Performance Pay in Denver
August 8, 2008
Performance Pay in Denver
- Paul Teske is Dean of the School of Public Affairs at the University of Colorado Denver, and Director of SPA’s Center on Reinventing Public Education Denver. His co-authored education books include “Pay for Performance Teacher Compensation: An Inside View of Denver’s Procomp Plan” (Harvard Education Press, 2007).
- Kim Ursetta is a 4th and 5th grade bilingual teacher at Newlon Elementary in the Denver Public Schools. She is currently on leave serving in her second term as president of the Denver Classroom Teachers Association. She has been teaching in Denver for 14 years, and is a National Board Certified Teacher.
Vaishali Honawar (Moderator):
Good afternoon, and welcome to Education Week‘s live chat. Our topic today is the recent controversy surrounding Denver’s performance-pay system, ProComp. ProComp was created in 2004 under an agreement between the Denver school district and the local teachers’ union, but the two sides have butted heads in the past few months over new changes proposed by Superintendent Michael Bennet. Joining us today are Kim Ursetta, the president of the Denver Classroom Teachers Association, and Paul Teske, the dean of the school of public affairs at the University of Colorado, Denver, who has co-authored a book on ProComp and is also conducting an external evaluation of ProComp expected next year. I am Vaishali Honawar, a staff writer at Education Week and I will be moderating this discussion. We already have many questions, so let’s get started.
Question from D.R. Harbert, Teacher, Denver Public Schools:
I just switched districts to come to Denver, and am excited about the ProComp program, having come from another district who claims to have performance pay, but isn’t based on performance at all. Before switching districts I had to do a lot of research about the controversy . . . and I have to say that a compromise between the two parties would seem to address both sides’ issues, while continuing to study and evaluate the fairly new program. Yes, new teachers need incentives to stay, but mid-career teachers can’t be neglected for already have chosen to be here. Is there any talk about coming to the middle? P.S. There is NO WAY I will go on strike, should the command be made. Not in today’s economy, and as a single woman living on a teacher’s salary.
I appreciate you commitment to come work in Denver -- a great school system with a commitment to ongoing improvement. I hope you will be a part of DCTA and help us to articulate what teachers want: * More time to teach and to plan and to analyze what we are doing. * An integrated system of induction, mentoring, evaluation, and professional development. * A compensation system that is balanced and fair, encourages things that make a difference in the classroom, foster a spirit of collaboration, and addresses the need to attract and retain qualified teachers. We are doing everything we can to help our members and the public understand what is at stake in this negotiation, and we believe we can reach a settlement with the help of a mediator.
Question from Joe Sulock, second grade teacher, NJ:
The research that I have read suggests that principal evaluations are not indicators of student success. Considering the lack of connection between teacher effectiveness and administrator evaluations, why is an administrative evaluation a part of the formula for the compensation teachers in Denver receive?
ProComp attempts to use multiple measures to reward teachers, in part since no one measure can be shown to accurately reflect everything about quality teaching (indeed, they all have flaws, including the student achievement measures). Many, perhaps, most organizations reward employees based upon significant input from their supervisors. So, using principals’ evaluations is perceived to be one of several measures to try to reward teachers - principals clearly should know which teachers are doing a good job, from their perspective, even if it doesn’t always show up directly in this year’s student test score data.
Probably underlying your concern, of course, is historical evidence that most principals give most teachers very high ratings, when not all can really be “excellent.” This is an issue that these plans must address carefully, in implementation.
Question from Michael Helms, Math Teacher, Camden County High School:
Why not pay everyone involved in the educational system from the federal, state, and local levels based on the performance of the students? Why only the teachers?
Teachers are “closer” to the students, and we do know from research that very good teachers can improve student performance more than any other factor that the education system can control. So, that is one reason to focus first or mainly on teachers.
Denver has developed a parallel plan to ProComp for its principals, and certainly some other districts award principals bonuses for enhanced student performance. Some other districts also have “group” plans, where everyone in a school or district is rewarding if student achievement goes up. Of course, some have even proposed, or experimented with paying the students for performance, or perhaps their parents.
