Updated 6/28
The key resources for fixing troubled schools are the people who already inhabit them.
That’s a bald statement, and it rings too much of a crazed street evangelist with a long beard holding a sign saying “Heal Yourself.” But consider what we know about approaches to fixing failing schools.
For almost a century, American public education has operated according to the industrial precepts that outside experts could import proven programs, and that they would turn around failing schools. These practices have left an expensive trail of tears and frequent recrimination.
For the past 20 years, industrial fix-it strategies have been joined by the accountability movement’s belief that calling students, teachers, and schools a failure will motivate and empower them to succeed.
Neither has worked well.
Carnegie’s ‘Networked Improvement Communities’
Instead, I’m attracted to the Networked Improvement Communities techniques developed by the Carnegie Foundation for the Advancement of Teaching. I’ve written about the Carnegie Summit, and last month I visited their Palo Alto headquarters to learn more about how aspiration links with method.
High above the Stanford golf course, their modernist headquarters looks toward San Francisco Bay with sweeping views that give rise to thoughts of “how does anyone get work done here?” But they do.
Under the leadership of Anthony Bryk, who blended his analytical expertise with boots-on-the-ground experience in Chicago’s school reforms, Carnegie has devoted itself to the learning-to-improve method since 2008. Its essential steps are captured in the chart below, on the foundation’s web site, and in the book that Bryk co-authored with Louis Gomez, Alicia Grunow and Paul LeMahieu.
The sixth of the Carnegie learning-to-improve principles is acceleration through a Networked Improvement Community (NIC): moving from what an individual educator knows to what thousands of educators know, combining and focusing their knowledge on specific problems.
Carnegie adopted the name from Douglas Engelbart, a pioneering scientist credited with inventing the computer mouse, whose work ultimately led to social media and cloud computing. Later in his career, he concentrated on how to build systems that help people get smarter about their work.
Building Blocks of Improvement
Consider the basic building blocks. There’s what you know. Then, there’s what you and your colleagues know. In education, just moving from the typically isolated teacher to a group of teachers is a huge reach. Learning Walks, Teacher Study Teams, and the Professional Learning Community are built to accomplish these interactions.
Now suppose that it were possible to connect what you know with what thousands of others know. Carnegie is building an alternative to the industrial-era knowledge transfer, consultants or academics distill knowledge and pass it back to teachers in bite-sized bits via professional development. As fifth grade teacher Jackie Funkhouser, whose reaction to working in a networked community is captured on the foundation’s web site, puts it, “I thought a researcher was someone who collected data, wrote about it, and told me what worked. Then I was the person who took the information and used it. Now I see that I’m a researcher too. I can collect data and use it in my teaching to make it better.”
Networks at Work
The foundation’s work provides several examples of NICs at work. The Math Pathways projects focus on the student graveyard of developmental math: those often awful courses that students who haven’t been successful at math are required to take in order to get into credit bearing math courses. They’re “where aspirations go to die,” said Bryk.
Math Pathways have engaged 20,000 students in 15 states, and have produced double or triple the success rates of conventional developmental programs.
For other applications of NICs, see the Student Agency Improvement Community (funded by the Raikes Foundation) and the Building a Teacher Effectiveness Network. A new report shows how the Fresno Unified School District used a NIC to expand college access for their graduates, expanding the range of schools students applied to by more than 50 percent.
Carnegie has been connecting the network improvement idea with online computer technology to allow teachers and other educators to try multiple plan-do-study-act cycles, essentially little experiments that can be repeated or abandoned, depending on the results. Jojo Manai and Susan Haynes, two of the developers, guided me through the system, which is called NILS, shorthand for Networked Improvement Learning and Support platform.
NILS helps in “surfacing tribal knowledge” Manai says. Teachers carry around a great deal of tacit understanding about their craft. The NILS protocols help make what teachers know both concrete and public.
Haynes and Manai demonstrate the system’s structure using the example of the Tennessee Early Literacy Network, formed to improve the proficiency of the state’s third graders. The network’s designers hypothesized that there would be three primary drivers: the way that schools related to families and communities, standards-aligned instruction, and support for struggling students.
