Category Archives: education

The Student Achievement Guarantee in Education (SAGE) contracts for MMSD schools will be on the agenda at Monday’s (3-10-2008) Special Board of Education Workshop meeting.  I have mixed feelings about the SAGE program because of the choices it forces school district to make.

A serious overhaul of the school funding system is needed and one of the things that should be addressed are the problems with SAGE.  Most of the proposals I’ve seen (Wisconsin Alliance for Excellent Schools, School Finance Network, Alan Odden…) would minimize or eliminate some of the issues discussed below.

I am all in favor of targeting resources (or the money to pay for resources) to children in poverty and schools with high concentrations of children in poverty.  I also think all four parts of the SAGE program are great:

Program Elements

SAGE promotes academic achievement through the implementation of four school improvement strategies:

• class sizes of no more than 15:1 in grades K-3;
• increased collaboration between schools and their communities;
• implementation of a rigorous curriculum; and
• improved professional development and staff-evaluation practices.

SAGE does this by providing districts with $2,000 per student in poverty at SAGE schools (next year it will be $2,500, the first increase since the program started over a decade ago).  I even like the fact that there are some strings associated with the money, that it has to be used in certain ways.  In this fiscal climate legislators and tax payers want to know that their money will be spent wisely and the preponderance of research (and here) indicates that the areas SAGE money can be spent are productive best practices.

The two of the biggest problems with SAGE are that 1) There are a limited number of SAGE contracts, meaning there is a cap on the number of schools (and children) that can benefit from the program (MMSD has 20 contracts);  and 2) SAGE does not direct extra resources to poor children in non-SAGE schools (it isn’t easy being a poor child in a rich school).  I’ll add a number 3, that SAGE does nothing for children after third grade).   As a result of these —  and the fact that SAGE funding is insufficient (it is an under-funded “mandate”) — the SAGE program promotes economic segregation in our schools.

Economic segregation was among the considerations in the recent West-side attendance area boundary discussions.  The Equity Task Force has weighed in with guidelines to minimize economic segregation.  I am an unapologetic believer in promoting integration as a key element of the social mission of public education.   However, the case for  economic integration does not rest solely on these ideals, significant research has demonstrated that poor children tend to achieve more in schools with an economic balance (and here and here and here…. Note that  —  like everything else in education research — there are no absolutes and that there are schools with very high poverty proportions where achievement is also high and schools with low poverty where achievement is not so high).   These finding are reflected in the local data below (see also the “Classmates Count” study).

Graphic taken from “Effect of Concentration of Poverty in School on Reading Scores (MMSD).”

The problems come in because unless there are high concentrations of poverty in individual schools, meeting the SAGE program requirements demands great expenditures from general operating budgets, budgets that are already stretched to near the breaking point.

For simplification, I am only going to do the math for approximate classroom teacher wages and benefits costs (this means that expenses having to do with community collaboration, curriculum, staff development, evaluation, specials teachers, facilities and supplies are not included).  A full time equivalent teacher costs MMSD about $76,000/year in wages in benefits.   There are 28 schools in MMSD serving K-3 (not counting the hand-full of students listed at Lincoln; there will be 29 schools next year).  Among those schools the average number of kindergartners is 72, to make the math easier (and more dramatic), let’s use a school with 63 kindergarten students (these are  crude estimates because the the way the numbers break down with 21/1 and 15/1 are crucial and the use of multi grade classrooms opens up some other possibilities for maximizing SAGE dollars).   At a 21/1 pupil/teacher ratio this would mean the school would require 3 kindergarten teachers and classrooms.

63/21 = 3.0.

At a 15/1 ratio the school would require 5 kindergarten teachers and classrooms.

63/15 = 4.2 (round up to 5…SAGE requires 15/1 or less).

At $76,000 per teacher the difference in cost is $152,000.  Using next year’s SAGE funding ($2,500/student in poverty) it would take about 61 students in poverty to make SAGE to pay for itself.

152,000/2,500 = 60.8 (round up to 61).

