10 Ways Data Can Sabotage Your Teaching

To effectively use data as a teaching tool, you have to understand its forms and limits to inform assessment design and data practices.

10 Ways Data Can Sabotage Your Teaching

by Terry Heick

How do you use data in your teaching? Or assessment?

Functionally, the purpose of assessment is to provide data to revise planned instruction. In seeing what a student understands, the best path for moving forward towards mastery can be planned based on those results—the ‘data,’ as it were.

While there are inherent (and perhaps crippling) flaws with any form of assessment (what is the best way to determine what a student understands?), assessment design is a topic for another day. For now, it can help to simply understand data better—for its benefits and perhaps more importantly, its limits.

Data champions like Rick Stiggins, Richard DuFour and Nancy Love have done wonderful work with—and for–data. Love’s book Using Data to Improve Learning for All is a well-written and thorough starting point for any teacher wanting to better understand how data can support improvement in their craft.

But it’s not that simple.

Using Data In Teaching

Data has become not just the catalyst of school improvement, but the galvanizing force in ed reform en masse.

From No Child Left Behind, down to the data teams and professional learning communities in small rural school districts, data has been called upon to bring science—and light—to the increasingly murky waters of school improvement. In many schools and districts, data isn’t just a tool, it’s the reality everyone–and every effort–is judged by.

That which is measured is treasured (and vice-versa), and many other possibilities fall away.

While this is problematic—assessment design being so far behind available learning domains, spaces, and tools—there are ways the damage done by a data-driven frenzy can be minimized. The key is not dismissing data as a valuable teaching tool but understanding its limits in pursuit of better assessment design and data extraction practices, especially in the form of data-friendly curriculum and instructional design.

So what are some of the ways that data (an otherwise faultless concept) can sabotage your teaching?

Using Data In Teaching? 10 Ways It Can Go Wrong

1. The assessments are imprecise

Test, quizzes, projects, and other assessments that are high on procedural knowledge and low on an ability to uncover what a student understands about the content and standards being assessed are not only not helpful, they’re harmful to students, and barriers to learning in general.

When the test has more gravity than the learner, content, or teacher, something is out of whack.

2. The inferences based on assessment result are limited or erroneous

Even with a well-written assessment, using data in teaching is only as useful as the educator making the decisions about how to best modify planned instruction based on that data. Item analysis—going through each and every question by each and every student to make inferences about what went wrong–is a pain-staking process, and is educated guessing at best.

When it’s done ham-fisted, this is a recipe for disaster–or at least an uneven and frustrating learning experience for students.

3. Assessment is infrequent

In a climate of persistent assessment, each ‘test’ (quiz, exit slip, concept map, drawing, conversation, observation, etc.) provides a snapshot of what the student seems to understand (see #1).

But if such assessments are infrequent, the opportunities to demonstrate progress and mastery are limited. Students struggle with exams for any number of reasons that have nothing to do with content knowledge, from anxiety to simply having a bad day.

The more frequent the assessment, the better. 

4. The assessment is poorly-timed

The right assessment at the wrong time is the wrong assessment. In an ideal environment, each student would get a completely personalized assessment pathway—the right content at the right time, the right assessment at the right time.

This is a challenge in a public school environment where educators may be tasked with an downright overwhelming workload of writing assessments, planning instruction, revising instruction, item analysis, data extraction, and so on.

Even in ‘data team’ environments, this can be too much to take on consistently no matter the can-do spirit of the teacher and school.

5. The data you’re using is dated

The right assessment given at the right time and even leading to the right inferences is of no help if it isn’t used right away. Understanding is perishable, growing and shrinking with experience and time. Data that is dated is spoiled milk.

6. ‘Depth of Knowledge’ isn’t factored

Most educators are well aware of the important of depth of knowledge—Bloom’s Taxonomy, for example.

But if an exam isn’t written with intentional use of such ideas—from low-level recall to higher-level “critical thinking” skills—the data has to be considered in that same limited light.

7. Data is not transparent or accessible to others

Teachers are gluttons for punishment, taking on the whole of the learning process themselves. Not only is this unrealistic, but hurtful to the learners. It takes a village, after all. The more that other content area teachers, administrators, parents, families, mentors, and even community members know about learner performance, the more inclusive the student support system.

8. Data sources are not diverse

From exit slips to tests, projects to computer-based assessments, self-assessments to peer-assessments, district assessments to national assessments, criterion-based to norm-referenced and otherwise, the more diverse the sources of data, the more complete macro view of understanding.

9. Inflexible curriculum that resists data ‘absorption’

If the curriculum is scripted, the pacing guide brutal, the curriculum maps rigid and static, the assessment pre-made and packaged—these factors are not conducive to absorbing and using the data in ways that immediately react to student learning needs.
10. There is too much data

Nancy Love explains in her book, “Simply having more data available is not sufficient. Schools are drowning in data. The problem is (using all of the) data…”

Too much data is worse than no data at all.

Image attribution flicker user tulanepublicrelations

By Terry Heick

Founder & Director of TeachThought, humanist, technologist, futurist, failed philosopher, macro thinker extraordinaire.

Please Login to comment
4 Comment authors
newest oldest most voted
Notify of
Brandy G

Fantastic! Thanks you!


Not many people in schools seem to know much about interpreting data, and the term is applied to virtually anything that can be presented quantitatively in a way that suggests objectivity. There is FAR too much pressure to make “data driven decisions” when few of the people charged with the most important decisions (or those who have the greatest institutional power) have little understanding of students, learning, teaching, or data interpretation.


Great article…

Here’s one anecdotal data point: Teachers are confused and overwhelmed…


Number 7 is true; however, the data is all too often used as a “got you” tool for other teachers, administrators, and support staff. When the group wants to use data to better everyone, rather than punish the teacher, it becomes more effective. Unfortunately education has become such a dog eat dog world that we lose the “it takes a village” mentality. Too many teachers and administrators are out to make themselves look good by making everyone else look “bad”.