1703 North Beauregard St.
Alexandria, VA 22311-1714
Tel: 1-800-933-ASCD (2723)
8:00 a.m. to 6:00 p.m. eastern time, Monday through Friday
Local to the D.C. area: 1-703-578-9600
Toll-free from U.S. and Canada: 1-800-933-ASCD (2723)
All other countries: (International Access Code) + 1-703-578-9600
by Daniel R. Venables
Table of Contents
In my 30-plus years as an educator, I have realized that many teachers and administrators find looking at and making sense of data an unpleasant task for many reasons. Some find it just too intimidating to plow through table upon table of numbers and try to make sense of them. Others have an unspoken but underlying belief that data do not really tell them anything important.
Others experience a painful flashback to the math and statistics courses with which they struggled in college. And for others—arguably, most—there isn't enough time in their busy workday to submerge themselves in graphs, tables, and charts to produce what they think will be, in the end, facts about their students' learning that they already knew or at least suspected. Add to this the fact that for other, more seasoned teachers, reviewing data was not done for many years in education, so their prevailing attitude is "We got by just fine without it."
Of course, in our post–No Child Left Behind (NCLB) and present Common Core State Standards (CCSS) era, we know differently. We know that to not review state and CCSS data is to put our students at certain, even measurable, disadvantage. Data aren't everything, to be sure, but they represent the best, most reliable way to see where students currently are in their learning and to identify instructional actions to get them to where we would like them to be. If we don't use data to do so, we are left to rely on our hunches for making important instructional decisions. As we'll see, hunches are sometimes wrong. Even if our hunches are wrong only 25 percent of the time, that's still a lot of time to spend trying to fix problems that may only be symptoms of greater, more encompassing issues that may well go undetected. That's time we don't have. Educators are experiencing an increasing sense of urgency to fix problems related to student achievement and can't afford to misdiagnose students' learning gaps. Now more than ever, we need to be right in what we believe those gaps are and in what we believe their root causes are. A careful look at quality, relevant data can help us achieve this goal.
In helping schools review and respond to their data, I have seen a tendency for administrators to provide year-end data to teachers and let teachers decide what to do with it. This approach often results in teachers and teacher teams giving the data a cursory look, with no intent to use the data to make instructional changes. This pattern does not reflect teacher disinterest in improving student achievement as much as it does their general discomfort with turning numbers into action.
This feeling has partly to do with the size and kind of data teachers are often asked to review. For example, large macrodata, such as end-of-course test data, seldom produce specific, nuanced instructional changes. In the absence of more detailed microdata, teachers respond to large data with an equally broad brush: "We can see our students with disabilities are sorely lacking in proficiency," or "Once again, our girls have outperformed our boys in reading comprehension." Little useful information is gleaned by taking such a cursory look at the data; therefore, teachers will revert to the status quo, and no real instructional change is likely to result. It's not that teachers resist changing instructional practice (although some do), but more that, in many cases, the data are revealing things the teachers already knew.
The problem lies in viewing only the macrodata and stopping there. It turns out that the kinds of information that can provide the impetus for real change lie in the details—the microdata. A good place to start, macrodata may tell us what is happening and perhaps even to whom it is happening, but to find out why it is happening and how to fix it, teachers and teacher teams need to turn to the more detailed and frequent microdata. To help review and implement this sort of data, I have developed the Data Action Model.
The Data Action Model is a systematic process for reviewing and responding to data. I have already noted that teachers are generally good at looking at data; the extent to which that review translates into changes in the classroom is quite a different matter. The Data Action Model helps. It begins with a structured look at the macrodata and then drills down into the smaller microdata, often bringing in additional relevant data to fully understand the root causes of any learning gaps that are uncovered. Then learning gaps are linked to corresponding instructional gaps, and the process culminates with a goal-driven action plan, complete with a metric for assessing the effectiveness of the plan once implemented.
The Data Action Model is composed of three main phases: Gathering and Reviewing Data, Identifying Gaps, and Planning for and Evaluating Action (see Figure 1). Each of these phases is incorporated into five data meetings. Each data meeting is broken down into manageable steps for teacher teams to follow. The process is cyclic in nature; once a team has met its data-driven goal at the end of one cycle, the team begins anew and explores new data (or returns to the original data set) to uncover other learning and instructional gaps needing remedy.
One cycle of the Data Action Model normally requires five separate hour-long meetings from start to finish. Depending on the nature of the data used in the first meeting, this meeting time might be shorter. The Data Action Model normally spans approximately nine weeks (one grading period) if the meetings are held weekly. This is because there is a four-week Implementation Period between Data Meeting 4 and Data Meeting 5 (see Figure 2).