I think the links get a lot weaker, if we think about paying state or federal education officials rewards for student performance, and they are therefore less likely to be effective.
Question from Don W. Long, President, Education and Democracy for Learners and Leaders:
Given the importance of maintaining the original collaborative spirit between the district and the union and the importance this national model has for teachers and students across the country, I wonder if both sides could agree to disagree for now. I believe the union is rightly asking that significant changes should not be made until informed by the only rigorous evaluation, which is expected next year. In the interim, the district could explore the spectrum of non-monetary performance incentives shown by research to be very effective, such as recognition, job enlargement and enrichment, teacher leadership, and opportunities to collaborate and to learn.
One of the guiding principles behind the development of ProComp was the use of data to drive decision making. This isn’t about change or no change, it’s about whether future changes have data to support them. There is no research to prove that the changes the administration proposes will improve student achievement.
Question from Ken Babcock, Supervisor of Employee Relations, Lincoln Public Schools:
Of the many components of the performance pay plan, is there one component that appears to be most promising in increasing student achievement.
Unfortunately, I don’t think we have enough research or data, from ProComp or any other program or experiment to know the answer to that. Obviously, most people who think about this believe that the direct measure of student test scores improvement will be the one that works best, since it is direct and quantifiable. But, there are technical and data collection problems with even that measure.
The ProComp pilot did show that when teachers jointly set student growth goals with their principal, the “quality” of those goals was positively correlated with student achievement gains, so that is promising.
Question from David Liebowitz, Doctoral Student, Harvard Graduate School of Education:
You’ve stated that you hope to postpone renegotiating ProComp until after the completion of the external evaluation. How do you balance the desire to implement policy influenced by research and the political and educational urgency of capitalizing on seemingly successful reforms? Kim Ursetta:
Things that have worked about ProComp, including balance,fairness, shared goal-setting, professional development -- would be undermined by diminishing incentives for career teachers to participate in the programs. We ought to try to build momentum for future reforms, but authentic reforms -- such as integrating induction, mentoring, evaluation, and professional development, as we have proposed.
We run the risk -- in the ready, fire, aim approach to reform -- to make things worse and diminish enthusiasm for future reforms among teachers and the public.
Question from Mark Alberstein, President, Educate For A Change, Inc.:
Isn’t the crux of the matter simply: “How does one define the term performance?” For years, teacher performance has been evaluated by a broad list of criteria. Now, because of politics, that’s all been rendered meaningless; brushed aside by test scores. Didn’t this seem absurd when the proposal for performance pay was first introduced?
I think most observors would agree that test scores are not the only thing that matters, although recent accountability programs sometimes make it seem like that.
ProComp uses a big list of different measures to reward good teachers, not just test scores. It does attempt to make the “subjective” evaluations more concrete and more directly tied to outcomes that the district values, rather than more mushy past approaches to evaluating teachers that have generally led to outcomes where most teachers get “very positive” evaluations from their supervisors, because they didn’t really matter. Question from Liz Barkowski, Doctoral Student in Education Policy, University of Texas at Austin:
We know through research that teacher quality is the primary determinant of student achievement. By increasing the number of truly “highly qualified” teachers, student achievement will reciprocally increase.
If the goal of ProComp is to attract and retain high quality teachers and in essence improve student achievement, then why is the union opposed to increasing recruitment/retention stipends for teachers in high-need schools and subject areas, especially when preliminary results are positive?
Why throw more money into mentoring and induction instead when we are losing 30-40% of teachers in their first 3 years anyway?
No one who has been a teacher -- or looked at the research -- would say that mentoring and induction are “throwing away money.” Studies show that money is only one factor in high turnover in the early years. Teachers who have an opportunity to work with veteran teachers on an ongoing basis are better able to make the adjustment to an intellectually, physically, and emotionally challenging job.
It’s a mistake to say it’s one or the other. We have to address all elements of teacher quality, including preparation, evaluation, professional development, and compensation if we truly want to make a difference in student achievement.
