You are on page 1of 41

To: From: Subject: Date:

Gary Orfield and Erica Frankenberg Sheldon H. Berman and the Student Assignment Core Planning Team Alternative Adjustments to the Student Assignment Plan January 14, 2011

Beginning last summer and continuing into the fall and winter, the Jefferson County Public School Districts student assignment plan has been repeatedly challenged in the courts, in the mayoral and legislative electoral campaigns, in the legislature, and in the public arena. Although the overall implementation of the plan in terms of registration, notification and transportation was smoother and more expeditious than in previous years, a problematic opening day to the school yearin which some elementary students arrived home very lateaggravated the situation and provoked negative publicity. The movement in favor of neighborhood schools has gained support from some voters and some parents. In response to a proposed neighborhood schools bill, and to better inform policymakers and the public, the district provided documentation and editorials explaining the assignment plan, and the superintendent engaged leading business and community groups in dialogue about the issue (Attachments A and B). However, it became clear that some adjustment(s) to the planparticularly to the elementary component might be necessary in order to reduce tensions within the community and address some of the implementation issues. ISSUES TO BE ADDRESSED In early September, the Superintendent reconvened the Student Assignment Core Planning Team to review the student assignment plan and make recommendations for any adjustments that might address the concerns raised by parents and the community. The team identified five major issues that we wanted to examine in this review. Length of bus ride in terms of time and distance: Although the district had invested a good deal of time and resources in creating express routes, reducing the use of depots, and reducing both the average ride time and the longest ride time on buses, the primary complaint reported in the media and by parents was the distance some students had to travel to their assigned school and the length of time they would spend on the bus as a result. Although parents who selected magnet schools and programs did not strongly object to these ride times, some parents whose children were assigned to schools in a different section of the county deemed the bus ride problematic. This was particularly the case for parents who lived in the east end of the county and whose children were transported to downtown or west end schools, but there were others across the suburban areas of the county who raised the same issue. Some parents expressed a concern that the schools distance from their home might limit their involvement at the school. Therefore, one factor to be considered in any recommended change is whether it reduces the average ride time and distance for students and/or reduces the number of students required to participate in extended travel times. Travel times for kindergarten students: The diversity guidelines in the student assignment plan have never formally included kindergartners. However, in order to avoid moving students from one school in kindergarten to another for first grade, the Student Assignment Office staff has done its best to assign kindergarten students in a way that is consistent with the diversity guidelines. This strategy has been particularly important to principals, who otherwise have the challenge of serving one group of students for kindergarten and another for first grade, as well as having to address parent concerns if the parent receives

Page 1

notification that the childs placement will be changing for the next year. Some parents have expressed concerns that kindergarten-age children should not be traveling on buses for long periods of time and they should be assigned to their resides school. Therefore, one of the issues to be examined is whether it is possible to shorten the ride times or better address the school assignment issues for kindergarten students. Parents ability to secure their first or second choice of schools: Choice of school has been an increasingly important factor in student assignment. In an effort to enhance choice, the district significantly expanded the number and variety of magnet schools and also increased the number of choices offered to parents from two schools to four. However, because parents are required to choose two A area and two B area schools, parents who receive their third or fourth or none of their choices of school are most likely to raise concerns about the student assignment plan. Therefore, any adjustment in the plan needs to at least maintain, or preferably improve, the likelihood that parents could receive their first or second school of choice. Because 42 percent of our parents do not choose their resides school for kindergarten or first grade, it is important to retain a high degree of choice to accommodate parents. Predictability of school to which a student might be assigned: Providing predictability in the choices offered to parents for the school a child might attend was one of the guiding principles used in the development of the student assignment plan. The current plan provides parents the opportunity to identify four choices among the 12 to 15 schools in their cluster, as well as two choices among the magnet schools. During the application period for the 2010-11 school year, 83 percent of parents received their first or second choice, and 94 percent received one of their four choices. The percentage of parents who received their first or second choice dropped significantly in some areas of the district after the application period ended because some schools were at capacity. Although this plan provides parents with a set of predictable choices, it does not guarantee that a child will attend the parents school of choice. In fact, parents who receive their third or fourth choice often regard this school as an assignment rather than a choice. Some parents would like to know where their child would attend school based on their home address and would like to see an attendance area associated with one particular school. One of the motivations for neighborhood schools is that people believe such a system would guarantee the assignment of their child to the school closest to them. Some districts with student assignment plans that provide parents with choices have published decision rules that guide assignment decisions, thereby enhancing predictability. Although we, too, use various facets to help make determinations, we do not publish them. Therefore, another factor to be considered in recommending any change to the plan is its impact on the predictability of the school to which a child would be assigned. Diversity of student population: A final factor in reviewing any adjustments to the plan is its impact on the racial, ethnic and socio-economic diversity within the student population at the school. As the foundation for any student assignment plan, the impact on the plans ability to sustain or enhance diversity is a critical element. Under the current student assignment plan, there are student populations that are not included in the diversity guideline, such as English as a Second Language (ESL) students, Exceptional Child Education (ECE) students in self-contained programs, and kindergartners. By not including ESL students in the diversity count, we do not recognize the full diversity within a school. The five factors described above became central considerations as the Core Planning Team developed and reviewed alternative adjustments to the student assignment plan. ALTERNATIVES CONSIDERED In the ten meetings that were held from September through January, the Core Planning Team brainstormed numerous potential adjustments to the student assignment plan. We then examined each possibility for its effectiveness in addressing concerns and for our ability to execute it effectively. A detailed analysis of the benefits, drawbacks, timeline for implementation and areas of concern addressed is attached (see Attachment C). It is clear that there is no option that solves all the problems and that any option has both benefits and drawbacks. The multiplicity of alternatives evolved into three larger sets of ideas with multiple ways of putting some of them into practice. The three options are not necessarily mutually exclusive and could be phased into the plan at different times.

Page 2

These three options included: Identifying areas of A students within the B areas and areas of B students within the A areas. Creating kindergarten or early childhood/kindergarten centers in each cluster. Restructuring the six current clusters. Area A within Area B & Area B within Area A During the development of the student assignment plan in 2007-08, we were particularly conscious of needing to move to a geography-based plan incorporating multiple criteria. Initially, we examined four criteria: Minority status, income status, educational attainment of adults, and single-parent households. Because there was significant overlap between income status and single-parent households, we dropped single-parent households as a criterion. Using the three remaining criteria, the planning group in 2007-08 was able to identify area A resides schools that represented higher-than-average minority status and lowerthan-average income status and educational attainment. Area B resides schools represent lower-thanaverage minority status or higher-than-average income status or educational attainment. At the time, we realized that there were pockets of students in each area that more closely represented the demographics of the other area and that some schools that were labeled as area A or area B schools actually would have met the guidelines for diversity if individual student demographics rather than geography were the criteria for assignment. In fact, we explored the possibility of identifying census block areas within a resides that were subareas representing the demographic factors of the other geography, that is a block group representing the demographics of area A within an area B resides or a block group representing the demographics of area B within an area A resides. We felt that identifying these pockets of diversity within a resides area might better define the demographics of the area and reduce the need to transport as many students to meet the diversity guideline. However, in the end our analysis did not identify these subareas in sufficient size or quantity to make a substantial difference. Therefore, we established 25 schools in which the entire resides was considered area A and 57 schools in which the entire resides was considered area B, even though there may have been groups of students who met the demographic standards for the other area within those boundaries. As we began to reevaluate the operation of the current plan, our attention returned to this earlier analysis. Staff suggested that we reexamine the demographic analysis, this time dropping the criterion of educational attainment since there was a good deal of overlap between educational attainment and income status. With only two criteria remainingminority status and income statusblock areas emerged within both the A and B areas representing the other population demographics1. In fact, in 16 schools there are a sufficient number of subarea A students within B and subarea B students within A in the schools resides area that the school would meet the diversity guideline with its resides population. In another 26 schools, there are enough subarea A students within B and subarea B students within A to offset some of the need to transport students in order to meet the diversity guideline. A map showing these areas is attached, as is the analysis of the number of students within each resides who would be identified as from subareas A or B (Attachment D). The benefits of this new A/B configuration are that it allows more students to remain in the resides school, reduces the number of students who are involved in the exchange, and better recognizes the diversity that exists within some neighborhoods. As a result, it would reduce the number of students being transported and potentially reduce the ride times and cross-county exchanges. The most significant drawback to this adjustment is that it creates a more complex definition of diversity that is harder to articulate to the community. Neighbors within a schools resides would represent different demographic populations with different chances of being able to remain at the resides school. In addition, 40 of the 82 non-magnet elementary schools would remain entirely area A or area B, with most of those being on the extreme ends of the county. However, we may be able to address some of the exchanges using the new A/B
Census block groups were used to establish subareas. A block group is a subdivision of a census tract and is the smallest geographic unit for which the Census Bureau tabulates sample data. A block group consists of all the blocks within a census tract with the same beginning number.
1

