Information is the Key to Data-Driven Teaching


Information is the Key to Data-Driven Teaching by Lucy Kosturko was originally published on Getting Smart.

Data provides us with evidence to guide future actions. An estimated 46 million people rang in January 2014 by seeking guidance from a smartphone app that gathers health and fitness data. That’s one-third of all U.S. smartphone owners. Quantifying our weight-loss activities helps motivate these efforts and also pinpoints which actions seem to correlate with our goals. With weight loss, it’s relatively simple: if x, then y. If you burn more calories than you consume, you’ll lose weight. Gather data on calories going in and out, use this information to make decisions (e.g., padlock your refrigerator) and you’re set.

Data and Personalized Learning

Data also helps guide educators: the more they know about a student--her strengths and weaknesses--the better they can tailor instruction to fit her unique needs for guidance and support. The Office of Education Technology suggests data points “provide teachers with just-in-time feedback on progress towards mastery of content and allow educators to personalize learning pathways for their students.” Informed by the relevant data, teachers can, ideally, provide a student with the support she needs, at the precise moment she needs it, in the specific formats she requires.

Unfortunately, the formula for academic achievement looks nothing like the simple if x, then y. With education the x represents an ever-expanding host of interconnected factors: parental involvement, parent education, socioeconomic status, early exposure to math and reading, spatial reasoning skills, reading skills, self-regulatory skills, self-efficacy, growth mindset, access to technology… And as research continues to expand, so does the list of dependent variables that affect overall academic achievement. Only in the past decade or so, for example, did we begin to call for increased student access to the Internet in school.

The quality of a student’s education is multifaceted, intricate, and interdependent. While high-stakes, summative assessments are often the focus of student evaluation and school accountability, they tend to provide only a peek into students academic performance. In other words, making informed decisions about a child’s education, data collection is much more complicated than tracking calories burned versus calories consumed. It’s easier to get skinny than it is to get smart.

From Data to Information

To paint a comprehensive digital picture of a student’s academic profile, one would have to collect far more data than mere mortals could evaluate. In other words, we’d have a deluge of data, but no information; a series of numbers, but no actionable conclusions; we’d have chaos, without Edward Lorenz’s theory to help us recognize the data’s underlying order. Too often, the result is not enlightened action, but paralysis. After creating daily assessments and then administering, collecting, and organizing the data, teachers have little time left for critical analysis and even less for integrating lessons-learned back into instruction.

Thank goodness for computers: collecting, organizing, analyzing, and displaying data is what they do best. Let’s walk through daily, data-driven instruction made easy for one of the most important skills students learn during their K-12 education: reading comprehension.

Running Records

For early reading instruction, elementary school teachers frequently use running records, which are viewed as a particularly informative data point. In fact, the most effective reading teachers report using running records in their day-to-day instruction. In addition, research consistently supports their impact on early literacy achievement. The breadth of data yielded by running records and other tools (such as comprehension assessments) allows instructors to make informed decisions about specific components of reading that need targeted instruction, the student’s independent and instructional reading levels, and how to form reading groups for guided reading.

However, conducting and collecting running records for an entire class can be time-consuming, rendering them almost impractical in today’s busy classroom. Teachers must sit down one-on-one with each student, listen as students read aloud, and jot down any errors or other behaviors the student makes while reading, without asking the student to stop, slow down, or otherwise stray from their natural reading cadence. After completing this cumbersome, error-prone activity, the teacher sums up the errors, calculates a series of formulae, and analyzes the data by piecing together the various components. Based on this analysis, the teacher is finally ready to make informed judgements about the best instructional course of action for this student.

Phew. One student down; 25 to go. Even the most well-intentioned, well-organized teacher will struggle to to consistently make data-driven, strategic, instructional decisions.

Scanned Document

Luckily, many  classrooms now have mobile devices and, thus, access to Reading Records, a FREE digital solution for conducting running records. By complementing guidance from in-service reading specialists with research-based best-practices, this tool allows educators to gather, evaluate, and analyze data about students’ reading development without cutting corners or compromising precious class time. By recording students as they read aloud, SAS Reading Records maintains the integrity of the data produced by running records. This flexible tool also includes several features to the traditional paper-and-pencil method, further enhancing the benefits of the assessment.

Real-World Impact

Educators have been quick to appreciate these benefits. Consider the following from a teacher at Riverview Elementary School in Raleigh, NC:

I've started giving students weekly assessments through Reading Records and rewarding those who score well. This motivated students to take the assessment seriously, and now they know making a good score is not as complicated as they feared. I choose the Lexile, and the students choose the book. As their' reading levels grow I increase the lexile. This process gives me feedback a lot quicker than sitting in front of each child for progress monitoring. Reading Records found us at the perfect time. The students were starting to mumble and grumble before we began to use this app.

I’ve chosen a single example here to make a key point: apps such as Reading Records do not reduce the complexity of reading comprehension, they equip educators with information, not just data. That distinction is crucial to the goal of supporting data-driven teachers.

Not quite if x, then y, but a lot closer. Now, go take a walk, and think about what I just said. Your fitness tracker will thank you.

Tags 8007

About Author

Lucy Kosturko

Lucy Shores Kosturko, PhD manages product development for SAS Institute's K-12 educational initiatives, a suite of cross-platform offerings promoting data literacy, artificial intelligence and computer science. After graduating with a B.A. in psychology and computer science from Rhodes College, she earned a M.S. in computer science and PhD in educational psychology from North Carolina State University. Lucy lives in Raleigh, NC with her husband and two daughters.


  1. I really appreciate the concept of data used to provide feedback and analysis of student performance and additionally as a tool for further student development. In the physical education classroom, we acquire a large amount of performance related data that is used to guide student development and improvement from one assessment to the next. This concept is one that really has application in the kinesthetic learning lab of physical education

  2. Robin Strayhorn on

    I liked what I read pertaining to your research, and it makes more sense to me to use methods such as those you have described rather than the program that we are presently using. We are using the ARC Reading Program, and student growth is not happening as it should. There are so many students feigning reading, and it cannot be managed well through individual and group conferences, which are so time-consuming and ineffective. By using online tools that will provide data and diagnoses for each learner, it seems more realistic and in line with 21st Century Learning and the tendency toward converting to 21st Century Classrooms. The best thing is that it would provide a more realistic assessment to determine the path students need to take to grow, and the automatic data is fool-proof.

    • Lucy Kosturko
      Lucy Kosturko on

      Thank you for your comment. We need to utilize tools that turn data into actionable information; not simply collect rows and rows of data. Please share your best practices with us in the future!

Leave A Reply

Back to Top