Question from karen green researcher, former teacher:
do you honestly think that a salary of 37,000 to $44,000 is enough of a draw in denver to attract quality teachers to areas of science,math or special ed? the union does not agree to it nor do the districts. why not raise the salary to the point of middle income as other professionals earn and base as you say the salaries on education and years of service ? if teachers are indeed doing their job they should be compensated. if not don’t you think they should be let go?
One goal of ProComp is to provide greater incentives for teachers in these areas of shortage. It may be that the actual incentive amount is not enough, and that is one issue that the proposed changes to ProComp are trying to address.
The idea of paying teachers differently based upon what they teach, where they teach, how well they teach, and not just based upon their inputs is central to these changes. ProComp is starting to do that, but it is just a start.
Question from Susan Norwood, Advanced Studies Specialist, Charlotte-Mecklenburg Schools:
As our system moves into year two of a pay-for-performance initiative, one of our goals is to shift the focus from compensation to instructional reform. What steps has Denver taken to ensure that teacher behaviors are changing to improve student achievement?
ProComp deals with nine areas -- including three different factors focused on student growth, and five other elements related to continuing improvement, including evaluation and professional development.
One of our goals is to establish a Peer Assistance and Review program that would coordinate induction and mentoring, evaluation, professional development, and other issues. We think this could be well integrated with ProComp and emphasize the system is about improvement, not just salaries.
Question from Jennifer Portillo, Teacher, Denver Center for International Studies:
Wasn’t the Pro-Comp trust set up so that as more teachers moved into it, that there would be sufficient funds to pay teachers’ in the future? If so, why does the DPS administration and people studying education issues keep insisting that too much money in the trust is not being used at this moment in time?
There has been some misunderstanding in the community about how quickly anyone expected the trust fund money to be spent - partly it depends upon projections (and now some actuals) about how many teachers would opt-into ProComp and how many elements they would earn each year. I wonder how those who have commented that DPS has spent too little of the ProComp money would feel if “too much” had been spent - probably not happy with that either.
Having said that, I think both the district and the union have an incentive to spend some more of the money on paying good teachers for good work, sooner than later. There certainly should be urgency on all parts to get better teaching for all of Denver’s kids, sooner than later.
Question from Ken Mellendorf, Professor, Illinois Central College:
Why should teachers desire to improve performance if a large pay raise across the board results from nothing more than participation?
A few years ago, Gallup did a poll for Phi Delta Kappan that showed that more than 80 percent of Americans agreed with the idea of merit pay for teachers. And the majority believed that the merit should be based on knowledge and experience. That’s what the single salary schedule does -- it rewards knowledge and experience.
But at the same time, we support maintaining the nine elements of ProComp that people can get incentives for. If you believe there should be more to it than a single salary schedule, then you should support our proposal to sustain and strengthen ProComp -- rather than convert it to a bonus system with quotas. Question from Jessica Lyons, teacher, Denver Public School:
The district’s proposal creates equity among the salary increases, so that newer teachers are getting raises that are about the same size as teachers with the most experience. Why is the union opposed to equalizing raises?
In our view, the district’s proposal make the system less fair and equitable. We have agreed with the idea of raising salaries for new teachers -- even to the extent their raises would be higher than some experienced teachers. The real inequity comes at the other end. The administration’s proposal would cap salaries at the 13th year, reducing life time earnings and retirement benefits for all teachers. It would create a bonus that would be solely at the discretion of a principal.
If you are weighing the two sides on fairness and equity, ours is better.
Question from Liz Barkowski, Doctoral Student in Education Policy, University of Texas at Austin:
What are the major proposed changes to ProComp’s plan and why does the union oppose those changes?
The district’s proposal would convert ProComp from a system of incentives to a system of quotas.
We are concerned that a teacher’s salary could go up or down by $9,000 based on things that aren’t even proven to improve student performance.
We are concerned that the district is working to implement the Tough Times, Tough Choices plan -- whether it fits Denver’s needs or not -- to cap salaries and reduce retirement.
We are concerned that district’s plan would violate four of the nine principles of ProComp, and that if they continue to press for such a radical change in the program, it may be a generation before another union will be willing to take such a risk.
Not every new idea is reform. Not every “reform” makes things better.