What to Do (and in what order)
Data Meeting 1
Reviewing existing data and asking questions
Data Meeting 2
Triangulating the data
Data Meeting 3
Determining gaps and goals
Data Meeting 4
Planning for action
Conduct a Strategies Search; then
PLC Meetings 1–4
Implementation period (four weeks)
Look at student and teacher work; troubleshoot obstacles; look at texts, research.
Data Meeting 5
Evaluating success and determining next steps
It should be pointed out that the data meeting template in Figure 2 presupposes that the data initially reviewed by the PLC are broad data, as in end-of-course or Common Core assessment data. These data are most likely reviewed in August or September, at the start of the school year. However, teacher teams will revisit this template once the year has begun (but before new macrodata such as Common Core assessment data are in); in this case, most PLCs are reviewing district benchmark data, teacher-designed assessment data, or pre-test data on an upcoming unit they are planning. With pre-test data or data derived from diagnostic assessments of untaught material, the cycle of meetings varies, as shown in Figure 3 (also see "Using backward-looking data" and "Using forward-looking data" in Data Meeting 5, and Figure 12 on page 111).
Reviewing pre-test data and identifying learning gaps
PLC Meetings 1 and 2
Implementation period (two weeks)
Note that when teams begin with pre-test data on a unit they are preparing to teach, the cycle is shortened considerably. Most units are approximately two weeks from pre-test to post-test. That means the full data cycle is at most four weeks, often shorter. It also means that teams should allow ample time between gathering pre-test data and teaching the unit so that they have time to decide on instructional strategies before actually teaching the unit. Otherwise, teams tend to rush straight from identifying gaps to teaching standards the way they have always taught them, without regard for what might actually be the best way to teach those standards. The result, too often, is that no actual instructional change occurs.
More than once, I have referred to the notion of cycles. Although the length of the cycles varies considerably—as in the case of the cycles depicted in Figures 2 and 3—the Data Action Model is necessarily cyclic in nature. Data are reviewed, gaps are identified, goals are set, and action plans are implemented, and that cycle is repeated as new data are reviewed, new gaps are identified, and so on. This is to be expected.
That said, it is also somewhat restrictive to think only in terms of cycles. The danger is that teacher teams will try to pigeonhole cycles into fixed time frames that, in reality, may or may not be applicable to the content targeted in any one cycle. Cycles vary in length and must remain flexible if they are to be useful. Teams should be aware of this point as they map out their cycles.
Initially, I organized this book by chapters that each described a step in the Data Action Model. However, teacher teams who actually used the model were more interested in what should be happening at each meeting—not at each step. They were spot-on. If the Data Action Model is to be truly useful to teams trying in earnest to follow it, it should not be organized by steps but, rather, by meetings.
Based on this important feedback, I completely reorganized the contents of this book to be set up primarily by meetings as opposed to steps. The result is more PLC-friendly, as it delineates what should be accomplished at each PLC meeting. Data Meeting 1, then, is the first chapter; Data Meeting 2, the second; and so forth.
Throughout this book, I use the terms PLC and teacher team to refer to the same thing. While I acknowledge that the term PLC encompasses some characteristics that may or may not be present in all teacher teams, for our purpose here, I use the two terms interchangeably. The same is true for my use of the terms PLC coach and PLC facilitator, both of which I use to refer to the teacher who leads the teacher team. Finally, I often refer to the Common Core State Standards (CCSS) simply as the Common Core, as they tend to be called in most education circles these days.
In the chapters that follow, we will explore these phases in detail and illustrate implementation of each one with vignettes of schools that are using the model. (All vignettes are fictional, describing schools that are composites of actual schools, and all names used are fictional.) Before we turn to what should happen during Data Meeting 1, it is important to be clear about what we mean by data, where we find them, what types exist, and how those can be used to improve teacher and student performance.
Copyright © 2014 by ASCD. All rights reserved.
No part of this publication—including the drawings, graphs, illustrations, or chapters, except for brief quotations in
critical reviews or articles—may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopy, recording, or any information storage and retrieval system, without permission from ASCD.
Subscribe to ASCD Express, our free e-mail newsletter, to have practical, actionable strategies and information delivered to your e-mail inbox twice a month.
ASCD respects intellectual property rights and adheres to the laws governing them. Learn more about our permissions policy and submit your request online.