Page 3

configuration so that students are assigned to schools nearer the center of the cluster. Finally, current A area schools that have subareas of B demographics within them may be adequately balanced for minority status, yet be less well balanced for socio-economic status; that is, these schools will still have a high percentage of students whose median family income is below the county average. Using the new A/B configuration has implications for the middle and high school components of the plan as well, since it enables us to identify subareas within the resides of those schools and their satellites. As with the elementary component, the A/B configuration allows us to more accurately recognize the diversity that exists in these schools. An analysis of the middle and high school plan components currently approved by the Board of Education reveals that changes may be necessary to address the differences in diversity revealed by the new A/B configuration. We are still reviewing precisely what those changes might entail. Tables showing the middle and high school A/B ratios under the current plan and under the new A/B configuration are attached (see Attachment E). Because the high school boundary changes do not go into effect until 2012-13, adjustments in the school boundaries to reflect the new A/B configuration could easily be made in time for the regular communication cycle for student assignment. However, the middle school component, with its new boundaries that were approved in the fall of 2010, is slated to go into effect in 2011-12. If the new A/B configuration were to be approved, it would result in additional changes to boundaries that would impact approximately 900 students, 80 of whom would be returned to the school boundaries that existed prior to the 2011. There are three options. The changes could be implemented this year, thereby immediately impacting these students for 2011-12. They could be made in a future year, creating a second adjustment in the plan. Or the implementation of the middle school component could be postponed until 2012-13, at which time all the changesthe boundary changes and new A/B configurationcould be integrated simultaneously. Postponing implementation would require immediate notification to 1500 students that their assigned school for next year would revert back to current resides area. The Core Planning Team felt that changes should only be made once rather than twice, in order to avoid a second adjustment in the plan and a second set of grandfathering. Therefore, we recommended against the second option of phasing in the new A/B in a later year. However, it wasnt clear whether the new A/B changes should be made later this year or whether implementation of the entire middle school component should be delayed. Both scenarios have logistical and political implications. The proportion of area A students in the district under the current plan is approximately 30 percent. With the new A/B plan, that figure would rise to 35 percent, more accurately reflecting the demographics of the county. Given this increase, it may be advantageous to move the diversity guideline from 15-50 to 15-55, thereby allowing a margin of 20 percentage points on either side of the districtwide average of 35 percent A area students. This change to 15-55 would actually allow one more school to meet the guideline, further reduce the number of students who would need to be transported, and further reduce average transportation time. However, it has several drawbacks. It could appear as if the Board is retreating on its commitment to diversity by broadening the guideline. It could also increase the number of schools that have higher concentrations of minority and high-poverty students. And it could provoke a conversation about expanding the guideline even further. This first of the three options could be implemented sooner than the other two because it does not require structural changes in the configuration of the clusters, nor does it necessitate the redesignation of space or buildings for kindergarten centers. It would sustain choice and diversity while reducing the number of students transported and the average length of bus ride. Our attorneys have advised the Core Planning Team that having two criteria (minority status and income status) instead of three would still meet the Supreme Courts guidance since it remains a geography-based plan. Kindergarten or Early Childhood Centers The issue of transporting kindergartners needs to be considered within the context that many of the districts pre-school children ages 3 and 4 ride a bus to school for an extended period of time. However, the ride time for kindergarten children is an area in which the district and the Board receive complaints. The Core Planning Team discussed the benefits and drawbacks of including kindergarten children in the

Page 4

student assignment plan or modifying the current way we assign kindergarten children. That issue will be discussed more fully in the next section on decision points. However, one suggestion that we explored to address the issue of bus rides for kindergarten students was the option of creating two to three kindergarten centers within each cluster. All kindergartners would attend these centers. These centers could also include pre-school programs so that they could become early childhood centers and be better aligned to support the needs of the 3-5 year olds. No elementary students would attend these centers. The primary advantage of kindergarten centers is that, because they would be centrally located, they could reduce the length of ride for some students and alleviate cross-county exchanges. All students would attend the center near their home, and the assignment of students within the diversity plan would begin in first grade. There would not be an expectation that some students would be staying in the same building for elementary school, thereby eliminating the issues that currently exist when some students are required to move to a new school for first grade. The primary difficulty with this option is space. There are three possibilities for identifying space for these centers. The centers could take over current elementary school buildings, with the elementary students being redistricted to other schools. The centers could be placed in high schools where there is existing space. Or the district could build new centers designed specifically as early childhood or kindergarten centers. Pre-school and kindergarten programs must be located on the ground level of a building, thereby making the identification of space or schools more challenging. Each of these alternatives has significant drawbacks. The repurposing of elementary buildings would require the displacement of elementary students to other buildings. This would be a politically charged issue. Placing these programs in secondary schools, if space were available, would require extensive renovation of the facilities to accommodate younger children and isolate them from older students. Building kindergarten or early childhood centers would take an extended time, and construction funds would have to come from general fund resources rather than construction funds. There are other, more general, drawbacks to consider. First, many students would actually ride the bus longer than they currently do since these centers would be fewer in number and more centrally located than our elementary schools. Second, the children would most often be riding the bus without the support of an older sibling. Third, the centers would substantially increase transportation costs because routes would have to be covered four or even five timesonce for pre-school, once for kindergarten, once for local routes, once for cluster exchange, and once for grandfathering. The kindergarten centers would provide predictability, reduce ride time for some students, and avoid crosscounty exchanges. They would also alleviate the need for the discussion of whether kindergartners were included or excluded from the plan. However, this option would actually increase ride time for many children, would be disruptive to the current school configurations, and would be costly to implement. In addition, every child would have to adjust to a different school environment upon entering first grade, thereby reducing the plans attention to stability. Restructuring the Six Current Clusters Another option that was explored by the Core Planning Team was changing the cluster configuration in order to reduce transportation time and ride length. Multiple ways of restructuring were examined, including creating subclusters within the existing clusters, moving to a non-contiguous cluster plan, and moving schools from one cluster to another. The primary objective of restructuring would be to create pairings of schools that would reduce ride time for students in the exchange. The idea of subclusters within each of the existing clusters was explored as a potential adjustment for the 2010-11 school year, but was not implemented. There could be two subclusters of six to eight schools within each cluster. Schools on one end of the cluster would be paired with schools nearer the center of the cluster, thereby reducing travel distance. Although parents would still be able to select two A area schools and two B area schools, the choices would be from a more limited set of schools. The drawbacks to this approach are that it may not realistically reduce transportation time, would increase the complexity of the transportation system, and would limit parent choice. Because we have created magnet programs within