Question from Stacy Holland, Teacher/Parent, Katy ISD:
What makes your system work? Houston ISD has a performance based pay program that is not working. How much is based on standardized test results?
Only part of it is based upon test scores, and that is probably one reason ProComp is perceived to be working. It is “working” in the sense that teachers, the district, and voters approved of the idea and actually implemented it, which, as you know, hasn’t happened in too many other places. Many other programs have been imposed, top-down and have been somewhat doomed from the start for that reason.
We don’t yet know that ProComp is “working” in the sense that it is driving achievement - more evaluation is needed for that, though some early studies and results are at least promising. And a broader conception of “working” would not just look at test scores, but teacher and principal satisfaction, teacher attraction and retention and other factors.
Question from Linda Bach NBCT 2002, special education teacher Sumter School District 2, Sumter, SC:
How do these programs effect teachers in special education classes and other pull out programs or fine arts teachers? Can there be equity in all special areas?
One of the things that people often don’t realize is that lots of teachers teach subjects or grades in which there is not an annual, required test. Less than 40% of Denver teachers teach in a class where the CSAP test is given.
ProComp recognizes that, and has a series of other elements - market incentives, school level rewards, teacher skills and professional development - through which teachers can earn bonuses and build their base salary. So, these teachers can still earn rewards and higher pay, even if they don’t have as direct a link to a student achievement test.
Question from Timi Biermann, teacher, Slavens K-8 School:
I am a teacher who has been with DPS for 4 years. If I switched form traditional salary to ProComp, I would make the same or LESS money than on the traditional salary scale. I investigated this matter completely by having meetings with the ProComp officers. This was not an assumption on my part. The ProComp officers informed me that when I arrived at step 13, I would make additional money through ProComp. However, their plan at the time stated they would close the ProComp window to all incumbent teachers the year I would be on step 12. ProComp was touted in the press as a system that would benefit all teachers who are willing to perform at the top of their profession. So why am I unable to get ANY additional money through ProComp? How will DCTA make ProComp a more equitable system, where all teachers have at least the OPPORTUNITY to earn additional money?
One of the guiding principles behind ProComp was the ability to allow everyone to participate. Granted not every one can be in a hard-to-staff school, Distinguished School, etc., but everyone can agree to set student growth objectives, all could participate in professional development.
By capping participation, the district’s proposal diminishes both the spirit of ProComp and the incentives for career teachers to take initiative.
We are working to make sure ProComp is sustained as a system of incentives -- where all can take initiative -- rather than a quota system -- where bonuses are given or taken away based on arbitrary factors. Question from Jim Schatzman, Educator, Aurora Public Schools:
I imagine the main worry that many teachers may have about pay for performance has to do with assessment. We already experience many apparent injustices regarding how teachers are treated in the public schools system. What steps will be taken to see that the implementation of pay-for-performance fairly reflects teachers’ effort and skill?
Much of pay for ProComp teachers is still based upon their experience, as well as their demonstrated skills and effort (as evaluated by their principal). ProComp is trying to tie some incentives for improving student performance to rewards for teachers.
Some from the private sector have suggested that much more of ProComp pay should come from direct measures of test scores, etc. For a variety of technical reasons, this would often misidentify teachers and reward some of the wrong ones. So, it is understandable for teachers to be concerned about such radical proposals.
ProComp strikes a balance between providing incentives, allowing good teachers to build their base pay, but also not putting all of the eggs in a basket that has some holes.
Question from Linda Tafel, National-Louis University:
What could other districts who are considering pay for performance models learn from the Denver experience--the good, the challenges?
Being flexible in the approach is one lesson. No one has all of the “right” technical answers to how teachers should be paid, to best enhance student performance. Research does show that the input based (seniority and formal education) single salary schedule does not relate directly to student performance. So, it is worth trying new approaches. On the other side, however, those with private sector experience sometimes just say “make all of teacher pay based upon student performance,” but that is very naive and ignores not only problems with measuring student performance precisely, but also the fact that this isn’t really how most people who work in the private sector are paid.
