Amanda Datnow and Vicki Park
To support teachers in discussing data, school leaders should ensure that the proper foundations are in place.
Schools and districts that successfully promote data-informed decision making often have one key feature in common: dedicated collaboration time for teachers. Whether teachers meet weekly or monthly in horizontal, grade-level groups or vertical teams that span grade levels, teacher team meetings are key to effective data analysis. As one principal noted, “Without collaboration and collegiality, data use is impossible.”
In data-focused team meetings, teachers may eagerly examine a wide variety of information on student learning, consider multiple interpretations of the data, help one another grow, and support one another in developing meaningful instructional improvements. These meetings approximate the professional learning community moniker that many districts have given them.
However, not all teacher team meetings that focus on data use are so productive. Indeed, teachers sometimes feel as though they’re engaging in contrived collegiality (Hargreaves, 1994). They may find that the data on student learning they’re examining are irrelevant to their practice. They may neither trust their colleagues nor view them as sources of support in instructional improvement.
Whereas teachers in the first kind of meeting often feel they never have enough time to work together, teachers in the second kind frequently glance at their watches, wondering, are we done yet?
How can leaders support teachers in having more productive team meetings that focus on data? In the course of our research on data use over the past decade (see Datnow & Park, 2014), which included many observations of teacher team meetings and interviews with teachers and administrators, we’ve distilled five essential components for quality conversations about data.
Component 1. Students are the shared responsibility of everyone.
Think of it this way: If you were only concerned about the achievement of the students in your classroom or if you felt competitive with your colleagues, you would have little motivation to assist your colleagues in examining data or developing new instructional strategies. School leaders play an important role in cultivating a belief that teachers share responsibility for all the students at the school, not just those in their classrooms.
One high school principal we spoke with encouraged teachers to use a wide range of data to look at students’ experiences holistically. For example, recognizing the importance of student engagement, the principal provided teachers with the names of students who were not enrolled in extracurricular activities so that teachers could reach out to them and get them more involved in the school.
A shared sense of responsibility is also a key element in an elementary school that has a policy of not allowing any student to fall through the cracks. Teachers make it their business to know one another’s students and their families. When one teacher shares that he or she isn’t having success in meeting a particular student’s needs, others offer information that may be useful—for example, a seating arrangement or teaching strategy that worked well with the student in their classroom or information they’ve learned about the student’s family that may enable other teachers to better support that student.
The concept of data-informed leadership as a shared enterprise is essential in data-use efforts. Leadership for data use is not an inherent property of someone in a formal role; rather, it’s distributed among a network of people with complementary expertise working together (Spillane, Halverson, & Diamond, 2004). This is precisely the role of a teacher team when it’s functioning productively. Teachers share responsibility and leadership for improving student achievement.
Component 2. Conversations about data include healthy disagreement.
A shared sense of responsibility for all students and common expectations of student achievement are important ingredients of productive data use among teachers. However, that doesn’t mean that teachers must always agree. A healthy level of disagreement is an essential component of quality conversations around data.
For example, teachers may disagree about whether a student’s poor performance on a recent district assessment accurately reflects what the student knows. Particularly when these assessments have consequences for the student’s course grade or class placement, it’s essential that teachers speak up and provide alternative opinions. Leaders must actively create a climate in which it’s possible to raise divergent ideas, especially when the ideas are in the service of improving teaching and learning.
In one district we studied, leaders established a firm expectation that decisions would be supported by data, but teachers had to decide how they would use the data in their teams. The principal acknowledged that staff members had “lively discussions,” and she recalled an incident in which teachers in one department argued about how often they should assess students. She said that “we fought, we fought, we fought,” but the team eventually resolved the issue through multiple discussions about the relevancy of data and the need for frequent formative assessment. They agreed that interim assessments would be useful benchmarks to monitor student learning and growth.
School leaders can play an important role in helping teacher teams develop the norms of effective collaboration, including healthy disagreement. For example, one school administrator promoted the following slogan with her teachers: “Whatever happens in your meeting stays in your meeting.” She believed that disagreements should occur: “If you don’t disagree, then there’s something wrong because you’ve got 10 different personalities” on the team.