Page 5

each cluster, parents would still have access to these magnets, thereby requiring transportation to all A area schools within the cluster. If implemented in 2011-12, it would also add another layer of grandfathering to the existing plan, with students in grades 4 and 5 assigned under the previous plan, students in grades 2 and 3 assigned under the current plan, and students in grades K and 1 assigned under the new plan. Moving to a non-contiguous plan might have the advantage of being more familiar to parents who had children in school prior to the current plans implementation. Because the non-contiguous plan has more clusters and fewer schools in each cluster, it might reduce transportation time or distance. However, it has some of the same drawbacks as subclusters and, because it is so different from the current plan, it would create significant changes in the choice of schools for parents and a far greater degree of grandfathering. In addition, the travel distances in the non-contiguous plan presented in 2008 or the non-contiguous plan prior to 2007 are not so significantly different from the travel distances in the current contiguous plan as to reduce concerns about the busing of children (as delineated in Attachment F). Returning to the noncontiguous plan prior to 2007 would also produce a greater differential in the percentage of area A and area B students between each cluster, thereby creating advantaged and disadvantaged clusters. The contiguous cluster plan has the benefit of greater efficiency in transportation over the non-contiguous plan once grandfathering is over. This efficiency would be diminished with a move to the non-contiguous plan. Finally, the Core Planning Team explored the idea of moving schools from one cluster to another to address transportation issues. The benefit of this approach was that it would leave the current clusters substantially the same, while addressing specific problems in cross-county exchanges. However, after examining a couple of case studies, the Team could not find a way to accomplish this effectively. Moving a school from one cluster to another did not seem to significantly address issues of transportation or distance, but simply moved the problem from one cluster to another. In addition, it had the potential to disrupt the current set of choices, add another layer of grandfathering for a small group of students, and reduce the diversity balance achieved in the current configuration of clusters. DECISION POINTS There are three key decisions related to adjustments in the student assignment plan that the Core Team believes the Board should consider. These include the timing of implementation of any change, the inclusion of ESL students in the diversity count within a building, and the inclusion of kindergartners in the plan. Timing Each of the options the Team considered has timing implications. The new A/B plan could be implemented most quickly because it entails the fewest structural changes. The restructuring of clusters and the kindergarten center options would not be able to be implemented until 2012-13 at the earliest. However, any new plan should take into consideration new demographic information from the 2010 census so as to avoid making a second set of adjustments at a later time. We had anticipated that detailed census data would be released in December, but only broader population data were released at that time. It is likely that the demographic information we need will be released some time this spring. Therefore, it may be best to consider these options now from a conceptual framework, and then wait for the census information to be released before creating a more concrete proposal. In addition, any redesign of the student assignment plan will have implications for the opening of school the following year. Significant changes in the plan diminish the likelihood of a smooth opening because they will entail new routes and, depending on the option selected, new pairings of schools. It would be beneficial to our students, our district, and our entire community to have a smooth opening in August 2011. Therefore, the start of the school year should be a consideration in the timing of any change that is selected. Inclusion of ESL Students Throughout most of the districts history of student assignment, ESL students were included in the plan as other (i.e., other than African-American), thereby assisting the district to achieve diversity by placing

Page 6

ESL units in high-minority schools. When the new plan was approved in 2008, ESL students were excluded from the diversity count based on the rationale that the district needed to place these specialized classrooms where there was space and where the students fit best into the instructional and support services of the schools. Over the past two years, the district has worked to distribute ESL classes more equitably across schools. The problem with not including ESL students in the diversity count is that it does not acknowledge the existing diversity within a school or give a school credit for inclusion of ESL students. Because approximately 50 percent of ESL students are area A students and 50 percent are area B students, they can have an impact on the diversity ratio in a school. At this point there is a consensus among members of the Core Planning Team that it would be more effective to include ESL students in the diversity count within the student assignment plan. This step would recognize the diversity within a school, provide a more realistic picture of the resides area of a school, allow more students to remain in their resides area, and enable us to better place programs to support a diversity balance among schools. Inclusion of Kindergarten Students in the Student Assignment Plan There are both significant advantages and disadvantages to including kindergartners in the plan. On the one hand, inclusion enables parents to request a choice of school for kindergarten and know that the assigned school will be the childs school for his or her entire elementary experience. This would provide educational continuity. Parents would not have to reapply for first grade and face the possibility of their child changing schools. (Parents would still have the option each year to apply for a magnet program school or to request a transfer if they were dissatisfied with the current placement.) Including kindergartners would also put the child in the same school with older siblings. Finally, it would ease the anxiety created by having a child attend one school for kindergarten and then transfer to a new school for first grade. Principals clearly support the inclusion of kindergartners because it enables them to more fully integrate the children and parents into the school community. However, the significant drawback to inclusion is that kindergartners would have a greater likelihood of being assigned to a school farther from home and spending more time on the bus. In fact, it is the kindergarten year that has been the flash point for issues about student assignment. The Core Planning Team looked at several ways to address this concern. Very few members of the Team felt that fully excluding kindergarten children from the plan and assigning them to their resides school was a viable approach. However, there were mixed opinions about three other alternatives. Some members believed that full inclusion in the plan was the most effective solution. Others felt that not including kindergartners in the plan but assigning them informally as we currently do would work if we were to use a decision rule in the placement of these children that took distance and ride time into consideration. The informal decision rules currently used do this now to some extent. Finally, some Team members suggested that parents could submit school choice forms prior to the kindergarten year for assignment in first grade. If they received their choice, the child could attend that school in kindergarten. Otherwise, the child would attend kindergarten at the school serving the area in which s/he resided (unless that school was at capacity, which would necessitate a placement at another school). The student would receive a first grade assignment based on the diversity guidelines, which might be at a different school, and would move to the assigned school in first grade. Experience with a prior assignment plan in the 1980s revealed that many parents who knew that their child would attend a different school for first grade would volunteer to move their child to that school for kindergarten. However, other parents who knew that their kindergartner would have to move in first grade put pressure on school and district staff to remain at the current school. Which of these alternatives would be most effective is still an open issue that the Team is continuing to discuss. CONCLUSION There is no one best plan for student assignment. However, the Core Planning Team believes that there is significant merit in the A within B, B within A option, pending a review of the 2010 census data. The inclusion of ESL students in the diversity count would further support this option as a solution to some of the issues raised by the current design of the plan.

