Do We Believe in Data?

As anyone who reads a newspaper knows, numbers and statistics are easy to manipulate. Sentences that begin with the phrase “studies show” are subject to skepticism, and for good reason. Data analysis can be met with a level of wariness born out of the fact that drastic action is sometimes taken too quickly because of information that is not plentiful or accurate, or that is not applicable across a wider spectrum. Being doubtful of data bias is natural, but for teachers, disregarding the validity of everything we learn about student performance can get in the way of achieving growth measures. Instead, we must consider how multiple data points work together to illuminate a path forward for increasing student achievement. 

Classwork

Teachers tend to believe the most in what students show us with their performance on assignments or projects, which makes a lot of sense. After all, why would we not believe in the validity of the tasks we design for students? Presumably, any work that kids complete under the guidance of teachers should be considered important. However, the reliability of performance data can become skewed when too many assignments are graded for completion or practice, rather than for content acquisition. For that reason, aligning classwork to learning outcomes is the best way for classwork to be an accurate data point. 

For example, suppose that a class is learning about the customs of a particular culture. There might be several skills to focus on, but the teacher must prioritize one or two as targeted learning goals that are explicitly shared with students. One possible focus might be to define specific vocabulary terms that are relevant to the culture being studied, which means that any classwork collected should specifically measure how well students are achieving the objective to define cultural terms in a way that demonstrates understanding. That way, any classroom data that is collected is geared toward helping students meet criteria for success that have been clearly laid out for them, and the teacher can be confident in the reliability of the information that is gathered.

Assessments

When we think of how students are assessed outside of more formalized settings (more on that shortly), two types of measures come to mind: formative assessments, which are typically less formal and administered throughout a learning process, and summative assessments, which occur at the close of a unit and have inherently higher stakes. While both are important, formative data is particularly helpful in letting teachers determine where student progress sits in relation to a desired goal. Ideally, students should be formatively assessed each day in some way, whether it’s simply through answering one question at the close of class or by completing a short assignment that indicates how well they understood the daily objective. 

In addition, there is a broad misconception that data must be quantitative to be meaningful, which couldn’t be more wrong. Teachers use a variety of qualitative formative data to assess student understanding of any given concept. For example, exit tickets are a tried-and-true method for getting quick information about what students understood from a lesson, and their responses are a vital guidepost for establishing instructional next steps. Without both types of data, teachers do not have enough information to get an accurate picture of where student comprehension lies. 

Standardized Tests

We know only too well that students are often tested via external measures, which are standardized assessments. While the general format of the test and the content standards that will likely appear might be somewhat predictable, the assessment itself is secure with outcomes that are largely out of any teacher’s control. Of all the data types teachers look at, standardized tests are widely considered to be the least reliable for a variety of reasons. For one thing, specific results are often not visible to teachers or students, so there is no way to go over answers to questions or learn from areas of struggle. In addition, national or state tests might be disconnected from a more localized curriculum.

Even with the potential drawbacks of standardized assessments, they do provide an important way to look at student performance across a variety of grade-level, standards-aligned criteria. In addition, the ability to examine variations that might exist from region to region (or even more significant, state to state) can highlight where kids are being held to inappropriate expectations. Having a concrete, objective measure of student success is therefore an edge standardized tests possess.

Data might feel like a dirty word because it comes with a stigma. However, when we look at what both quantitative and qualitative measures have the power to tell us, the information can only be helpful in advancing student achievement. Deciding that data analysis is a waste of time is akin to treating a medical condition without looking at blood work, or flying a plane without a navigational system. The only way to accurately know what students need is to measure their progress with multiple types of data that, when put together to form a bigger picture, give us the information we need to move instruction forward in a way that benefits everyone.

Written by Miriam Plotinsky, Education World Contributing Writer

Miriam Plotinsky is an instructional specialist with Montgomery County Public Schools in Maryland, where she has taught and led for more than 20 years. She is the author of Teach More, Hover Less, Lead Like a Teacher and Writing Their Future Selves. She is also a National Board-Certified Teacher with additional certification in administration and supervision. She can be reached at www.miriamplotinsky.com or via Twitter: @MirPloMCPS

Copyright© 2024 Education World