Of course, disagreements need to be resolved in a respectful manner that’s focused on the goal of improved practice. The inability to manage conflict is one of the chief reasons teacher collaboration may not lead to instructional improvement (Horn & Little, 2010). When teacher teams can’t deal with conflict productively, they tend to avoid it, thereby reducing teachers’ opportunities for professional growth (Achinstein, 2002). Developing and agreeing to a set of norms is one way teams can address this issue. Common team norms include allowing a safe place for sharing all ideas, no matter how underdeveloped they may seem; listening with respect; anchoring key decisions in evidence on student learning; not making assumptions about other teachers’ beliefs and practices or about their students; and keeping the focus on what’s best for students.
Component 3. Conversations about data engender trust rather than suspicion.
Some teachers mistrust data use because they fear that the data may be used against them. It’s crucial to establish trust between leaders and teachers and within teacher teams—and that requires having respect for all involved.
If you walk into a meeting with a feeling of respect for your colleagues, you’re more likely to be nonjudgmental when you examine the data they share. You’re more likely to assume that the patterns in the teachers’ behavior are rooted in their own circumstances and knowledge, rather than stemming from ill intent.
Teachers who have high expectations for students and continually embrace opportunities for professional growth have a difficult time respecting teachers who appear to be doing the bare minimum or who have deficit views of their students. But what if they asked one another what might be at the root of their actions and ideas? Many teachers experience fear and lack confidence in their ability to change. If team members understand the challenges their colleagues are facing, they can better understand those colleagues’ decisions and respond more helpfully—for example, by redirecting the conversation in a more positive manner, focusing on the assets that students bring to the classroom rather than the deficits.
One principal we interviewed said that teachers were initially defensive about their students’ test results. She explained to the teachers that data use was about finding out how the school needed to improve. Developing trust also involved continually reminding teachers that data would be used to ask questions and chart plans for improvement, rather than as a “gotcha, you’re doing a poor job” tactic, as one superintendent noted. This superintendent believed that creating a sense of trust was a “top-down, bottom-up, and side-by-side process.” In other words, for trust to develop within teacher teams, there must be trust between the district and the school and between the principal and the teachers.
That said, developing and maintaining trust within teacher teams can be an ongoing struggle, particularly as teachers come and go from teams. One teacher explained that in a meeting she attended, a colleague had removed some benchmark assessment test questions from her analysis because she hadn’t taught the content yet, which resulted in a score of 80 percent proficiency for her class. Other teachers who had retained the problems achieved far lower scores. The teacher stressed that honesty was important in data discussions and wished that her colleague had been more transparent. This problem can be addressed at the outset by establishing some of the norms of behavior discussed here.
Component 4. Data teams take a solution-oriented approach.
Being solution-oriented doesn’t mean adopting quick fixes or jumping to conclusions when analyzing data. Educators with a solution-oriented approach act when the evidence is clear and compelling, rather than selectively choosing data points to make a case.
For data use to truly improve teaching and learning, teachers need to engage in reflective practice. They ask why a particular approach or practice has worked well and identify lessons to be drawn before proposing it as possible solution. As one teacher explained,
This isn’t always easy. If I’m getting bad data back, I’m thinking, “What am I doing wrong, and what do I need to fix?” Some people think, “Oh, I’m a bad teacher.” They do that emotional shutdown. With our team, we say, “Oh man, my data sucks. … What are you doing that’s right, and let’s do that.”
Focusing on the future and on being positive are crucial features of a nonjudgmental, solution-oriented culture. One principal candidly explained,
We do have very typical challenges, struggles, and problems, but we don’t vociferate over it. We just say, “OK, this is how it is. What are we going to do?” We’re very solution focused, very forward thinking; there’s not a lot of griping at all.
As part of a continual improvement process, productive teacher teams use data discussions as a launching pad to discuss student thinking and understanding. For example, as one grade-level team examined students’ performance on a writing test, the teachers discussed whether students needed help with reading comprehension or more specifically with vocabulary development. Should they teach students prefixes and root words? Should they reconsider how they taught reading comprehension strategies in general? Several teachers commented that written tests didn’t capture the rich storytelling abilities of their students, many of whom were English language learners. They decided to experiment with an oral retelling section as part of their assessment process to better home in on student strengths and needs.
