SCHOOL FOR THOUGHT

Moneyball and CompStat: Implications for Education Policy

I have been reading a very important book entitled Teachers Matter: Rethinking how public schools identify, reward, and retain great educators, by Marcus Winter. I propose to do a series of posts that excerpt from this book, showing how Dr. Winter develops his argument. Today’s excerpt (pp. 1-2) examines the potential of data-driven teacher evaluation and compensation. As you read, bear in mind that most current administration systems hire and compensate teachers on the basis of factors (like experience and qualifications) that are unrelated to teacher effectiveness.

• In his 2003 book Moneyball, Michael Lewis explained how in recent years the Oakland A’s managed to post one of the highest winning percentages of any team in Major League Baseball despite consistently having one of its smallest payrolls. The A’s trick, it turns out, was a data-driven assessment for identifying players who would contribute to a team’s success. Instead of relying on the traditional subjective analysis of baseball lifers and the usual array of official statistics, the team’s general manager, Billy Beane, hired a group of numbers geeks to develop statistical models that could better predict a player’s ability to produce runs in the future. Bean’s insight, which was already percolating among baseball outsiders, was that a team with better information about player quality has an advantage on the field. Data analysis didn’t ensure you’d win the World Series, but, all else being equal, it gave you a better opportunity to do so.
• In hindsight, it’s surprising that the smart people working in baseball before Beane came along thought that scouting reports correctly identified the best players. Scouts’ eveluations were often based on little more than viewing a few of a player’s at bats. Nonetheless, for little more reason than clinging to tradition, clubs simply did not bother to improve upon these clearly incomplete measures of a player’s worth for many decades. Failure to distinguish correctly beween the most and least effective players cost cost teams millions of dollars each year as they overpaid for mediocre talent.
• As in baseball, many features of the current system for employing public school teachers seem reasonable on the surface. For example, it seems self-evident that experienced teachers and teachers who have earned advanced degrees would be more effective than their junior colleagues.
• Sports franchises have recognized that they can use data to identify inefficiencies that will save them money and increase their performance. Policy makers have used advances in the development and use of data to save lives.
• In the early 1990s, New York City was a pretty difficult place to live. Crime was out of control, and it appeared that there was little that could be done about it.
• Then came CompStat. Adopted by Mayor Rudolph Giuliana under the guidance of his police chief, William Bratton, this revolutionary data system tracked crime precisely. The influx of data allowed the NYPD to focus its efforts on the most troubled neighborhoods, and to hold precinct captains accountable when things went wrong.
• Looking back, you have to wonder why it took so long for the police to adopt CompStat. Identifying the areas where crime is persistent would seem essential to anticipating it. Once technology was capable of maintaining and analyzing information about street crime, the obvious next step was to put such data to good use. Public schools are only now beginning to learn the fundamental lessons of Moneyball and CompStat.