Page 7

CORE PLANNING TEAM MEMBERS Sheldon Berman - Superintendent Joe Burks Assistant Superintendent High Schools Rick Caple Director Transportation Barbara Dempsey Specialist Parent Assistance Center Amy Dennes - Assistant Superintendent Elementary Schools Dena Dossett Director District Planning Craig Garrison IT Project Manager Jeanne Giberson Manager Transportation Operations Jack Jacobs Executive Director Student Relations and Safety, Interim Director StudentAssignment/Gheens Academy Marti Kinny ESL Academic Program Consultant Sandra Ledford Assistant Superintendent Middle Schools Stephanie Malone Assistant General Counsel Dan McCubbin Retired General Counsel Kathy McGinnis Director Elementary Student Assignment Frank Mellen Partner Wyatt, Tarrant, and Combs Rosemary Miller General Counsel Bernard Minnis Assistant Superintendent Diversity, Equity, and Poverty Programs Mike Mulheirn Executive Director Facilities/Transportation Bob Rodosky Executive Director Accountability, Research, and Planning Ralph Stephens Demographic Analyst

Page 8

Attachment A

Important Facts About Student Assignment


Elementary Assignment 80 percent of applications from parents of kindergarten students that were received during the application period (Feb. 1 Mar. 9), were assigned to their first or second choice. 87 percent of applications from parents of first grade students that were received during the application period (Feb. 1 Mar. 9), were assigned to their first or second choice. Of the applications from parents of kindergarten students who reside in Area A that were received during the application period (Feb. 1 Mar. 9) and listed the resides school as a first choice, 66 percent were assigned to the resides school. Of the applications from parents of kindergarten students who reside in Area B that were received during the application period (Feb. 1 Mar. 9) and listed the resides school as a first choice, 93 percent were assigned to the resides school. Middle & High School Assignment The proposal approved by the Jefferson County Board of Education on September 13 reflects minor boundary modifications of the current middle and high school plan. The middle and high school plan is not a new assignment plan. Applications for magnet schools, magnet programs, or transfers to schools outside the resides area will be continued. The boundary changes would only affect 1200 of the current fifth grade students (approximately 6% of middle school students). The proposed boundary changes would impact approximately 180 students living in Area B. Transportation On September 13, 2010, the transportation goal of bus rides no longer than 75 minutes was met. Student drop-off times have improved since the beginning of school on August 17. Bus routes are no longer than the plan in place in 2007 and are often shorter. Numerous depots have been eliminated and more routes are direct routes to schools, thereby reducing ride times. Enrollment The district achieved its projected enrollment 99,849 students, which includes Early Childhood, Exceptional Child Education, and Special School students. Student Transfers The district approved 49 percent of all transfer requests received. Forty percent of elementary transfers, 59 percent of middle school transfers, and 62 percent of high school transfer requests were approved. The number of transfer requests this year increased by only 59 from the number submitted in 2009-10. Diversity The plan is beginning to achieve its diversity goal in first and second grades. 59 percent of the elementary schools met the districts diversity guideline a ten percent increase over 2009-10. 76 percent of middle schools have met the diversity guideline. 76 percent of high schools have met the diversity guideline. Important note: Area A is the term given to neighborhoods where the average household income is below $41,000, the average education is a high school diploma or less, and the minority population is more than 48 percent. JCPS diversity guideline calls for schools to have between 15 and 50 percent of its students from Area A.

Page 9

Neighborhood schools. at what cost?


On the surface, the Williams-Seum neighborhood schools plan seems simple. But implementing the plan would be disruptive, costly, and restrictive for Louisville families. Disruption for families Boundaries for all JCPS schools would be revised. Thousands of students currently attending their resides or neighborhood school would be eligible for new school assignments. o 33 percent of elementary students (15,300) o 43 percent of middle school students (9,100) o 36 percent of high school students (8,900) Younger siblings of current students would have no guarantee of placement in the same school. Added Costs Because children are not evenly divided throughout the county, the plan would create overcrowding in some schools and vacancies in others. Community expectations that their children attend the closest school would place significant pressure on the district to build five new elementary schools and one new high school, or transport thousands of students to the next closet school with vacant seats. The cost to taxpayers to build and furnish six new schools would exceed $200 million. The plan, which would allow students to retain their current school assignments, would require the district to operate a more complex, multi-tiered transportation system in order to transport the estimated 7,500 students who would be grandfathered into their current schools (over five years) and all new students assigned to their closest school. Additional transportation costs would be approximately $22 million. Severely Restrict Choice Popular districtwide magnet schools and programs such as traditional schools and Louisville Male High School would very likely be available only to the students living closest to them. In summary, the impact of passing the Williams-Seum bill would: Disrupt families Change every schools boundaries and make 33,300 students eligible for new school assignments. Add costs Potentially cost more than $200 million for 6 new schools. Eliminate choice Access to popular JCPS magnet schools like Louisville Male would be severely limited for students who dont live within the schools boundaries.

Page 10

Jefferson County Public Schools Board of Education Statement Regarding Proposed Student Assignment Legislation The Board of Education of Jefferson County Public Schools (JCPS) believes that policies regarding student assignment and school attendance are best handled by local boards of education. In the strongest terms possible, we respectfully request that the Kentucky General Assembly not adopt legislation that constrains local control over this important policy area. For public schools in Louisville Metro, passage of legislation to allow children to attend the school nearest their home would have a significant negative impact on parental choice, the efficient and equitable allocation of resources, and would, in large measure, resegregate our schools. Local school board members have a deeper and more nuanced understanding of the needs and specific circumstances of their school district than is possible from a state-level perspective. Through extensive, regular, direct contact with parents, employers, and community leaders, school board members are highly attuned to the issues confronting their community. As locally elected officials, school board members are directly accountable to parents and voters in their community, and are responsive to addressing the concerns raised by their constituents. In our role as members of JCPS Board of Education, we work hard to develop policies after weighing input from a variety of stakeholders, balancing at-times competing community values, and seeking workable solutions to complex problems. We continually evaluate the success of those policies, and the ability of district administration to implement them. We are ready, willing and able to alter course and makes changes when problems arise. We are actively engaged in that work right now to address the challenges we face with the design and implementation of the student assignment plan. In 2007, the U.S. Supreme Court ruled that the former JCPS student assignment plan was unconstitutional because it used the race of individual students to determine where a student attended school. After listening to significant public input through public forums, scientific random sample surveys, and other means, after consultation with national experts, and after extensive board discussion and deliberation, the JCPS Board of Education adopted a set of principles to guide the development of the current student assignment plan. We believe that those guiding principles - diversity, quality, choice, predictability, stability, and equity - reflect fundamental values our community holds, and remain the correct ones to rely upon as we review, adjust, and implement the student assignment plan. While our challenge as a board is to find a workable balance of those guiding principles, it should be noted that fully 88 percent of over 3,000 survey respondents said it was important for students from diverse backgrounds to be brought together in schools. Passage of the proposed legislation on student assignment would significantly undermine our ability to remain fully responsive to our communitys aspirations for its school system and, ultimately, our collective future.

Page 11

Neighborhood schools. at what cost?


On the surface, the Williams-Seum neighborhood schools plan seems simple. But implementing the plan would be disruptive, costly, and restrictive for Louisville families. Disruption for families Boundaries for all JCPS schools would be revised. Thousands of students currently attending their resides or neighborhood school would be eligible for new school assignments. o 33 percent of elementary students (15,300) o 43 percent of middle school students (9,100) o 36 percent of high school students (8,900) Younger siblings of current students would have no guarantee of placement in the same school. Added Costs Because children are not evenly divided throughout the county, the plan would create overcrowding in some schools and vacancies in others. Community expectations that their children attend the closest school would place significant pressure on the district to build five new elementary schools and one new high school, or transport thousands of students to the next closet school with vacant seats. The cost to taxpayers to build and furnish six new schools would exceed $200 million. The plan, which would allow students to retain their current school assignments, would require the district to operate a more complex, multi-tiered transportation system in order to transport the estimated 7,500 students who would be grandfathered into their current schools (over five years) and all new students assigned to their closest school. Additional transportation costs would be approximately $22 million. Severely Restrict Choice Popular districtwide magnet schools and programs such as traditional schools and Louisville Male High School would very likely be available only to the students living closest to them. In summary, the impact of passing the Williams-Seum bill would: Disrupt families Change every schools boundaries and make 33,300 students eligible for new school assignments. Add costs Potentially cost more than $200 million for 6 new schools. Eliminate choice Access to popular JCPS magnet schools like Louisville Male would be severely limited for students who dont live within the schools boundaries.

