Data overload does not always compute: a teacher-turned-researcher’s perspective by Emily Hodge

When I was a new teacher, each day threatened to overwhelm me with its moving parts.  Did I respond to every parent email?  Did I post the homework on Blackboard?  Did I let fifth period leave without reminding them of their homework?  Where did I put Robbie’s paper when he handed it to me late (and why did I never develop better routines, as Wong & Wong told me to do in The First Days of School?) 

Professional development days and frequent meetings—always during my precious planning period—introduced a host of other responsibilities, each seemingly with its own acronym: RTI, IEP, AYP.  Many of these responsibilities were related to data: how to collect it, where to put it, or how to interpret it.  We used one computer program to input our Response-to-Intervention data, and track the effectiveness of our Level 1 and 2 interventions.  We used another program to analyze students’ historical performance on our state test and anticipate their score trajectory for the current year.  We used results from an online practice test to identify “bubble kids” who were close to proficient on the state assessment, as well as specific standards to reinforce in our instruction.  We used yet another program to input our curriculum units to check their alignment with the standards.

It seemed to me that the general principle was, “If some data is good, more data must be better,” and I confess that “data” quickly become a four-letter word for me. I exaggerate, of course, but I suspect that most teachers reading this will identify with how overwhelming the constant emphasis on data can be in schools today.  I nodded my head in Paul Goren’s piece when he started listing the many data-related terms that buzz around a teacher, such as “data-driven decision making” or “performance metric.”

Big data sets can help me to see patterns across a state, or trends among thousands of teachers.

Of course, I felt like I used data to make decisions all the time!  Didn’t I modify my instruction constantly based on student work, student questions, exit slips, and pre-tests?  I made sure to note whose eyes glazed over and when, who seemed jumpy, and who just wanted to put his or her head down and sleep today.  I felt that I could tell if students understood the concept of “main idea” from their writing, without an online pre-test telling me so.  This kind of information did not feel like “legitimate data,” however.  It was personal, subjective, based on my interactions with students, and usually left no written record.

It seemed to me that there were two kinds of data – the kind of data I inputted in computer programs to be collected at the district or state level, and the kind I collected from my students every day.  Sometimes, the central data collection seemed so disconnected from the immediacy of my classroom that it was hard to see its utility.  I occasionally harbored the thoughts of a disgruntled conspiracy theorist.  Who exactly wants this data?  What are they going to do with it?  Why else would they want it except to keep tabs on me?  Is Big Brother watching?    

Even though it was difficult for me to understand the usefulness of so much centrally collected data as a teacher, I recognize its value now as a researcher.  Big data sets can help me to see patterns across a state, or trends among thousands of teachers.  However, despite the fact that my research findings may benefit teachers and students in a general way, I believe that we must also remember the importance of the local data that teachers collect every day to help their students, much of which will never be documented in a data management system.  It is a difficult balancing act, but I believe that we must guard against fetishizing data for data’s sake, while still trying to use data judiciously to help students, as both teachers and researchers.