Failing Fast to Reduce Habitual Absenteeism

Fail fast is an expression that is thrown around with some regularity today (mostly with Silicon Valley types). The basic idea behind the concept is that we want to stop wasting time doing things that don’t have an impact (or “creating products that nobody wants” in Silicon Valley speak). Instead, we are meant to be learning from our failures and rapidly improving our ideas based on what we learned. When I work with teachers on the Innovation Box process, I encourage them to fail fast to accelerate learning. For the most part, teachers seem comfortable with the point I am trying to get across, but I realized after a recent session I don’t illustrate what this means with a real-world example. The purpose of this write-up is to tell the story of how one teacher has been feeling his way through the process of developing an innovation and how the concept of fail fast shows up in his work. Matt Prunier is a high school social studies teacher at an alternative school in New Hampshire. When Matt started with Innovation Box he framed his challenge as “How might we solve habitual absenteeism?” As Matt noted during the Innovation Box launch in his district, “If a kid doesn’t come to school there is absolutely no way they will graduate and for many kids this is their last chance at school, so we have to figure out how to get them here.” Matt was aware that this was a huge problem and lots of smart educators before him had tried to solve it. What’s more, he knew that most previous efforts had made little difference or failed completely. Matt determined he needed to approach the problem from a different perspective to have an impact.


Matt was aware that past efforts to improve student attendance by using text messaging had mostly failed, but Matt saw his approach as different from previous efforts in two ways. First, instead of messaging parents, he would message students directly (through a school-approved app). Second, instead of sending an informational message (e.g. an attendance report for the week), he would send a message of support that included an invitation to return to class. After a few weeks of collecting baseline attendance data at the beginning of the school year, Matt launched his text messaging experiment. Matt started by texting five different students on the day they were absent. He reported nearly immediate success when the first student he targeted texted back and returned to school the following day (see images). However, the experiment was not a success across the board. In fact, out of the five students that Matt texted, two responded positively and returned to school the next day. For two other students, they didn’t respond, and they did not immediately return to school. For the final student, she did not respond and did not return to school. Early on Matt made a key assumption; if students felt like they were missed at school that they would want to be there and would be less likely to be absent. Based on this assumption, Matt decided his idea was to connect with kids via text message when they were absent to reinforce that they were missed in the classroom. To test this idea, Matt randomly selected a group of students that he would text when they were absent. An important component of Matt’s idea was that the message would be a positive message of inclusion in the classroom community. For example, “we missed having you in class today.”

Matt found the early test results confusing. Was this a failed experiment or something else? For at least two kids the experiment proved successful, they returned to school. For some reason it didn’t work with three others. As he reviewed the students that he had selected for the experiment, he observed that the students that had returned immediately after receiving one text message were students that he had already developed a trusting relationship with. For two of the three students where the text messaging approach did not work, he had started to develop a relationship, but it was still in the early stages. For the student that never responded and did not return to school, Matt had not developed a relationship at all. This observation led Matt to make a new assumption, texting and inviting students back to class only works if you have already developed a relationship (or to remind or reinforce a feeling that students may already have about being welcome in the classroom and school). A different approach may be required to intervene with students where a deeper relationship is not present. Matt’s early experiment helped him start to define the limits of the impact of his idea for addressing habitual absenteeism. It seems that directly texting students and inviting them back to class on a day they have missed is effective at getting students to return to school if there is already a positive teacher-student relationship established. The early results have encouraged Matt to continue this innovation for all the students where he has already developed a strong relationship. Matt has also learned that a different approach may be required with the students where a relationship is not as string. Matt is not abandoning the approach of connecting through text but plans to alter his messages to see if students respond differently to broadly pro-social messages (e.g. “most students do not miss school”) or informational messages (e.g. “you have missed 8 days since the beginning of the month”). The principle of fail fast is really learn fast. The failure is not what matters, but it is the opportunity to learn and improve that failure affords us. Matt’s first experiment only took a matter of weeks for him to learn something interesting. Matt is learning enough with these early experiments to make decisions about how to improve and iterate his idea so that it has a greater impact on students. Matt is learning quickly.

Joseph MillerComment