Page 12

Sheldon H. Berman, Ed.D. Superintendent

The Impact of BR 111 on Parental Choice in Jefferson County Public Schools


BR 111, AN ACT relating to school attendance.

Revised January 3, 2011

Department of Accountability, Research and Planning Dr. Robert J. Rodosky, Executive Director

Page 13

BR 111, AN ACT relating to school attendance. (2011 Regular Session) Sponsors: Senator Dan Seum, Senator David L. Williams, Senator Elizabeth Tori Bill Summary The bill would permit a parent or legal guardian to enroll a child for attendance at the school nearest to the child's home, except in cases in which there are academic or skill prerequisites for attendance at the school. If the number of children living in a schools attendance area were to exceed the capacity of a school, children residing the shortest distance from the school would be given first priority in assignment to the school. If a child were to be denied attendance at the school closest to his or her home due to the school having reached enrollment capacity, the child shall be given priority for assignment to the next closest school. A child currently attending a school could not be displaced to permit the attendance of another child. A parent or legal guardian would be permitted to enroll a child at a school other than the closest school if the child meets any academic or skill prerequisites for enrollment and there is attendance capacity, with the permission of the district.

Page 14

Impact of BR 111 on Jefferson County Public Schools


Would Significantly Reduce Parental Choice The bill would severely impact Jefferson County Public Schools (JCPS) robust parental choice plan, which includes traditional schools, career-themed high schools, and magnet schools and programs. This would significantly reduce educational options for thousands or tens of thousands of families in Louisville. Would Eliminate the Current Plan that Offers All Families a Fair School Selection Chance No student assignment plan, including the one proposed in BR 111, can provide all families with their first school choice. The bill would supplant the current JCPS student assignment plan, which offers every family a fair chance at selecting a school regardless of where they live. Would Resegregate Public Schools The bill would resegregate our schools by race and income, negatively impacting equitable educational opportunities and student achievement across the district. It would be a profound departure from a long-standing community commitment to diversity, with lasting and detrimental consequences for the future of Louisville Metro. New District Boundaries Would Prohibit Many from Their Current Resides School The bill would eliminate every existing school boundary in Jefferson County. With an entirely new set of boundaries, many families would not be in the attendance area for their current resides school, because other students live closer to that school, or they live closer to a different school. Siblings of a current student who is attending a school other than the school nearest their home could not be given priority over other students who live closer to the school. Complicated Student Assignment Plan Would Create Massive Disruption Although the concept in BR 111 seems very simple, the bill actually would create an even more complicated student assignment plan than the current one, adding confusion and additional transportation costs. Many students could not assigned to the school nearest to their home, because many schools would exceed their capacity. This would require a significant number of students to be offered attendance at the school next closest to their homes, if space were available. This complicated musical chairs scenario of students being bumped from school to school would create a disruptive, unworkable plan for the district and its families. Would Eliminate Local Control of Student Assignment Student assignment has always been the responsibility of local school districts, and for good reason. Each district is different, and local decision makers have the clearest understanding of how to best meet the needs of families in their districts. A one size fits all mandate dictated from Frankfort will never meet the needs of individual counties.

Page 15

Would Significantly Reduce Parental Choice The current JCPS student assignment plan is designed to provide the greatest number of families with their choice of schools. Indeed, the choices of a large majority of parents are accommodated under the current plan. For the current school year: o 80 percent of kindergarteners and 87 percent of first graders were assigned to their first or second choice of schools during the enrollment period, excluding magnet or optional choices. o 42 percent of parents of kindergarteners and first-graders chose a school other than their resides school as their first choice. o Of the remaining 58 percent of parents of kindergarteners and first-graders who chose their resides school as their first choice, 86 percent were assigned to their resides school. While not all students are assigned to their first or second choice of schools, many parents of those students ultimately express a high degree of satisfaction with the educational opportunities provided to their children. Of the many magnet schools and programs offered by JCPS, only Manual, Central and YPAS have admissions standards based on academic or skill prerequisites for attendance. Access to all other magnet schools and programs would be trumped by a formula based on proximity to a school. The defacto dismantling the system of the magnet schools and programs would result in significantly reduced choices for families and diminished opportunities for students to pursue their educational and career interests. This will be detrimental to individual students and the entire community. Would Eliminate the Current Plan that Offers All Families a Fair School Selection Chance No student assignment plan, including the one proposed in BR 111, can provide all families with their first school choice. The bill would supplant the current JCPS student assignment plan, which offers every family a fair chance at selecting a school regardless of where they live. JCPS is striving to provide a system of excellent schools throughout the district so that parents will have greater choice and be satisfied with their child education no matter where their child is assigned. The current student assignment plan works to balance the preferences of parents with the goal of ensuring diversity and equitable educational opportunities across the district.

Page 16

Would Resegregate Public Schools Housing patterns in Louisville Metro are such that BR 111 would, to a significant degree, resegregate JCPS public schools by race and socioeconomic status. JCPS has conducted an analysis of the distribution of elementary school students by race and income if students were assigned to the school closest to their homes. For the 89 elementary schools in the district, the analysis found that: o While the district-wide elementary student population is 34 percent AfricanAmerican: 10 schools would be 90 percent or more African-American. 17 schools would be 90 percent or more White/Other. o While the district-wide elementary student population has 60 percent of students eligible for free or reduced-price lunch (F/R): 29 schools would be 80 percent F/R or greater. 11 schools would be 30 percent F/R or less. The severe concentration of students by race and socioeconomic status would create a host of new challenges for many schools. The decimation of the magnet school structure would reduce educational opportunity for all students, particularly minority and low-income students. The educational and social experiences of all students would be diminished if schools were allowed to become resegregated, having a profound impact on the future of our city. New District Boundaries Would Prohibit Many from Their Current Resides School The bill would eliminate every existing school boundary in Jefferson County. With an entirely new set of boundaries, many families would not be in the attendance area for their current resides school, because other students live closer to the school, or they live closer to a different school. Students entering the school system who are siblings of current students at a school other than the school closest to their home could not be given priority over other students who live closer to the school.

Page 17

Complicated Student Assignment Plan Would Create Massive Disruption BR 111 would create an even more complicated student assignment plan than the current one, adding confusion and additional transportation costs. JCPS conducted an analysis of the impact on student capacity of elementary schools if students were assigned to the school closest to their home. The analysis found that: o 30,500 of the 45,747 elementary school students in the district currently attend a school other than the school closest to their home. Each of those 30,500 students would be given priority to attend the school closest to their home, except that students would be allowed to continue attending their current school if they chose. o 32 of 89 schools would be over capacity by a total of 5,494 students if all students were assigned to the school closest to their home. Ten schools would be more than 200 students over capacity. If all students were assigned to the school nearest to their home, many schools would exceed their capacity. This would require a significant number of students to be offered attendance at the school next closest to their homes, if space were available. If those next closest schools were also at capacity, students would be offered the subsequent next closest school. This would result in a confusing, complicated musical chairs scenario with students being bumped from school to school until all slots are filled. The current student assignment plan allows a student attending a school based on the previous student assignment plan to continue attending that school. The proposed legislation would add complexity by requiring a second layer of grandfathered student assignments based on the current plan. This would require additional transportation costs, and potentially, extended travel times for some students. The combination of all of these factors would create a complicated, confusing, expensive, and disruptive plan for the district and its families. Would Eliminate Local Control Over Student Assignment Student assignment has always been the responsibility of local school districts, and should remain so. Each district is different, and each is confronted with a unique set of issues and challenges. Local decision makers have the clearest understanding of how to effectively and efficiently meet the educational needs of students in their districts. A one size fits all mandate dictated from Frankfort will never meet the needs of individual counties and their citizens.