When teams allow room for experimentation and for investigating possible solutions, it is more likely that data discussions will directly connect to both student and teacher learning.
Component 5. Data teams know what they’re expected to accomplish.
Leaders play a significant role in setting the expectations for teachers’ discussions about data. However, it’s important that these expectations are balanced. Too much structure could stifle teacher discussion; too little could allow discussions to get off track.
To provide such structure, some districts and schools have developed or adopted data discussion protocols. Many teachers we interviewed reported that these protocols guided productive work. For example, one teacher explained that during discussions about a recent benchmark assessment, the teachers asked the following protocol questions: What did the majority of students do well on? Do you have an explanation for that? What strategies did you use? The teachers also examined the assessment itself, asking such questions as, How was this question worded? and Do these students know what the words in the question mean? To understand varying student achievement levels, teachers also reflected on how deeply particular standards had been taught. Then, they made plans to improve instruction.
However, in other cases, protocols were not optimally used. In one group we observed, the point of the meeting appeared to be making sure that team members discussed all parts of the protocol. With the focus on form completion, teachers spent less time engaging in a discussion of future instructional activities. In another data team’s discussion, teachers quickly completed the protocol using one-word answers and then spent most of their time in a discussion of potential instructional strategies that did not refer back to evidence collected in the protocol discussion. In neither case did the tool meet its intended goals.
Communicating broad expectations of what will be accomplished in data discussions is important, but teachers should not feel hamstrung. They should not fear that if the principal stops by to observe their team meeting, there will be consequences for not being on task. Leaders would be wise to share the mantra that the goal of such meetings is instructional improvement—and that data can be a crucial part of the discussion.
In one school we visited, teachers were encouraged to use a wide variety of data, especially observational data. Within their professional learning communities, teacher teams used their observations of students inside and outside the classroom, in addition to assessment data, to develop instructional goals, create formative assessments, and determine the next course of action. For example, a teacher noted that a student in her classroom asked insightful questions during literature discussions but struggled with fluency when confronted with written assessments. This led the team to develop a more fine-grained plan to improve the student’s reading achievement—one that did not rely on assessment scores alone.
Leaders must also recognize that some teams might need more scaffolds than others. For a team that struggles with communication styles, developing new instructional strategies, or learning how to use data, having a facilitator can help.
At one school we observed, an instructional coach helped connect the data analysis to instructional strategies and nudged teachers to move beyond swapping stories. When teachers shared the strategies they used, she would push them to reflect by asking, “Why did you do that, as opposed to using another strategy?” If a teacher talked about a strategy or a lesson that worked well, the coach questioned, “Why do you think it worked? Do you think it would work with all students?” When teachers came up with new strategies they wanted to experiment with, the coach offered instructional resources and support in the teacher’s classroom. Data discussions became the building blocks that helped build instructional capacity.
The Need for Data Leadership
These essential components for quality conversations about data are not rocket science, but they do require careful attention to norms. They speak to the importance of leadership in setting the right tone and expectations for data use among teachers. Leaders must create a culture of inquiry that allows for authentic teacher engagement in all stages of the data-use process.
Achinstein, B. (2002). Conflict amid community: The micropolitics of teacher collaboration. Teachers College Record, 104(3), 421–455.
Datnow, A., & Park, V. (2014). Data-driven leadership. San Francisco: Jossey Bass.
Hargreaves, A. (1994). Changing times, changing teachers. New York: Teachers College Press.
Horn, I. S., & Little, J. W. (2010). Attending to the problems of practice: Routines and resources for professional learning in teachers’ workplace interactions. American Educational Research Journal, 47(1), 181–217.
Spillane, J., Halverson, R., & Diamond, J. (2004). Towards a theory of leadership practice: A distributed perspective. Journal of Curriculum Studies, 36(1), 3–34.
Amanda Datnow is professor of education studies and associate dean in the Division of Social Sciences at University of California, San Diego. Vicki Park is assistant professor in the Department of Educational Leadership at San Jose State University, California. They are coauthors of Data-Driven Leadership (Jossey-Bass, 2014).
November 2015 | Volume 73 | Number 3
Doing Data Right Pages 10-15