we’d like to see welcomed and valued in our schools?

​The challenge for us, in operating any kind of community we want to be inclusive—as public education surely must be—is this:

I have many thoughts about how we show that we value all colors, what Shelley calls “teaching to identity”—too many thoughts to list here. There are educators inside APS and in other school divisions who are doing this exceptionally well, and they are my teachers.

Our APS superintendent talks about knowing every student “by name, strength and need.” We have an obligation to identify and address the needs, to be sure. But we’d be doing our students a great injustice if we don’t also help them name and build on their individual strengths and identities, and assure them that our school communities are better because of their presence.