Systemic School Reform in Pennsylvania:
A Policy Proposal for Grades 1-8
Robert P. Strauss*
August 28, 1999
* Professor of Economics and Public Policy, H. John Heinz School of Public Policy and Management, Carnegie-Mellon University, Pittsburgh, Pennsylvania, 15213-3890. Phone: 412-268-4798. Email: RS9F@Andrew.CMU.Edu. Home Page: http://www.heinz.cmu.edu/~rs9f/. The opinions and views in this paper are those solely of the author, and do not represent those of Carnegie-Mellon University or its Board of Trustees.
The purpose of this paper is to describe and explain what can create systemic, constructive change in Pennsylvania’s system of public education. It is not a piecemeal reform, but rather a very major revision to the existing system of public education. The new system is composed of the same major pieces, but the process by which they relate to each other is redesigned with the result that better outcomes in terms of student learning are very likely.
The core ingredients of this set of policy proposals are:
1) high levels of quality information about student outcomes;
2) a material increase in parental involvement in public education and the introduction of strong elements of choice for parents which retain the structure of public schools;
3) revision and clarification of rights and responsibilities of stake holders that entails revising materially the environment within which school directors and superintendents make decisions; and,
4) the continued presumption that local control is the mantra of public education in Pennsylvania, and that much of the existing superstructure of education remains in place.
A central consideration of addressing 3), which no state has effectively done, is that unless Pennsylvania’s teacher force is materially improved in terms of content knowledge as they are replaced during the retirement cliff, the chances for significantly improving student outcomes is low. That is, unless the teacher hiring decision is cleaned up in Pennsylvania, all bets are off.
The core assumptions are that high stakes testing at the beginning and end of the school year: 1) inform stake holders (children, parents, teachers, school administrators, and school board directors) what ‘problem’ prospectively they need to solve during the academic year, and 2) indicate, afterwards, what success in solving the educational ‘problem’ they achieved during the year.
Providing parents with this information and requiring them to choose who should teach their child in the subsequent year changes in a very radical way the incentive structure facing all the stakeholders. Currently, assignment of student to teacher is a delegated responsibility by the General Assembly to the local school boards, which in turn have delegated it to their administrators. The idea of parental choice of teachers is to statutorily delegate this responsibility instead to parents (who can choose not to exercise it). School administrators implement these choices on behalf of the General Assembly and their delegates, the local boards.
Very quickly parents will take it upon themselves to demand information about what has and will happen in the classroom, and school boards, administrators and teachers will be forced into meeting these demands. During the school year, parents will also monitor what happens and become engaged in working with their child and school to ensure that gain in student learning, what everyone agrees is the desired outcome of the educational process, in fact occurs.
It is widely agreed that where parents are involved in the schools, student achievement is greater.
Under the proposed public school choice model, choice is introduced in a sequence of opportunities: First, choice among teachers within the building.
Second, if no choices within the building are acceptable to parents, then they are, under this model, enabled to elect to send their child to another public building in the same district. This in effect is a magnet portion of the model.
Third, if intra-district choice is not attractive to the parent, then the parent may take his child to a building in another district if the district is willing to accept the payment. This across district movement is associated with the loss of per-pupil state aid from the origin district, and the state adding to, if necessary, the destination district’s budget sufficient funds. The destination district must, however, be willing to accept the student.
Fourth, as a last resort, significant vouchers are available to allow parents for whom these public education choices are not acceptable.
Reliable information linking median gain scores to teachers is critical over time to enable parents to make informed choices.
II. Parental Choice of Teachers and Cleaning up the Hiring Decision
It is likely that increased parental scrutiny of who teaches their child will do two things:
a) it will strongly encourage continuing teachers to be more responsive to children and parents as to what their educational activities will be and encourage them to make meaningful self-investment in professional development that improves their subject and pedagogical skills; and,
b) it will accelerate the process of retirement, and thereby create an opportunity in many districts to renew their teacher force at a lower cost.
Central to improving the quality of Pennsylvania’s teacher force is to radically improve the content knowledge of its new hires. Given the statutory and judicial setting in which school boards currently make the hiring decisions, improving in a systematic way the quality of hiring decisions requires some very major tinkering.