Page 18

Degree of Public School Segregation in Greater Louisville Project (GLP) Cities and 2009 NAEP Trial Urban District Assessment (TUDA) Districts
Percent of Schools 80%+ Minority or 80%+ White

School District

City

County

State

Students Enrolled in District 2007

Greater Louisville Project Districts Memphis City School District Charlotte Mecklenburg Schools Columbus City Guilford County Schools Duval Davidson County School District Omaha Public Schools Wake County Schools Jefferson County Public Schools

Memphis Charlotte Columbus Greensboro Jacksonville Nashville Omaha Raleigh Louisville

Shelby Mecklenburg Franklin Guilford Duval Davidson Douglas Wake Jefferson

TN NC OH NC FL TN NE NC KY

115,342 131,176 55,269 72,389 124,740 73,715 47,763 134,401 95,871

89.6% 55.6% 53.0% 41.5% 33.3% 31.1% 26.1% 10.5% 4.7%

TUDA Districts Detroit City School Disitrct District of Columbia Public Schools City of Chicago School Distric t 299 Houston Independent School District Miami Dade Baltimore City Public Schools Los Angeles Unified New York City Public Schools Boston Philadelphia City School District Atlanta Public Schools Milwaukee School District Fresno Unified Cleveland Municipal City Austin Independent School District Charlotte Mecklenburg Schools San Diego Unified Jefferson County Public Schools

Detroit Washington Chicago Houston Miami Baltimore Los Angeles New York Boston Philadelphia Atlanta Milwaukee Fresno Cleveland Austin Charlotte San Diego Louisville

Wayne D.C. Cook Harris Dade Baltimore Los Angeles New York Suffolk Philadelphia Fulton Milwaukee Fresno Cuyahoga Travis Mecklenburg San Diego Jefferson

MI DC IL TX FL MD CA NY MA PA GA W CA OH TX NC CA KY

107,874 58,191 407,510 199,534 348,128 81,284 693,680 989,941 56,168 172,704 49,991 86,819 76,460 52,954 82,564 131,176 131,577 95,871

96.9% 90.0% 86.8% 86.7% 86.2% 85.6% 82.9% 79.9% 79.7% 79.6% 79.4% 75.1% 71.4% 69.8% 64.7% 55.6% 47.7% 4.7%

Notes: 1. Data for the following Greater Louisville Project comparison cities was not available: Birmingham, Cincinnati, Dayton, Indianapolis, Kansas City, and Richmond. 2. "Minority" includes Black, Hispanic, Asian/Pacific Islander, and Native American. Sources: 1. Characteristics of the 100 Largest Public School Districts in the US: 2007-2008. NCES, USDOE. 2. The Nation's Report Card: Trial Urban District Assessment Reading 2009. NCES, USDOE. 3. The Nation's Report Card: Trial Urban District Assessment Math 2009. NCES, USDOE. 4. Digest of Education Statistics: 2009 Tables and Figures. Table 92. Enrollment, poverty, and federal funds for the 100 largest school districts. NCES, USDOE.

JCPS. JGL. 9-27-10.


Page 19

Attachment B

Administrative Offices ____________________________________________________________________________________


VanHoose Education Center P.O. Box 34020 Louisville, KY 40232-4020 (502) 485-3011

October 6, 2010

Jewish Community Relations Council c/o Matt Goldberg Executive Director Louisville, Kentucky Dear Faith-based Community Leader: You are undoubtedly aware of the controversy that has arisen in recent weeks surrounding the Jefferson County Public Schools (JCPS) student assignment plan. At this point, there is really no educational issue of greater import to Louisville Metro than how we sustain diversity in our schools. I appreciate being invited to meet with you at noon on October 28 to engage in a dialogue about how we can best set a course that builds a consensus within the community. To set the stage for that conversation, I would like to offer a few thoughts and also share several documents that help to frame the issue. I am confident that each of you is familiar with our school districtsand our communitysstoried efforts to maintain diverse schools. Our presentation to the U.S. Supreme Court in 2007 was supported by numerous briefs from educators, social scientists, psychologists, and business leaders who cited the advantages of a diverse learning environment. Those advantages include: Prepares students to be effective citizens in our pluralistic society Furthers social cohesion and reinforces democratic values Promotes cross-racial understanding Reduces prejudicial stereotypes Enhances life opportunities for students of all races Better prepares the workforce for a global economy Creates more willingness to live in diverse neighborhoods For thirty-five years, our school district has defended its commitment to school diversity against one legal challenge after another, even as other communities abandoned the struggle. We recently completed an analysis (enclosed) of more than two dozen districts that serve urban areas and have large student populations. This study reveals that Louisville has the smallest proportion of highly segregated schools in the nation. Where fewer than 5 percent of JCPS students attend schools whose student populations are either more than eighty percent minority or more than eighty percent white, in most of the comparison cities that figure exceeds 75 percent. I believe that the circumstances we now face are very different from those of past years. The school district is in only the second year of implementing a student assignment plan that had to change as a result of the Supreme Court ruling. A multi-dimensional plan required different configurations of school pairings to achieve diversity, particularly with the inclusion of socio-economic status. I know that you realize how challenging it was to develop a plan to sustain diversity while meeting the Courts guidance. Countless hours were spent on this effort, including consultation with national experts as well as many local individuals and groups. As outlined in the enclosed document No Retreat, new magnet schoolswith their specialized learning opportunitieswere added as a key strategy for encouraging families to send their children beyond their resides school. We have had a successful launch of these magnets, yet it takes several years for a magnet to become established and attractive to parents who may live at some distance from the school. For example, Brandeis Elementary at 2817 W. Kentucky St. is a