So, balance and flexibility are key to making it work, as well as compromise and negotiations among the key stakeholders, rather than top down imposition or incentive plan.
The reality is that these plans are hard to implement too. The current negotiations between DPS and DCTA show that agreement at one point in time, doesn’t mean the parties will agree later - learning and adaptation needs to take place in implementation as well as planning.
Question from Ben Jackson, Teacher, Bruce Randolph School:
We know money is not the only thing that attracts teachers to work at hard to serve schools, but we know it is an important factor. How has the union come to conclude that bonuses for teachers in those hard-to-fill jobs need to be raised, but not by nearly as much?
In the spirit of negotiations -- back and forth between parties to achieve an agreement -- we have been open to the district’s interest in a number of areas, including raising salaries in the first five years, taking a look at whether the incentives for hard-to-serve and hard-to-staff are sufficient.
We would hope the final agreement strikes a balance among those interests. We cannot accept a plan that puts so much attention on the first 13 years that it doesn’t prepare for the future and doesn’t recognize the value of experienced teachers.
Question from Lenna Ojure, Director of Teacher Education, Washington and Lee University:
I noticed that student achievement is one of the factors considered in the plan. Are other indicators also being considered such as improved attendance, greater participation in higher level courses,etc? How exactly is achievement being measured?
There three basic areas that look at student growth and development -- intended to make sure that the plan is not just “pay for test scores.” One of those elements is the Colorado statewide test (CSAP), but there is also an element for Annual Objectives, set collaboratively by teachers and principals, and Distinguished Schools, that rewards significant progress in a variety of areas which include test scores.
As your question recognizes, student performance needs to look at a wide range of issues -- including attendance, parent satisfaction, and student growth (not just the bottom line score for a group of students compared to their predecessors).
Question from Joe Farragher, National Account Executive, WIDE World at Harvard’s Graduate School of Education:
I am intrigued to learn about some of the reasons given by the 50% of your teachers who have not opted into ProComp as of yet?
People have a variety of reasons for not opting in the ProComp, and we haven’t done a comprehensive study of the motivations.
Some may look at the chart and see that it is in their economic interest to continue outside the system.
Some may be philosophically opposed to some or all of the elements.
Some may harbor some distrust that the system will be drastically changed and end up being to their disadvantage.
We have been working to develop trust in the system -- and find ways to balance the needs of a diverse teaching force. Maintaining the balance of salary factors and using the salary schedule to enhance incentives is one way to build continued reliance that ProComp is fair and workable. Question from Jack, Teacher John Bowne High School:
Won’t this program (as some studies show) put morale within the school culture at a low? Studies indicate that pay rise for performance creates resentment among fellow colleagues. It has also been suggested that teachers should not be held solely accountable by their salaries for all of these NCLB regulations which bind our hands. What do you think of this in relation to Denver’s system of pay?
Again, the balanced approach of ProComp helps here. Many of the incentives are not related directly to NCLB/test score issues. The fact is that ProComp puts more money on the table for teachers salaries, money that Denver voters approved. So, teachers should see it as a positive appraoch (and they do not have to opt-in, if they don’t). In addition, data from the pilot program that led to ProComp showed that teachers did not compete with one another around the bonuses, to lower morale, instead they tried to help their colleagues improve their teaching, to do a better job and to earn more money. This was one of the cases where empirical evidence from the pilot allayed a concern like you have expressed here.
Question from Kim Bridges, School Board Member, Richmond Public Schools:
How does the intense public interest (and thus, media coverage) in this new system impact the parties’ ability to iron out differences? And once that is done, how “transportable” do you think most of ProComps features will be to other school systems? Are there some elements that are more localized than others or do you believe that most of the program could be used elsewhere?
I think that the parties are closer than most realize and that some compromises can work this out, advantageously for both sides. I actually think the public and media attention is good for teacher compensation reform across the country, because it raises the level of awareness and knowledge of ProComp and similar programs. While I don’t think other districts should “xerox” ProComp and apply it immediately, they can and should borrow elements, combine them perhaps in different ways to fit their unique local circumstances, and make it work for them - and they probably don’t need the 6 years of preparation that it took for ProComp, since the ice is now broken.
