This is the second post in a two-part series exploring how education leaders can maximize student learning in an era of big data. In Part 2, we take a closer look at the different types of data school and district leaders should consider using to bring about change. Read Part 1 here.
Now that we’ve identified the three types of data to use as a school leader, let’s take a closer look at how exactly to implement them successfully.
Outcome Data
Think of outcome data as summative data. To become data driven, we have to begin at the end. How will your school define success? This question takes us back to the first obstacle: too much information. Schools that struggle with data often make the mistake of looking at every piece of data and creating an initiative to address it. The result is often that these schools implement programs a mile wide and an inch deep.
Even worse, this lack of direction often confirms the maxim that a person who goes down every road often goes nowhere. As leaders, the first challenge is to work with your team to create a clear vision of success for your school, and then define how you will measure this success. Education has many competing stakeholders and a tremendous responsibility. Without a clear vision, it can become easy to make the mistake of trying to be all things to all people. Limiting your number of outcome data measures is an important first step.
This is the time when you, your staff, and other stakeholders should have the important conversations and answer questions such as, “What is valued here?”, “What does success look like?”, and so on. If done effectively, these conversations can become the foundation for creating a learning community. Your team may want to focus strictly on academic achievement and benchmark scores, or they may bring more of a whole-child focus, adding in surveys about school climate, mental health, and achievement.
Predictive Data
Once we have decided how we will measure success, we want to utilize predictive data. This is the data that will predict our outcome measures. For example, many schools use benchmark testing as an outcome measure, so the next question a data team would ask is, "What drives improved test scores?" This data measure is akin to formative assessments in teaching that allow us to monitor progress.
The following engagement measures can be great predictive data sources:
- Attendance and work completion: These can reliably predict that those students who show up and turn in their work are going to have more success than most.
- Engagement: The amount of engagement can predict which students experience success, and a lack of engagement can predict problem behavior.
- Teacher engagement: Engagement is motivation in action. If your staff is consistently struggling to find the motivation to go to work, you may want to examine the environment and morale at your site.
- Staff attendance and turnover: These can speak volumes about the school climate and certainly carry over to student achievement. If staff are being paid and still don’t want to teach, there's little doubt that's impacting achievement.
In general, I have come to realize that if we can create a climate where staff and students are showing up, doing their best, and turning in work, learning is occurring. Ideally, a data-driven school will identify a small number of outcome data measures and use these to identify a small number of predictive data measures that will drive outcome data. Once this is done, the team should develop a plan to increase those predictive data scores and, ultimately, the outcome data.
Implementation Data
Once an intervention has been decided upon, the team will want to identify implementation data measures. Implementation data helps a school answer the question, "Did we do what we said we would do?" There is no way to evaluate the effectiveness of education programs if we cannot reliably say whether the program was actually implemented as intended.
These measures should be observable and measurable. Can you and your team see or touch this measure? The National Implementation Research Network (NIRN) is a national research collaborative focused on this kind of data. You may want to explore their resources as you work to develop your data-driven processes.
Being a Data-Driven Leader
The exciting part of becoming a data-driven school is guiding your staff and stakeholders through the process of creating a clear vision for your school, and then creating a plan to achieve that vision and put it in place. In even the most ideal circumstances, this process is difficult, and bringing that vision into reality is certain to be challenging. However, when a team has had the opportunity to understand their data and take ownership of future directions, chances of success improve.
Data can either provide your school with a compass to help achieve its vision or it can overwhelm staff and leave your school adrift. A data-driven school needs to understand the amount of data out there and which measures matter to them. This requires a disciplined focus, not allowing all the information and good ideas available to create chaos.
The effectiveness of efforts to develop a data-driven school will depend on the leadership quality of the school. In education, it is easy to be overwhelmed with good ideas. As John Hattie explains in Visible Learning, almost any initiative will show some kind of improvement. Additionally, often the goals of schools will shift with the latest news story, vacillating from achievement to equity to bullying to social-emotional learning to school spirit to career readiness and so forth. Unfortunately, each good idea and well-intentioned focus spreads staff thinner until there is little focus on anything other than summer vacation. However, with a disciplined approach to data-driven decision making, we can create a focus and measures that bring out the best in your team.
The views expressed in this article are those of the author and do not necessarily represent those of HMH.