Out of a class of 63, this means a poverty proportion of 96.8% is required for SAGE class size reduction to be “fully funded.”  No K-3 schools in Madison are currently at or above this level.  The closer you get to that 96.8% the less general operating money is needed.   Here is a chart for percentage of kindergarten students in poverty and local implementation costs (the unfunded portion) based on the assumptions and calculations above:

This creates a dilemma.  Maximizing SAGE dollars pulls toward concentrating poor children; best practices pushes toward balancing poverty at the school level.

SAGE also creates a related dilemma in the allocation of contracts between big schools with low poverty and small schools with higher poverty numbers.  Using the contract in a big school can bring in more SAGE dollars, but will also require more local dollars also.  Using the contract in a small school will mean fewer total students will benefit and may mean fewer students in poverty benefit.  I’m going to use Gompers and Chavez to illustrate this (see here).

Gompers (2007 figures)

154 K-3 students, 60% low income, about 93 SAGE funded students,

at $2,500/student = $232,500 in SAGE dollars.

Cost differential for 15/1 ratio (four more classrooms) = about $228,000.

Chavez (2007 figures)

482 K-3 students, 27% low income, about 130 SAGE funded students,

at $250,00/student = $325,000 in SAGE dollars.

Cost differential for 15/ratio (12 more classrooms) = about $912,000.

So fully implementing (K-3) a SAGE contract at Chavez instead of Gompers would bring in more money,  serve more students and more students in poverty, but at an additional cost to the district of about $684,000 per year.  Tough choices.

In Madison these choice are made even more difficult by the fact that we have about seven schools between 23% and 33%  poverty level, but only enough SAGE contracts for two or three these schools.  These schools vary greatly in size, and the exact percentages cannot be known till after the third Friday counts in September, further complicating the issue and making the equity based choices even more elusive.

In the past Madison has worked around some of these issues via implementing various levels of SAGE (K-1, K-3, whole school…) and using local funds to reduce class size in non-SAGE schools.  Madison has also won praise for leveraging federal, state and local monies to maximize the impact of all the dollars (see: Resource Distribution in the Implementation of Class Size Reduction Policy: Looking Inside the Black Box of District Practice, MMSD is “Maxwell”). Last year was the first year the district moved away from locally based class size reductions.  Without a successful referendum in November 2008, it won’t be the last.

In closing, there are some questions surrounding what options a district has in transferring SAGE contracts.  Last year the administration analysed choices based on the assumption that contracts could be moved (and here). Recently, the Board of Education was advised that “neither the statutes nor the administrative rules expressly prohibit the transfer of a contract.”  The DPI guidelines from February of 2007 state:

Transfer of contracts has been allowed when SAGE schools have been closed, consolidated, or moved to new buildings to ensure the benefits of the program could follow the students to their new location.

• Within the term of a five-year SAGE contract the contract may be moved by the district from an existing school to a different school more in need of the program only with the consent of the recognized representatives of both the staff and parents of the school giving up the contract
• At the end of a five-year contract the district board may transfer a SAGE contract from one school to another the SAGE requirements will immediately apply to the school to which the contract is transferred.

I don’t know what decisions the Board might make on Monday.  With a matter this complicated and with budgetary and equity consequences for the entire district, I believe that in the absence of guidelines or policy directly addressing the issues, these discussions and decisions should take place as part of the budget process and not as a separate item.  I also wish the Board the best with these very difficult issues.  Last, I hope that the community understands that there are no easy or clear choices and that the Board must weigh many factors and options with an eye on what is best for the district as a whole.

Thomas J. Mertz

[Note post edited at 5:42 PM, 3-09-08 to correct mathematical error. The new version uses  a school with 63 kindergarten students as an illustration, the first version used a school with 72.  Because of MMSD policies and the way the numbers work out the cost differences for a school with between 63 and 91 students in a grade would not be as dramatic (only one more teacher required).  The district cannot know if a particular school will hit a sweet spot (64, 65, 66,..) or a sour spot (62, 63, 91, 92…).]