Page 20

Jewish Community Relations Council October 6, 2010

math/science magnet, while Coleridge-Taylor Elementary at 1115 W. Chestnut St. and Kennedy Elementary in the Park-DuValle neighborhood both offer the Montessori program. Year after year, these schools are filled nearly to capacity, and each draws students from more than two dozen zip codes. However, it took time for these magnets to build both their programs and their strong reputations for high-quality education. This year, we began phasing in the visual and performing arts program at Western Middle School at 2201 W. Main St. Of the 206 sixth-grade applicants, nearly 80 percent were students from outside Westerns former resides-area boundary. As a tool for promoting diversity through choice, magnets unquestionably work; however, parents need time to become familiar with them and to develop confidence in their programs. Although magnet schools and reliance on choice provide effective ways to enhance diversity, they cannot sustain diversity without some assignment of students so that schools meet the districts diversity guideline of no less than 15 percent and no greater than 50 percent of students coming from A areasthat is, areas with a higher percentage of minority and low-income residents and lower levels of adult educational attainment. It is for that reason that the district has always described the student assignment plan as a managed choice plan. I have enclosed two maps to help you visualize our elementary plan. One map shows which areas are identified as A areas. The other shows the arrangement of the six elementary clusters within which parents are asked to list four choices of schools they wish their child to attend, two from area A and two from area B. Over the past two years, the new student assignment plan has demonstrated success in enhancing the diversity of our elementary schools. As of mid-September, 59 percent of our elementary schools were within the new plans 15-50 guideline for first and second grades (which is an increase of ten percent over last year), and another 34 percent were moving toward the diversity guideline. Therefore, two years into the plan, 93 percent of our schools are making progress in achieving the new multifactor diversity guidelines, and we can say with confidence that the plan is working. In recent months, the combined impact of the Supreme Court ruling, the media coverage surrounding Wake Countys (NC) abandonment of voluntary school desegregation, and the repeated legal challenges to the JCPS student assignment plan has communicated to the public that our plan is voluntaryrather than a response to comply with a court order. As a consequence, many people perceive this moment as an opportune time to advocate for neighborhood schools. Although our transportation system is now more efficient than it has ever been, with shorter average ride times and higher levels of service than are provided in most of the urban districts around the country or even in districts throughout Kentucky, transportation has become a central issue in the debate over student assignment, serving as one rationale for neighborhood schools. Some of you may be wondering about the bill pre-filed by Senators Williams, Seum, and Tori, and its implications for our student assignment plan, magnet schools, and district stability. According to our internal analysis (enclosed for your information), the bill would resegregate our schools by race and income, negatively impacting equitable educational opportunities and student achievement across the district. I believe it would constitute a profound departure from a long-standing community commitment to diversity, with lasting and detrimental consequences for the future of Louisville Metro. It is an impractical plan in which many parents would not be able to have their children attend their neighborhood school due to capacity issues at that school. In addition, it would have significant adverse consequences for parental choice and our magnet school options. I have enclosed two editorials from The Courier-Journal (August 29, 2010) and the Lexington Herald-Leader (September 1, 2010). Both editorials make it clear that there is strong support in multiple sections of the state for diversity in student populations and for local control of educational decisions. The JCPS administration and Board are working very hard to ensure that our plan and our programs offer a highquality education to students in every corner of the district. With the communitys support and involvement, JCPS has created a unique balance among students who attend schools in their neighborhood, attend magnet schools of choice, attend one of their preferred schools within their cluster, or are assigned to schools other than their parents first or second choice. This year, 86.7 percent of first-graders parents were able to receive their first- or secondchoice school. It is interesting to note that 42 percent of parents of kindergartners and first graders requested that their child attend a school other than their resides school. Other important details about the student assignment plan can be found on the enclosed fact sheet. At this point, the options that exist for modifying this plan, short of totally dismantling it, are limited. I trust you recognize the importance of diverse schools to the future of our students and our community and, therefore, appreciate the dilemma we now face. The March 16, 2010, editorial from The

Page 21

Jewish Community Relations Council October 6, 2010

Courier-Journal, plus the September 20, 2010, article from The Boston Globe (both enclosed) leave no doubt as to how high the stakes are. It is clear to me that the controversy over busing will not disappear, no matter which path the district follows. Busing is an emotion-laden issue and one that people take very personally. The key to supporting Louisville Metros social and economic progress is to navigate through the issue in a way that demonstrates the leadership that the community needs in order to achieve our educational and diversity goals. As faith-based leaders, you clearly understand that leadership doesnt mean doing whats easy or whats popular. It means doing what is right for children and for the future of our community over the long term. In reality, we are at a critical point in the community dialogue about how and whether to place a priority on diversity in our schools. I will continue to work with community leaders to frame the issue and, hopefully, reaffirm the merit of diversity as a fundamental community value. Because this topic has become laden with political ramifications, it is even more critical that leaders such as you, who do not have to be concerned with ballot boxes, be willing to help us find and support the right path through this issue on behalf of our communitys future. In the end, it truly is a community decision, a community choice. There are strong proponents speaking in favor of neighborhood schools, and some of the arguments being put forward can, indeed, be compellingbut the consequences for our community, both immediate and long-term, would be grave. So where do we go from here? On behalf of our school district, I am seeking more than just your counsel; I urgently need your active engagement. We have a common interest in preserving school diversity and the benefits it yields for our residents, our workforce, and our attractiveness as a city. The challenge is: how do we defuse the tension around this issue and help the community at large, individually and collectively, understand that diversity is a productive choice for themselves and their families? Again, I appreciate the deep thought you are giving to this vital issue. Thank you for inviting me to meet with you, and I welcome any questions or feedback you would like to give me in the coming weeks as we continue to grapple with this important matter. Sincerely,

Sheldon H. Berman, Ed.D. Superintendent SHB:jb Enclosures

Page 22

Page 23

Page 24

Page 25

Page 26

Attachment C

1.10.11 DRAFT

Options for Adjustments to the Student Assignment Plan


1. Establish pre-school/kindergarten centers which would be separate from students in 1 through 5 grade. These centers could be located in (a) existing school buildings, or (b) in secondary buildings, or (c) in new construction. Benefits Drawbacks Timeline for Implementation Does it address concerns? Centers could be strategically located in These programs must be located on ground level, Could be implemented by 2011-12, but Addresses needs of P1 students clusters to minimize the distance traveled limiting the number of existing facilities which could for optimal implementation should be P1 students would have a greater possibility of for some students. At least two centers be used. delayed until 2012-13. being near the home, depending on the proximity would be required in each of the current of the pre-school/kindergarten (P1) center to the clusters. Lack of educational continuity between kindergarten home. (P1) and first grade (P2). Length of ride P1 students would not expect to stay in There would be shorter bus rides for some st th same building for 1 -5 grade. P1 students would not be in the same building as students and longer bus rides for other students, older elementary-aged siblings. depending on the proximity of the preschool/kindergarten (P1) center to the home. Many P1 students would have longer bus rides Provides for diversity because most would not be attending the Would not provide for geographic diversity neighborhood school. Provides for choice Parents could continue to choose magnet Transportation would have to cover routes four, schools. Magnet programs could not be st possibly five timesonce for pre-school, once for P1, implemented until the P2 (1 grade year). Not once for local routes, once for cluster exchange, once sure if parents could choose a center out of their for grandfathering which might require increased local area or cluster. transportation costs. Predictability in placement Students would be assigned to one of the centers The funding source determines if the pre-school based on the home address. programs are full or half-day and the number of days students are in session. If all students in the preschool/kindergarten center were on the same schedule, an increase in the number of instructional hours or days would require additional funds.
st th

Page 27

1.10.11 DRAFT
1a. Converting some existing elementary buildings to pre-school/kindergarten (P1) centers. Benefits Drawbacks Can use existing space and would not If current buildings were re-purposed, it would require construction of new facilities. displace current students and require redistricting of current elementary school boundaries which would be a politically charged issue. Timeline for Implementation Could possibly be implemented by 2011-12, but for optimal implementation should be delayed until 2012-13. Does it address concerns?

1b. Pre-school and Kindergarten (P1) centers located in secondary buildings Benefits Drawbacks Space is available in several secondary Extensive renovation of facilities would be required buildings. to eliminate the possibility of contact with older students. Would not require elementary students to be displaced or boundaries redrawn. Extensive renovation of restrooms, cafeteria, outdoor space and classrooms would be required to accommodate small children

Timeline for Implementation Would be difficult to implement by 201112 because of required building renovations.

Does it address concerns?

1c. New construction of pre-school and Kindergarten (P1) centers. Benefits Drawbacks Currently there is an issue of limited space Would need at least 2 additional buildings in each for pre-school programs in some cluster which would require construction funds. elementary buildings; centers located on General funds would have to be used for the other property may help relieve some construction of these facilities. overcrowding. Would not require elementary students to be displaced or boundaries redrawn.

Timeline for Implementation Would take at least 2 to 3 years to build the centers

Does it address concerns?

