69传媒

When I became governor of Pennsylvania in 2003, one of my first acts was to create an Office of Health Care Reform to come up with a plan to provide quality, affordable health care to all of our residents. Searching for successes to build on, my staff quickly discovered an extraordinary set of hospitals in southwestern Pennsylvania that had virtually eliminated the most dangerous type of infection that patients contract while in the hospital for unrelated illnesses.

These hospitals were saving lives while cutting wasteful spending so that they could put more money into patient care. And their strategy was common sense: They collected data on every patient; tracked the patient鈥檚 progress and any infections he or she developed; and used the data to pinpoint the exact sources of problems. Once a problem was identified, they fixed their own processes and systems so that it never happened again. Then they tracked the results to make sure the steps worked.

In 2007, every member of the Pennsylvania legislature voted to require all hospitals and nursing homes to replicate this successful strategy. The law will ultimately save well over 3,000 Pennsylvania lives each year, and improve the quality of life for tens of thousands more.

It is far past time that we put the same principles to work by using data to ensure our shared quality of life and economic prosperity in an education system that prepares every student for college or the high-skills workforce of the competitive global economy.

In Pennsylvania, we have spent the past six years building our academic infrastructure鈥攁nd the results are starting to show. Step by step, the commonwealth has put in place the fundamental components of quality education: prekindergarten programs; adequate school funding that focuses on proven programs and extra support for struggling students; an overhaul of teacher preparation, continuing training, and certification requirements; and transforming our high schools.

Yet as recently as three years ago, the only honest way to describe our data systems would have been 鈥渄ismal.鈥� Data-driven state-level policymaking has been a core value since day one, but it has taken far more time to build the systems that support local data collection and analysis in education.

Like virtually every state, Pennsylvania now has an information-management system that follows every individual student who attends our public schools. Our state-of-the-art data warehouse collects information on student demographics and attendance, staff characteristics and assignments, courses and enrollment.

But state leaders are not interested in compiling data for data鈥檚 sake. Our goal is to put data systems to work so that by the time our students graduate from high school, we can award them a diploma with confidence that they are well prepared for the real world that awaits them.

A sound approach requires longitudinal data systems that follow individual students鈥� progress over time. Access to data on student performance allows policymakers to know the state鈥檚 real graduation rate and the percentage of high school graduates who must take remedial courses in college; superintendents to understand which schools produce the strongest academic growth for their students; principals to know which grades, programs, and teachers are having the greatest success; teachers to know where their incoming students need help and whether their outgoing students are succeeding; and parents to know exactly how their children are progressing.

To achieve these goals, state longitudinal-data systems geared to making the high school diploma count should have the following five characteristics:

1. Start with early-childhood education, collecting demographic and assessment data on children in all early-childhood-education programs鈥攏ot just those provided by school districts鈥攁nd linking the data to the students鈥� K-12 records. Pennsylvania is at the forefront in this regard.

2. Measure each student鈥檚 annual academic growth, providing teachers and parents with a record of whether every child is making progress, in addition to whether he or she is currently on grade level.

3. Provide early-warning signs no later than middle school to alert parents, teachers, and principals that a student needs help鈥攇uiding the creation of an individual achievement plan based on the student鈥檚 weaknesses.

4. Include diagnostic tools for core high school classes to provide regular feedback on the skills that students are mastering and where they are struggling, so interventions can be successfully deployed and educators can make midcourse corrections.

5. Link pre-K-12 records to college matriculation, retention, and graduation data, as well as to workforce information, to track student success after graduation.

But none of this matters鈥攏one of it鈥攗nless we do a better job translating data into a format that is truly useful to classroom teachers, and unless we hold our principals and superintendents accountable for creating a school culture that revolves around working with data, talking about data, and using data to change teaching practice and deploy financial resources.

In Pennsylvania, we have taken steps in this direction鈥攔equiring all principals to receive training that includes a heavy dose of how to use data, and vastly improving our district- and building-level data tools to make them more user-friendly.

But like many other states, we are still struggling with how to create the teacher-level 鈥渄ata dashboard,鈥� and the accompanying training, that teachers see as a real tool to help them be successful.

In addition, we have a state-level responsibility to use these same data to ensure that students have access to the opportunity to learn. Course-taking patterns that show low-income students being shepherded away from college-prep classes; achievement data that suggest the newest teachers are being deployed to the toughest schools, with pernicious outcomes; and program-evaluation data that indicate we are not getting a return on certain taxpayer investments must all be seen as cause for alarm and, most importantly, action.

Just a few months ago, Pennsylvania鈥檚 independent health-care watchdog announced that the number of patients who died from hospital-acquired infections dropped by 300 from 2006 to 2007, when our new quality-control law was in effect for six months.

When used to change the way we do business, data systems work.

Thankfully, President Barack Obama and U.S. Secretary of Education Arne Duncan included $250 million in the American Recovery and Reinvestment Act to support state efforts to implement longitudinal-data systems, and they have wisely linked state commitment on this issue to the deployment of fiscal-stabilization and 鈥淩ace to the Top鈥� investments from the federal economic-stimulus money.

Now, it鈥檚 up to policymakers and educators to take advantage of this once-in-a-lifetime opportunity to do the right thing for our students and our economic future.