Welcome to #BeyondTheMean! Check out this post to see what this blog is all about.
Very often, while working with educators on issues related to data analysis and research use for decision making, I am met with the terse refrain: “Standardized tests are invalid. This is a waste of time.” While there are certainly legitimate concerns about both standardized testing policy and the nature of the tests themselves, data-driven school improvement is about so much more than standardized testing data.
Think for a minute about the types of data collected in your classroom or school data systems. Educators collect mountains of data! Attendance data, behavior data, intervention data, formative assessment data, homework data, family income data, residence data, language data… the list can go on and on. Instead of focusing your attention on the one type of data that you don’t like, consider digging into data-driven school improvement by reviewing the data that you do find valuable!
Quality data analysis for school and classroom improvement is rooted in the idea of triangulation. When we triangulate data, we use information provided by multiple data points to help us understand a phenomenon. Your data is like a constellation of stars. By itself, it is a beacon that draws your attention; but when you connect those beacons together, you can see a miraculous picture. This is one reason that I preach and teach the importance of exploratory data analysis in the classroom. Exploratory data analysis is a method of data analysis that helps you pull back your view and see the whole picture for what it truly is. The best part about it is that it requires a lot of different kinds of data to give it value.
Here are some real world examples of outcomes I have helped schools see through the exploratory data analysis process…
Once, while helping a school review data related to the impact of virtual learning during the COVID-19 pandemic, we observed that a group of English language learners were not logging into the district’s learning management system. The group’s initial reaction was that the families of these students must not have understood the instructions because they didn’t speak English and the district lacked the ability to translate the instructions appropriately. When we looked at all of the data together, we realized that it wasn’t an issue with their English Learner population, but rather an issue of location. We looked at the student’s home addresses and found that students living in a certain neighborhood were not logging in. The discovery was simple – they lacked internet access. By viewing the whole constellation of data, the school was able to respond to the students needs more quickly and accurately.
Another school I worked with was struggling with the varying levels of readiness among their Kindergarten students. On its surface, it appeared that race was a leading indicator and was suggesting segregated student groupings. When we added a student’s prior placement into the data constellation, we saw that all the less-ready students attended the same pre-school program. It wasn’t a race issue at all, it was a prior placement issue. The school was able to work with that particular pre-school to help them boost Kindergarten readiness.
Finally, I was working with a school that was examining barriers to access for advanced placement and dual credit courses. Again, they thought race was a leading indicator and cited a large volume of research that supported their theory. When we re-did their analysis with a larger data set, we found that poverty was the actual leading indicator and that race was heavily correlated with poverty in their district. Instead of taking a race-based approach to removing barriers they took a poverty-based approach to removing barriers and their enrollment numbers steadily began to rise.
These are just a few examples of the power of exploratory data analysis. Do you notice what they all have in common? A conspicuous lack of standardized assessment data! You don’t have to use standardized assessment data to begin your journey into the world of data-driven school improvement. In fact, in many instances, you shouldn’t use standardized assessment data! The world of standardized assessment is extremely broad, with many different kinds of tests designed to measure different outcomes and inform different types of decisions. If that’s news to you, check out my blog post Standardized Testing is Not a Monolith.
If you’re ready to start digging into data-driven school improvement and aren’t sure where to begin, here are a few resources that can help get you started:
My book, Exploratory Data Analysis in the Classroom, is a step-by-step instruction manual to help you learn to collect and analyze data with a broad lens that emphasizes the whole student. You can read a sample here.
If data analysis skills are a concern, check out the free resources in The Repository. There you will find six auto-analysis tools that allow you to simply upload a spreadsheet, select a column, and see your results. You may also benefit from my video series on spreadsheet fundamentals, which offers screen casts to teach you basic spreadsheet functions that can expedite your analysis process.
Finally, if you’re still unsure about how data analysis can exist outside of the realm of standardized testing, check out some of these other blog posts:
Good luck on your journey friends and let me know how I can help!