Moving Beyond the Single Data Point
“In God we trust; all others bring data,” said the management guru W. Edwards Deming. At the Data Quality Campaign, we champion the power of data to improve student achievement. However, if there is not trust in the quality of the data—and how they will be used—this information will not change conversations, decisions, and actions in order to help students succeed.
During these summer months, principals and superintendents are sitting down to make personnel decisions, some of which will be shaped for the first time by value-added teacher scores. One of the hottest questions in education is whether to name individual teachers publicly with their value-added performance data. Publicly disclosing these numbers alone—which are neither in context nor useful—gives parents incomplete information, puts misguided pressure on administrators, and sabotages vital trust with teachers.
A favorite mantra at the DQC is “data as a flashlight, not solely as a hammer.” The predominant culture around data use in education has been too focused on compliance and punishment. That needs to change. Data are only as good as they are useful—and that includes teacher-performance data. Accountability and transparency are important uses of these data. But it is also crucial to get these data into the hands of teachers themselves as soon as the information is available. Too often, they are denied that opportunity.
In 2010, the Los Angeles Times first published individual teachers’ names alongside their performance rankings. The numbers came as a shock to many Los Angeles teachers, who first learned of their own performance data on the front page of the newspaper. The paper reported in August 2010:
“...[One teacher] said the numbers were important and, like several other teachers interviewed, wondered why she hadn’t been shown such data before by anyone in the district. ‘For better or worse,’ she said, ‘testing and teacher effectiveness are going to be linked. ... If my student test scores show I’m an ineffective teacher, I’d like to know what contributes to it. What do I need to do to bring my average up?’ ”
The system is broken when reporters can get more information about how teachers are doing than teachers can themselves. This delayed and denied access to their own data undermines teachers’ trust in the data and efforts to use the information in the classroom to make the best possible instructional decisions.
Families need data to make decisions about their kids. Helping states empower parents with data is a major aim of the DQC. Taxpayers, principals, and school administrators also deserve information about the effectiveness of teachers, but publicly releasing individual teachers’ names and student-performance data in the name of transparency and accountability will not get us the results we desire. We have the technology to transform data into actionable information. What we need to do now is to tailor this information to meet the needs of the stakeholders based on the questions they are trying to answer. It all comes down to getting the right data to the right people at the right time.
Value-added scores are important, but they are only one slice of the apple. Single measures of student growth do not paint the full picture of how a teacher is doing, and they do not empower parents and other stakeholders to make the best decisions. It is not enough to pay lip service to this fact before releasing isolated data, as if parents and others could easily find the rest of the information they would need to form a full picture of a teacher’s effectiveness. Responsible policymakers and practitioners provide stakeholders with a full picture without sending them on a scavenger hunt for context and meaning.
The good news is that many states are developing better performance-evaluation and -development systems that take into account observations, student surveys, team teaching, and multiple measures of student growth. Equally important, they are working to get these contextual, useful data into the hands of teachers, principals, and administrators who can use the information to drive change and improvement. Louisiana and Tennessee, for example, have enacted legislation that specifies stakeholder access to teacher-evaluation data and prohibits public disclosure of teacher names. And recently, the New York state legislature passed Gov. Andrew Cuomo’s bill that prevents public naming of teachers, while making teacher-evaluation data available to parents who request it.
Other states can build on this model by publishing aggregate, school-level teacher-performance information based on comprehensive teacher evaluations (including, but not limited to, value-added scores). Further, states and districts can provide tools and guidance to help parents and teachers understand the data and what they mean for their children and schools. And teacher preparation should include data-literacy training. According to "Data for Action 2011,” the DQC’s state analysis, which studies states’ progress toward building and using longitudinal-data systems, only 10 states have policies that require data-literacy training both for state approval of teacher-preparation programs and for teacher and principal certification. We must do more to train all stakeholders—particularly teachers—and encourage them to use data effectively.
By ensuring appropriate access to meaningful data, states can achieve transparency, establish trust, and equip decisionmakers with useful information.
What’s in a name? A lot more than a single data point.
Vol. 31, Issue 37, Pages 28-29