Feeding into these are secondary drivers (see screenshot from NILS demonstration) and specific change ideas. One of these is the student journey map, essentially a careful chronology of a student’s struggles to read. Collected from students throughout the state, the maps included the usual background issues—homelessness, death or disability of a parent, and drug or alcohol abuse—and also interviews with students about their experiences learning to read and write.
The maps revealed common problems, such as the lack of coordination between educators, but also problems specific to individuals. One student had no books at home, so the teacher started sending them home with the child.
A member of the Tennessee project, or someone like me, who is trying out the system, is invited to initiate his or her own small experiment by initiating a plan-do-study-act cycle, a tool of improvement science that dates back half a century to the work of Edwards Deming.
I tried it, posing the question of whether I would get better results in getting parents to come to meetings if I invited them personally rather than sending out a broadcast email or note. I vowed to keep track of participation at the next meeting.
Playing With The System
Then, the system worked to keep me on track by asking me when I planned to complete this experiment, how I would carry it out, and what resources I would need. It kindly provided me with some resources from the people at the hub of the system at Carnegie. I didn’t have to pay any attention to them, but I could.
In the ‘Do’ phase of the experiment, NILS works like an on-line lab notebook, providing me a date to enter the data I had promised to collect: how many people came to the meeting, and allowing me to make any observations I wished. I saw that not all the parents knew one another, and that those who were isolated seemed uncomfortable.
By the time that I got to the ‘Study’ part of the cycle, I realized that there was a big flaw in my method. Just counting attendance didn’t distinguish between those who were personally invited and those who got the email, and I hadn’t kept track of how many people I had personally invited who had not attended, and some of the people I had invited to the first meeting didn’t show, but they came to the second meeting.
In the “Act” section, I concluded that the process was encouraging and that I should adjust it a bit and try again.
All Together Now
Now, imagine this all connected. My little experiments may resonate with an educator across the state, and I may learn from someone in the next county or country. Haynes provides the example of the Student Agency Improvement Community. Its Stress Reappraisal Activity has been tested across schools in New York, California, Virginia, Minnesota, Georgia, Washington, Hawaii, Ohio, Wisconsin, and Delaware, beginning in one classroom with one teacher, and expanding to whole grades and schools.
“Teachers adapted the change idea’s implementation in their classrooms to correspond with certain times of year, when students received the most benefit from learning about how reappraise stress, such as before taking an exam, prior to a public speaking event,” said Carnegie staff member Rachel Beattie.
The specialists at the Carnegie network ‘hub’ then “curate this wisdom and share it back with the network, which prompts ideation for further changes,” she added.
When this happens, structured networks can spread and scale improvements.
The secret is in the iteration, little things leading to big ones. As Manai told me, getting people to “break through the chains of accountability and see the system,” requires that teachers and other front line educators see the connectedness of their work.
The cycle of experiments also requires honesty. Some, probably many, of the little experiments don’t work. It’s hard for educators to admit that something isn’t working. As systems learning expert Peter Senge, who spoke to the Carnegie Summit in March said, if you say you don’t know the answer, “it’s like undressing in public.”
Taking Practitioners Seriously
As I talked with Bryk and LeMahieu in their offices I was taken with the dignity and status that the Networked Improvement Community approach offers to practitioners. When teachers work on a project at Carnegie, they are not simply stakeholders; they are valued as knowledge producers. They work together and learn from each other in ways that are rare in today’s top down education establishment.
I believe the Carnegie approach can add heft to the series of reforms that the folks in Sacramento call ‘The California Way,’ this state’s departure from leading reforms with test-and-punish accountability. Think what applying the Math Pathways techniques could do for the 1.37-million English Learners in California public schools.
There’s an understanding at Carnegie that massive changes are needed, are on the way, in education, and there’s also a method for getting there.
Graphic diagram: Carnegie Foundation
Update to correct attribution of quote and editing errors.
Save