To do this, the laws governing school board director conduct will need to be changed so that they:
a) positively affirm they and their agents (the school administrators) are to make decisions solely for educating each child to his/her intellectual capacity
b) agree not to directly or indirectly personally benefit from the programs or activities of the district in a manner that differs from any citizen of the district (more precise language on self-dealing is contained in my May 13, 1999 Congressional testimony on my web page), and
c) agree to disclose publicly their personal and business finances which are independently audited for conformity with b)’s conflict of interest prohibitions
In economic terms the legal framework is changed to eliminate much of the rent-seeking behavior which school directors engage in. It is only realistic to then pay them for their 400 to 500 hours/year of volunteer time. At $7,500/director and 4,500 directors statewide, this creates some new budgetary costs of $38 million per year.
Given school districts are agents of the General Assembly, they should be paid by the General Assembly. Receiving a monthly check cut by the General Assembly will have more than symbolic value, it will remind both the General Assembly and the school directors where much of public education funding is coming from. It also is a recognition by the Executive Branch that the General Assembly is the constitutional, central responsibility for public education in the Commonwealth.
III. The Choice Parents will Have Available to Them
In August, school boards will effect the mailing to each parent the teachers available in the upcoming school year. Parents will be provided information about which teachers are available to choose among, the beginning and ending average and quartile student test scores from last year, some information about the teachers (college of BA, MA, professional development, awards, years of experience) which will be especially important for new hires without a track record in the district. It would be especially helpful to the parents if they are also provided with the prior Spring’s test results of their child and some building-wide comparisons. Parents would not be guaranteed their first choice, and would not be forced to take less than, say, their third choice.
Parents will be asked to indicate:
1) do they want to let the District make the teacher assignment? Yes No
If no, rank their choices from 1 to n:
2) if none of the above in that building is acceptable: rank teachers 1 to n in other buildings in district.
3) if none in 2) is acceptable, the parent is to indicate a teacher in a building in another district which is acceptable to the parent and which has indicated it is willing to accept the child.
4) if none in 3) is acceptable, the parent requests a voucher in the amount of $3,500.
IV. Sequencing the Changes and the Nature of the Information Available
Getting school directors affirming their decision making and ethical responsibilities, disclosing and ultimately being audited, and paid are urgent first steps which can be done relatively quickly should they be agreed to legislatively. It highlights that reform means refashioning local control to focus on students learning to their intellectual capacity. Raising the ethical standards that govern how board decisions are made should erase much of the public cynicism about the ability or willingness of public education to change. It also should create initial support from superintendents, teachers and their unions who often complain that boards meddle for reasons other than making sure that education works.
Undoubtedly both parents and school administrators will need to learn how to express and manage the preferences for teachers, so it is suggested that the first year be one in which preferences are advisory rather than mandatory. As a result, there would be limitations on the amount of choice put into place (so, vouchers and inter-district choice would not be available in the first year), and the rankings could be advisory to principals in the first year. In subsequent years, no child can be assigned any teacher below his/her third choice.
State testing in late spring 2000 for each grade level should be performed, and the results accumulated by building, teacher, and grade level for provision by the local Board to parents in August of 2000 for the advisory year of teacher choice in 2000-2001. The tests should permit the determination not only if students ended the school year at grade level in basic skills, but also differentiate in terms of achievement. This requires state testing at each grade level in, say, May, 2000.
In mid-September, 2000, the beginning of the year state tests should be administered and results provided to teachers showing the average and distribution of ‘starting value test results’ but not individual student results. Parents and students would receive the individual results. The superintendent would be obligated to report publicly to the Board by no later than October 1 of 2000 what the building level scores (mean and distribution) were and what each teacher was starting with.
In the spring of 2001, the state would again test, and the Superintendent would be obligated to report publicly to the Board of the results on the same basis.
In August of 2001, parents would receive their choice forms, and have available, for the first time, the gain score data by teacher which would enable them to make informed choices.
By making publicly available the beginning and ending student test results, school boards would begin to develop oversight procedures, and superintendents and their staffs required to report on promises made and promises kept.
V. A Comment about the the Fall and Spring Tests
There are three major vendors of tests which can be administered in the fall and spring: Iowa Tests of Basic Skills, Stanford-Benet etc. They cost on the order of $3/child/test retail; moreoever, quantity discounts are available. The tests examine reading math etc. and give grade equivalent analysis. Using national tests has its strengths and weaknesses, but is the only way to find out where the kids are at on a consistent basis, and also to demonstrate nationally to employers that the state is serious about getting its kids ready for higher education and/or the labor market. It is imaginable that parents are acutely aware of preparing children for the rigors of the next century’s labor market and the international competition in skills.