Perhaps the biggest flaw in the initial version of ProComp was not the plan itself, but the chasm between how ProComp was described and what it actually contained. During the ballot initiative process numerous groups — including ProComp’s designers and proponents — labeled it as “Pay for Performance,” a term quickly adopted by media reports and ultimately assumed by voters.
After three years, ProComp came up for review, and in the ensuing spat between DPS and the DCTA, the disparity between promise and process became evident. As a May 2008 editorial from the Rocky put it:
“But we also are very familiar with how ProComp was sold to the public in 2005 during a tax-hike campaign to pay for it. Indeed, we took part in the salesmanship. There is no mystery at all about what voters were told: that ProComp was a pay-for-performance agreement…”
Describing ProComp as Pay for Performance, as I argued some months ago is clearly inaccurate, if not misleading. However, as the program evolves, this electoral bait-and-switch is increasingly less relevant. ProComp’s promise has faded, but the process is here to stay.
To start, let’s finally call ProComp what it is: differentiated pay. Simply put, ProComp varies teacher compensation based on criteria other than years of service. In most (perhaps all) professional occupations, differentiated pay is standard. Doctors, lawyers, software engineers, even college professors with the same tenure are paid different amounts.However, in our public secondary schools, pay is determined almost entirely on seniority (see the DCTA salary schedule). In the professional world, mention differentiated pay and barely a ripple ensues; in education the debate over variable compensation is seismic.
Within the confines of introducing differentiated pay, ProComp 2.0 makes some substantial progress. Having largely inherited ProComp, the current DPS administration deserves credit for extracting some value out of a system with severe limitations. If you are given a wiffle bat and ball, no one should expect you to knock a home run out of Coors field.
ProComp’s foundation remains basically unchanged, structured in four groups: Knowledge and Skills; Professional Evaluation; Market Incentives; and Student Growth. The first two groups are predominantly concerned with credentials, and have scant correlation with student learning or any quality component. The second two groups introduce some basic concept of supply and demand to the teaching profession, and begin – gingerly and indirectly – to address academic performance.
The initial version of ProComp devoted about 60% of possible dollars to the first two groups and 40% to the last two. ProComp 2.0 flips this ratio, with the first two groups reduced to 32%, thus reserving over two-thirds for Market Incentives and Student Growth. This transfer is critical to move to a system where rewards follow needs, for it is these two groups that have the most potential to aid reform.
Moreover, ProComp 1.0 had 90% of possible dollars structured as an increase in base salary (which continued through future years of service regardless of future results), while ProComp 2.0 keeps about 70% limited to the year the criteria is met. Finally, the overall amounts have increased significantly, with potential ProComp 2.0 dollars equaling over $17,800 (on a base of $36,635), compared to the previous total of $9,901 (on a base of $32,971). The average teacher bonus is now estimated at about $6,000, which is enough to be meaningful.
Differentiated pay is important both in concept and as an alternative to the rigidity of the single-salary structure in an educational system that increasingly demands accountability and flexibility. The introduction of market Incentives is the simple acknowledgment that across a large school district, some teaching jobs are harder to staff (and often harder to do) than others. Teaching reading to affluent students in Elementary School is far different that teaching science to poor students in High School, and the single-salary structure denies the opportunity to pay teachers differently.
But ProComp’s details still contain more than their fair share of devils. Most disheartening is the Student Growth group, where academic achievement is rewarded merely for being above the DPS average. In a school district where overall proficiency lags the state by some 35%, this is akin to bragging about being of above-average height in a tribe of pygmies.
ProComp rewards teachers at schools rated above average on the DPS School Performance Framework (SPF). Designation as a “Top Performing School” means only that a school ranks above the DPS average on absolute CSAP scores. While I’ll keep the names anonymous, among the long list of schools that therefore qualify are 1) a K-8 with 37% of 8th graders proficient in math; 2) a Middle School with 28% of 8th graders proficient in math; and 3) a High School with 36% of 10th graders proficient in math. If these are classified as top performers, just imagine the bottom.
All three of these schools also manage to slip above the DPS average on the “High Growth School” metric (how much CSAP scores improve year over year), with scores of 59, 57 and 64 respectively (50 being average). And all three also qualify as “Hard to Serve,” despite having free and reduced lunch populations below the DPS average.
Therefore, despite providing an education that should make most people quiver, these three schools have hit the trifecta of High Performance, High Growth, and Hard to Serve. All their ProComp teachers will receive an annual bonus of over $7,000 on these three criteria alone.
ProComp’s flaw is not just that the pay is not truly linked to performance; it is that the performance requirements are far lower than what students need to be successful. It is now an accepted truism that urban education reform requires adherence to high standards. In rewarding continued mediocrity, ProComp is embedded with the very same flaw it is supposed to cure.
Lastly, there is little difference between schools (and teachers) that are slightly good versus those that are truly great. Nor is there any exception for a below-average teacher who happens to be in an above-average school. In essence, DPS is grading a dysfunctional school system both pass/fail and on a curve. The smallness of this progress can only be appreciated if one remembers that previously there was no grading at all.
ProComp 2.0, as with many upgrades, is judged first compared to the old version, and not the original marketing campaign. Again, in the shift from credentialing to market incentives and student growth, there is considerable improvement, and more potential. However the core issues remain: taxpayers are funding a $25M annual program that does not directly confront the seminal problems of academic underperformance and the achievement gap.
To ultimately judge ProComp as a success or failure, one needs agreement on the objectives for Denver’s students, and some way to measure if those goals are met. Most of the debate thus far is simply about where and how much money is spent. Merely emptying the coffers is not sufficient, however there is no articulation of what specific educational outputs ProComp is supposed to produce. While this is politically expedient, allowing both DPS and DCTA to claim that the plan fits their (often disparate) aims and is responsible for any positive trends, as we saw this past summer it makes it hard for ProComp to evolve.
I remain hopeful that the incremental shift from the single-salary schedule to some differentiated pay may yet open the door to more meaningful changes. We sometimes forget that initial innovations are often underwhelming. Public education reform is frequently described as “building the plane while flying it.” It takes a historical perspective to remember that the first airplane flight at Kitty Hawk took all of 12 seconds and went just 120 feet. Even with its improvements, ProComp is unlikely to have a meaningful impact. But it may yet spawn programs that will.