The Bar Raises (Or Why Good-to-Great Teachers May Also Not Like Performance-Based Evaluations)
Earlier this week, the inestimable Michael Goldstein, founder of the MATCH collection of charter schools, asked an important question on his Starting an Ed School site related to Good to Great, management guru Jim Collin’s famed profile of top-performing companies. Noting that most of the 11 firms identified by Collins have slid into mediocrity, have been bought out, or, as in the case of the now-infamous Fannie Mae, went bust, Goldstein speculated on several reasons why the firms have fallen downward, including perhaps gaining too much public acclaim for their success.
Goldstein fails to consider legendary economics editorialist Henry Hazlitt‘s famed declaration that economic changes — especially after the emergence of new technologies and sectors — tend to turn today’s high-skilled workers and top-performing companies into middling players. One of Collins’ top performers, the now-defunct Circuit City, could not overcome the advent of the Internet and firms that have emerged in it such as Amazon (which made it easier for consumers to bargain shop for music and gadgets), the growth of retail giant Wal-Mart (which is now one of the two dominant players in nearly every category of retailing), and the success of rival Best Buy (now being buffeted by the same forces that brought down Circuit City) in addressing the emergence of consumer electronics as a lifestyle. Another good-to-great firm, Pitney Bowes, is mired in the business of mail service, which is also being adversely affected by the Internet (in the form of e-mail) and by Wall Street’s emphasis on fast-growing tech firms over Pitney and low-growth old-school industrial outfits.
At the same time, there’s another reason why good-to-great companies end up sliding into mediocrity is the consequence of raising the bar on performance. This is one that matters for school reformers, especially those working on overhauling how we evaluate and compensate teachers, and the good-to-great teachers that we want to reward for their work. This is the consequence of raising the minimum bar on performance. And this reality is one reason why even some top-performing teachers are resistant to moving away from traditional subjective classroom observations to performance-based systems using objective data on student performance growth on standardized tests.
Right now, many of the teachers who are good-to-great have gotten that way haphazardly. After all, ed schools, which account for nearly all new teachers entering classrooms, don’t recruit aspiring teachers for subject-matter competency, empathy for all kids regardless of background, or entrepreneurial self-starter ability — and they do an even worse job training them. More importantly, once a teacher enters a classroom, the desultory processes for evaluating teacher performance (including shoddy school leadership on this front), often means that no one is actually measuring how well teachers are doing.
As a result, those who are doing good and great in improving student achievement are as much where they are because the bar for teacher quality within districts is just so low as because of their well-honed talent, skill, and results. And so these teachers set the minimum bar for good-to-great work, especially in reform-minded districts where efforts to overhaul evaluations and teacher performance management are just in its infancy.
One can easily see this in the success of IMPACT, the performance evaluation system launched by D.C. Public Schools during Michelle Rhee’s tenure. While the district has had some increases in teacher turnover, most of the teachers leaving are those DCPS has either dismissed through the evaluation process or have been identified as being low-performers; just 9 percent of DCPS’s top-performing teachers in the 2010-2011 school year have left the district by the following school year, while 22 percent of the least-effective teachers have left for other employers. Based on those numbers, one can easily see DCPS getting rid of nearly all of its worst-performing teachers from the 2010-2011 school year within five years, leaving most of its top-flight instructors and merely effective teachers in place.
But there is a price to pay for being the ones setting a higher bar on minimum levels of performance. That comes in the form of new hires being as good — and after four year, perhaps even better — than the good-to-great teachers in place. As the bar for performance is raised, those who had originally set the high bar are no longer the top performers. Those who were considered great five or 10 years ago will merely be considered just good. And those who are good? Well, they are mediocre, or worse, low performers. As with technological and economic changes, changes in teacher quality will end up turning those who are top-performers into average players — and this is will likely happen because, on average (and unless they are continuously improving their work) a teacher is no better after 25 years on the job than they were after four.
This reality, which is as applicable in other aspects of working life (and sports) as it is in education, is absolutely scary for anyone who is at the top of the teaching game today. Being the best today — or just being really good — isn’t assured, especially in a climate of higher expectations. And thanks to Value-Added Analysis of teacher performance through the use of standardized test score growth data, along with other methods such as student surveys, a profession in which one could think of themselves of being the best (even if information proved otherwise) and keep their jobs regardless of performance (thanks to near-lifetime employment policies in the form of tenure) is now like every other profession. Or to paraphrase the old Wall Street statement: Past performance doesn’t guarantee future results or long-term employment. And if you are a good-to-great teacher or today (or those who have been named Teachers of the Year in their states, especially in those where such titles are often popularity contests instead of being based on actual metrics), moving to performance-based evaluations isn’t exactly comforting, even if you know this will have great benefits for both the profession and, more importantly, our children.
This unexpressed fear (and, to be honest, a reasonable one) is one reason why even good-to-great teachers would rather go with the multiple measures approach to evaluations (which includes the use of inferior classroom observations) than simply use student data as the overwhelming factor in measuring performance. As the Bill & Melinda Gates Foundation’s own Measures of Effective Teaching initiative has shown,the multiple measures approach is less-effective in evaluating performance than using Value-Added data alone or using it along with student surveys. But if you are a good-to-great teacher worried about keeping your job, weakening the system used to measure your work helps you keep your job for a while longer at the point when you are no longer good-to-great. In some ways, it is also likely the reason why some reformers, especially those who consider themselves bold, are also willing to weaken evaluations (and also oppose publicly releasing teacher performance results).
School reformers must consider how to address these unexpressed concerns as they push harder to overhaul teacher evaluations and how we recruit, train, manage, and compensate teachers. Because these concerns do play into the hands of the education traditionalists, especially the National Education Association and the American Federation of Teachers, who already have these teachers in the rank-and-file. This means playing to the better natures of our good and great teachers, and perhaps, even coming up with new career paths for these instructors once they are no longer in the ranks of the top-performing.