Question from Adonna Pennington, Teacher Secondary, Sheldon ISD:
I got my bonus last year, hree in Texas. But it was a struggle. I had 147 Freshmen, averaging 28 per class in a low socioeconomic area(73% on free lunchor asst.) the first year only 12 students received an acceptable rating. This year, no one was below 60%. Even with my masters almost completed, IT WAS TOUGH!
My question is, without varous incentives, how are we to keep our teachers working hard like this? Without a bonus or not, I am in it for the long hall.
PS Don’t ask me how I did it. Basically, I didn’t have a life last year. LOL
I believe that teachers go the extra mile because they care about children and they believe in what they are doing. I don’t believe it is too much to say that every teacher should be paid as a professional and treated as a professional. As you said yourself, you are in it for the long haul.
We’ve been willing to try some incentives to help create a culture of cooperation, lifelong learning, setting and meeting goals. If you take away those incentives after the 13th year, it sends a strong signal that they aren’t important.
Question from Karen Kramer, stay at home mom, Tolland, CT:
Teachers today are saddled with more challenges year after year. Some may say teachers may act as substitutes for parents, due in part to lack of parental involvement or any other factors that keep parents disengaged from their children. Are there any provisions in ProCamp that address more parental involvement or responsibilities? Is it being looked at as a partnership in the community with parents and teachers and administrators working together?
Thank you for the question. Parents and teachers seem to be the only ones who really understand how important parent-teacher communication is to improving student achievement. Active, involved parents understand that students are not always getting the help and support they need at home.
ProComp does not address that directly, but DCTA has been working with parents to address some key issues, including discipline and social promotion, that go right to the heart of better cooperation between parents and teachers.
Our main bargaining proposal to address this issue is our effort to designate about an hour and a half a week in time that is not teaching, not meetings, not required duties -- but time to plan, analyze, and reach out to parents.
Question from Tracy Herman, Researcher, Hezel Associates, Syracuse, NY:
As other states consider a teacher compensation system, what newer factors do you think the research evaluators should be looking for that may differ from what the Denver evaluator is examing?
The challenge for research, is to get as much good data prior to the change in compensation policy, as well as afterwards, so that stronger conclusions can be drawn. With any compensation plan, there are selection effects as well as outcome effects. Since ProComp is an opt-in system, it is possible that the teacher who opt-in are the ones who expect to earn more incentives - if they actually do, it may not be from the incentives themselves, but from the types of teachers. Or, it may be the incentives - it is hard to sort out. Good pre/post data will help evaluate that better.
Question from Scott Reynolds, Curriculum Specialist, Charlotte-Mecklenburg Schools:
In hindsight, what advice would you give to large public-school districts beginning to embark on a compensation reform plan similar to that of ProComp?
Working with the different stakeholder groups, including the unions, rather than trying to impose the ideas from above, is one lessons of ProComp’s successful development. Another, I believe, is recognition that the “science” of compensation reform still has a ways to go - even student longitudinal performance and teacher value added measures, which make lots of common sense to utilize, have some real technical problems, with measures, variability, etc, that make them less than perfect for paying teachers. So, multiple measures of performance make sense, and they also conform with how many employees in other fields are really paid.
Another lesson is that implementation is harder than it might seem. Unless your district has cutting-edge personnel, IT and human resources systems, it will take some considerable time and effort to connect these, in order to actually pay the right teachers the right amount, in a timely manner.
Vaishali Honawar (Moderator):
Thanks very much for all the great questions, and many thanks to Ms. Ursetta and Mr. Teske for their time. We still have many questions but are out of time, so we’ll have to stop now. A transcript of this chat will be available on Education Week’s Web site shortly: http://www.edweek.org/chat/
The Fine Print
All questions are screened by an edweek.org editor and the guest speaker prior to posting. A question is not displayed until it is answered by the guest speaker. Due to the volume of questions received, we cannot guarantee that all questions will be answered, or answered in the order of submission. Guests and hosts may decline to answer any questions. Concise questions are strongly encouraged.
Please be sure to include your name and affiliation when posting your question.