California does not have to start from scratch in building new education data system

January 17, 2019
Three faculty members sitting at a table discussing student data.

Los Medanos Community College faculty members attend a workshop on student learning outcomes and assessment.

In his first budget, Governor Gavin Newsom has proposed spending $10 million for a new statewide longitudinal education data system. That prospect has brought hope that California will create a system that provides useful information to local, regional, state and other policymakers; tracks student data from kindergarten through college to the workforce; and enables evaluations of policies and practices.

Brad Phillips

To make this a reality, however, we must develop a useful and usable K-20 data system. We must avoid making the same mistakes that other states have made and avoid creating a system that languishes and cannot reach its potential.

In my experience founding the California Partnership for Achieving Student Success (Cal-PASS) two decades ago — a precursor to what is being proposed — and my work since then helping other states develop and improve their systems, I learned many lessons about what makes a system useful, easy to use and able to provide actionable data for policy and practice decision-making.

Cal-PASS was a voluntary collaboration in which California schools, colleges and universities shared student data and brought together teams of educators to mine that information for insights to reshape curricula and help more students enroll and succeed in college. All of California’s community colleges, most universities and two-thirds of the state’s school districts belonged to the Cal-PASS data-sharing consortium.

So, what have we learned?

First, a “build it and they will come” philosophy does not work. Jay Pfeiffer, who was the principal architect of Florida’s longitudinal student data system and a great mentor, emphasized that for the system to be successful it must support policy and practice at the local, regional and statewide level by answering questions that educators ask: What have students learned before coming here and how do they perform when they leave? Developers must consult deeply and early with those who would use the system to understand their information needs.

Second, developers of a state data system must realize that what is useful at a state level is different from what is useful for teachers and local administrators, but ensuring relevance and usability of the data for all users is essential.

Third, discussions typically focus on technical matters — developing and implementing the data system — and political issues such as governance, as if merely having the data and knowing who controls it will inform policy and solve student success problems. At Cal-PASS, we were driven to create a system that supported educators in answering specific questions designed to improve student success and their transition across the segments — from K-12 to college or university. The data was merely the tool for the courageous conversations that needed to take place at the local, regional and state level.

Fourth, local institutions, those that enter the student data into the system, need to see and experience a return on investment. In other words, the data provided by schools and districts and incorporated into a statewide system needs to become useful in their local work of educating students. Without this, data quality will suffer. The problem that we have noted among state data systems nationwide is that the research questions that support informing state policy do not necessarily connect with local institutional needs. This disconnect does not have to exist if the system is thoughtfully developed.

Finally, almost universally, statewide data systems generate reports that are difficult to interpret, resulting in educators trying to decipher what the data means and how to put it to use. In our work, we came to realize that this approach is based on the assumption that educators should have the skills of data analysts. They are not analysts and there are decades of failed professional development efforts designed to turn educators into analysts that prove it.

If a student data system is to be useful, it must provide the information needed by educators in a way that is easily understood. Local administrators, faculty and staff don’t have the time, resources or wherewithal to pore over innumerable rows and columns of data — like some Where’s Waldo game — to identify the problems they need to act on. The same goes for policymakers. The reporting system must begin with end-user consultation about what they need and how they need it presented.

California has an opportunity to create a useful and usable state longitudinal data system that supports action at multiple levels and in multiple ways. But let’s learn from other states and California’s early success with Cal-PASS to create a system that is responsive to local, regional and statewide practice, policy and research needs. Students, the institutions that provide their education, our communities and the state will all benefit. If this effort is to be successful, developers must start with the end in clear focus.

•••

Brad C. Phillips is President and Chief Executive Officer of the Institute for Evidence-Based Change (IEBC) and founder of the California Partnership for Student Success (Cal-PASS).

The opinions expressed in this commentary represent those of the author. EdSource welcomes commentaries representing diverse points of view. If you would like to submit a commentary, please review our guidelines and contact us.

To get more reports like this one, click here to sign up for EdSource’s no-cost daily email on latest developments in education.

Share Article