School & District Management Opinion

How Improvement Science Is Leading to Gains at an Urban High School

By Contributing Blogger — February 25, 2016 9 min read
  • Save to favorites
  • Print

This post is by Susan Fairchild, Vice President of Knowledge Management and Chief-of-Staff, New Visions for Public Schools, Christine Han, Knowledge Management Program Officer, New Visions for Public Schools, Jennie Soler-McIntosh, Director of Community Engagement, New Visions for Public Schools, and Elaine Maskus, Resident Researcher at New Visions for Public Schools.

For the past school year, we’ve closely partnered with Stacey King, Principal of New Visions Charter High School for Advanced Math and Science II (AMS II), to get under the hood of her attendance system and refine it using an improvement science tool kit. In Part 1 of our two-part series on designing effective attendance systems, we highlighted the hidden complexity of daily attendance. We also identified what we believe are the core components of an effective attendance system (e.g., intervention criteria that trigger a response on the part of educators, a centralized list of students and their attendance patterns). Now in part 2, we introduce the driver diagram and how we are using it to support the improvement science work taking place at AMS II.


A driver diagram is a tool that explicates a theory of change. There are three parts to a driver diagram: an aim, a set of primary drivers, and a set of secondary drivers. The aim specifies how much an outcome should improve, the timeframe during which the improvement should occur, and the target group the improvement should impact. Primary drivers represent essential system components that most directly impact--or are hypothesized to impact--the aim. Nested within these larger system components are secondary drivers, which detail the specific processes necessary for the core systems to function.

Driver diagrams are important because they explicate the mental models that people carry. When clearly expressed in a driver diagram, assumptions can be questioned, discussed, and changed--especially when there are opportunities to put those assumptions to the test. Driver diagrams are also important because school staff and support organizations--like New Visions--often focus their work on secondary drivers (otherwise known as silos). Driver diagrams uncover where adjacent work is happening and allow for alignment across work streams.

At AMS II, we began our improvement science work with a precise aim and a strategy to achieve that aim. Figure 1 provides a driver diagram that captures AMS II’s specific aim--graduate 94 percent of the class of 2016 by August 2016. The diagram also illustrates what we believe are the high-leverage places to intervene--scheduling and programming; academic interventions; attendance; and curriculum. These four primary drivers represent essential categories of work, but they are incomplete because they fail to specify what the work is. The secondary drivers offer a level of detail that more clearly delineate specific tasks that give greater definition to the primary drivers. (See Figure 1.)

Figure 1.

At New Visions, two different teams are helping support the attendance work at AMS II in different ways. One team is refining the daily attendance process while a second team focuses on building family and community engagement. Both bodies of work exist within the Attendance primary driver, but the secondary drivers differentiate the work (e.g., fidelity to attendance-taking and reporting protocols and regular, consistent family/ community engagement). Without a driver diagram, these two bodies of work might seem independent or detached; and yet, they are deeply interconnected. One driver, fidelity to attendance-taking and reporting protocols, is reactive and responds to both the actions of the students and the staff. The other driver, family and community engagement, is proactive and works to build and deepen relationships.


Daily attendance is a surprisingly complex process. A school the size of AMS II (546 students) generates close to 4,000 data points on daily attendance each day. This level of complexity necessitates better understanding. To do so, New Visions staff paired up with AMS II’s attendance staff to document the minute-by-minute actions that defined the daily attendance process. This included a multitude of tasks--preparing the bubble sheets used by first period teachers to scan into the NYC Department of Education attendance reporting system, integrating multiple data sources and verifying attendance throughout the day, uploading the official attendance, reporting out, parent outreach, etc. These constitute just a few of the daily attendance tasks, which can take anywhere from seconds to close to an hour to complete. We also recorded the number of interruptions to this work throughout the day. The daily attendance work has naturally occurring pauses (e.g., period 1 attendance and period 2 attendance take place at specific times), that, aside the actual work, totals about 70 minutes. But, when we considered interruptions that disrupted the work within each phase of the attendance process, it takes over three hours. If AMS II could save two hours a day by streamlining the attendance process, that potentially creates about 360 hours of freed up time by the end of a 180-day school year.

We hypothesized that increasing the fidelity to the attendance protocols and streamlining the process would create time savings and more time for parent outreach, thereby improving daily attendance. Our aim was to improve fidelity to the attendance protocols by 75 percent by the end of June 2016 by using the three primary drivers visualized below. (See Figure 2.)

Figure 2.

Primary Driver No. 1 - Accurate Attendance Data

Observations revealed that school staff were reconciling 11 different data sources (e.g., students swipe in every morning with their ID cards, students who forget their ID cards manually sign in, students who arrive late sign a late log, seven period attendance data files, cutting data). Based on this observation, we worked with the school to develop an attendance verification tool that integrates the different sources of daily attendance into a single tool. This “micro-tool” supports a specific process previously accomplished manually. With a more automated way of generating daily attendance, we then introduced the school staff to New Visions’ attendance heat map--a tool that is updated daily and contains detailed information on each student’s attendance patterns. You can read more about our attendance heat map here.

Primary Driver No. 2 - Streamlined Workflow for Entering, Viewing and Responding to Attendance Data

When we observed the AMS II daily attendance work, the coordinator responsible for daily attendance was interrupted 42 times for a total of 120 minutes. Based on this data, Principal King adjusted the process and the workflow. Now, instead of allocating the attendance work to one staff member, major attendance tasks are divided across three staff who each oversees a different part of the process. A school aid manages the late logs, inputs data into the verification tool, calls parents, and records outreach. A second school aid manages the period attendance and cutting. Students who potentially cut class are logged into the verification tool, which alerts the Deans. The Attendance Coordinator, now removed from high traffic, disruptive spaces, manages the verification process across the multiple data sources, calls the parents of students who are absent for the day, records the outreach, and reports out to different staff throughout the day on the daily attendance status. This re-designed workflow lives alongside Primary Driver 3.

Primary Driver No. 3 - Process Completion Data and Ability to Evaluate Consistency of Outreach Against Absences

As noted in the first blog, we collected, tagged, and reviewed all of AMS II’s attendance protocols to establish “expected” staff responses against all types of absences. We noticed there was a gap between expectations and reality. In response, we developed a second micro-tool--an outreach audit tool--that allows AMS II staff to more easily display absences against parent/guardian outreach. This allows school staff to get a better sense of how their first line of intervention-- calls to parents are being implemented--and how fidelity to outreach is influencing changes in attendance patterns.

The improvement science work is already changing AMS II’s daily attendance process; soon we will gain a better sense of how these changes are influencing attendance rates. As the attendance team works to improve the school’s daily attendance system, what else may be influencing AMS II’s high year-to-date attendance rate of 94 percent?


AMS II activates a family and community engagement plan at the start of each student’s high school career (i.e., the moment students and families begin the high school application process). As early as eighth grade, a series of meet-and-greets, open houses, and community-building events introduce prospective students to the school. A summer bridge program also prepares students for the academic demands of high school. Simultaneously, Principal King and her team of social workers identify students with low middle school attendance, test scores, and/or grades. Based on these middle school risk factors, they plan supports these students may need during their freshman year. The AMS II team visits the homes of struggling students with the understanding that building trust with families is key to increasing their involvement in the school and their child’s academic success. Remarkably, all this work takes place before a student begins their freshman year.

Once the students reach high school, teachers plan learning experiences explicitly designed to draw upon student interests, connect to real world applications, and deepen students’ engagement. For example, the hip-hop therapy program uses hip-hop music instead of traditional counseling to engage students who are dealing with challenging socio-emotional issues. AMS II’s science team also partners with Rocking the Boat, a local non-profit organization that works with teachers to develop hands-on science activities using the Bronx River. Zach Stellato White, lead science teacher at AMS II, jumped on the opportunity to build an entire unit of study around the Bronx River, explicitly aligning science learning standards to the specific concepts covered through the Rocking the Boat experiences. What could have been an isolated, one-day field trip was the launching point for a thoughtfully crafted series of field trips tightly woven into the curriculum. Students returned from these field trips excited and energized for the work ahead. One student even wrote a letter to the editor of a local newspaper chronicling the Rocking the Boat experience.

The decision to establish relationships early in a child and family’s high school career and to reinforce those relationships often through a sequence of intentionally designed engagement opportunities reflects a uniquely proactive approach--one that recognizes the deep interconnectedness of family involvement, student engagement, and student attendance. In other words, student attendance exists within the context of deep relationships. No less important, AMS II is putting into place a highly responsive, daily attendance system that relentlessly tracks the 3,800 pieces of daily attendance data each day. One driver is in the service of the other; and it is the interaction of these two drivers that propels AMS II to a 94 percent graduation rate.

The opinions expressed in Learning Deeply are strictly those of the author(s) and do not reflect the opinions or endorsement of Editorial Projects in Education, or any of its publications.