Page 28

1.10.11 DRAFT
3. Sub-clusters within the current contiguous clusters. Benefits Drawbacks Could possibly have shorter bus rides for Sub-clusters would not result in a savings of some students. transportation time. Transportation implemented express routes and revamped the depots for the Could continue to provide the same 2010-11 school year, which decreased ride time for number of choicestwo in Area A and the great majority of students. For example, in two in Area B. Cluster 4, depots have been designed to serve fewer buses which decreases the time buses wait at the depot and reduces the time for the students to Would simplify the choices for parents change buses. would choose from 6-8 schools rather than 12-15. Transportation will still continue to have routes among most schools in the cluster because of the magnet opportunities available to students. Would reduce parental choice options from 12-15 schools to 6-8. Would require another layer of grandfathering, thereby not produce any savings in transportation time or expense and would create greater complexity in the operations until all grandfathered students moved into middle school. Timeline for Implementation Could be implemented by 2011-12, but for optimal implementation should be delayed until 2012-13 Does it address concerns? Addresses needs of P1 students Length of ride For some students there could be shorter bus rides Provides for diversity The sub-clusters would make it possible for there to be geographic diversity Provides for choice It would narrow choice. However, parents could continue to choose magnet schools, magnet programs, optional programs or other schools in the cluster, if there was capacity. Predictability in placement

Page 29

1.10.11 DRAFT
4. Review current cluster configuration to determine if there are schools that could be moved from one cluster to another. Benefits Drawbacks Timeline for Implementation There may be reduced travel distances for May not be able to achieve the desired diversity in Could be implemented by 2011-12, but some students because the average ride the school buildings for optimal implementation should be time may decrease. delayed until 2012-13 Any changes or modifications might require another Continuing with the clusters allows for layer of transportation to provide educational choice and access to a variety of programs continuity (grandfathering) for students assigned and opportunities under the current plan. Disrupts current choices and cluster assignments for those assigned to new cluster(s) thereby creating more confusion for parents. Does it address concerns? Addresses needs of P1 students Not necessarily Length of ride Possibly Provides for diversity Could provide for geographic diversity Provides for choice Parents could continue to choose magnet schools, magnet programs, optional programs or other schools in the cluster, if there was capacity. Predictability in placement

5. Review cluster plan from 2007-08 to determine if it is feasible to return to that cluster configuration. Benefits Drawbacks Timeline for Implementation Fewer schools in a cluster may reduce Some clusters would be greater that 50% A, requiring Could be implemented by 2011-12, but transportation time or distance some changes in the assignment of schools to for optimal implementation should be particular clusters. delayed until 2012-13 In a non-contiguous plan, it is easier to th th modify the clusters to help with diversity, Only 4 & 5 grade students in 2011-12 are assigned capacity and other demographic trends. under the previous cluster configuration. Students nd rd who will be in the 2 & 3 grade in 2011-12 have been assigned under the geographic plan. Students st in kindergarten and 1 grade in 2011-12, may possibly be assigned under a different plan. Need to determine if we can implement several different plans at the elementary level at one time. There are fewer choices in the cluster. There is a greater differential in the percentage of Area A and Area B among the clusters with some clusters having high percentages of Area A students and some having low percentages of Area A students. Some civil rights and fair housing advocates are concerned that this plan would not address socioeconomic issues.

Does it address concerns? Addresses needs of P1 students Not necessarily Length of ride Possibly Provides for diversity Some clusters have >50% Area A Provides for choice Parents could continue to choose magnet schools, magnet programs, optional programs or other schools in the cluster, if there was capacity. Predictability in placement

Page 30

1.10.11 DRAFT
6. Review definitions of Area A and Area B. The Core Team studied other ways to examine geographic data to determine if it is possible to capture smaller areas of diversity within larger geographic areas. If the variables of income and minority were used without the variable of educational attainment, small pockets of diversity were found within block group areas that are currently all Area A or Area B. The following is an analysis of the 2-variable plan with the inclusion of Area A blocks within Area B schools and Area B blocks with Area A schools. Benefits Drawbacks Timeline for Implementation Does it address concerns? Would allow more students to remain in The plan could be so complex it would be difficult to Could be implemented by 2011-12, but Addresses needs of P1 students the resides school and fewer students may articulate to the community. for optimal implementation should be Would require fewer students to be involved in be involved in the exchange. delayed until 2012-13 the geographic exchange. The data may be at such a discrete level that an Length of ride Recognizes the diversity that exists in individual student or family could be identified. There would be more students who had the some neighborhoods opportunity to attend the resides school, The data for census track and block is 10 years old resulting in shorter rides for some students. If Area A were redefined using the and may not reflect the current diversity of track and Other students may still have long ride times. variables of race and income: block. (However, 2010 census data might be released Provides for diversity soon enough to be used for 2012-13.) Recognizes diversity that exists in some 44 of the elementary schools neighborhoods. have a mix of Area A and Area B This proposal may support diversity in selected Provides for choice in their current boundary, schools but may not impact all schools. 38 Parents could continue to have choice for thereby offsetting some of the elementary schools remain at 100% Area A or Area B magnet schools, magnet programs, and within need to move students either in in the resides area. the cluster. or out of the school. Predictability in placement 16 of these 44 elementary This does not fully address the issue of ride times schools (Gutermuth, Kerrick, between schools which are farthest apart Hazelwood, Semple, Gilmore Lane, Okolona, Fern Creek, Luhr, geographically in the clusters, although it may decrease the number of students involved in those Watterson, Cochrane, Klondike, exchanges. St. Matthews, BreckinridgeFranklin, Chenoweth, Field, Would require changes to middle and high school Wilder) would be between 15boundaries 50% Area A in the resides populations. Would reduce the number of students being transported and potentially reduce ride times or cross-county exchanges. Gives a more realistic assessment of the diversity in the schools and provides a more balanced mix of students from Area A and Area B.

Page 31

1.10.11 DRAFT

Student Assignment Plan Decision Making Points


1. Includes students in the ESL program as part of the Area A/Area B diversity count in a school building. Benefits Drawbacks Timeline for Implementation Would recognize the existing diversity in a Would need to add a number of programs to have Could be implemented by 2011-12, but school/neighborhood area. centers at more schools. for optimal implementation should be delayed until 2012-13 Provides a more realistic picture of the Adding ESL units could require adjustments at resides area of a school schools which are near capacity. Would allow more students to remain in the resides school. Can strategically place programs to support balance among schools. Does it address concerns? Addresses needs of P1 students Not necessarily Length of ride Possibly Provides for diversity Yes. Provides for choice Yes. Parents could continue to have choice for magnet schools, magnet programs, and within the cluster. Predictability in placement

Page 32

1.10.11 DRAFT
2a. Excluding Kindergarten (P1) students from the exchangethe current student assignment plan does not include kindergarten (P1) students in the diversity guideline; however, in some instances, kindergarten (P1) students are currently assigned away from the resides school to keep siblings together or because of capacity issues. Benefits Drawbacks Timeline for Implementation Does it address concerns? Shorter travel distances for many Would cause an additional transition for many Could be implemented by 2011-12 with Addresses needs of P1 students st students. students between P1 and P2 (Kindergarten and 1 proper notification of parents. P1 students would be assigned to the resides grade) location, limiting choice. Higher comfort level for many parents. Length of ride There may be issues with capacity at some schools if In general, there would be shorter bus rides for all the kindergarten (P1) students were assigned to P1 students the location that serves the home address. Provides for diversity Would not provide for geographic diversity at Would not provide continuity between P1 and P2 the P1 level st (Kindergarten and 1 grade) Provides for choice Parents could continue to choose magnet Elementary-age siblings could be assigned to schools, magnet programs, optional programs different schools or other schools in the cluster, if there was capacity. Elementary principals expressed the desire to have Predictability in placement th all students Kindergarten (P1) through 5 grade in Many students would have to change the plan to provide educational continuity and build assignments at the end of the kindergarten relationships. (P1) year. Some elementary schools would be over capacity if all Kindergarten (P1) students remained at the resides location.

Page 33

Attachment D

JEFFERSON COUNTY PUBLIC SCHOOLS District Projected AB By Min/MHHI 2011 (DRAFT 10/6/2010)

This map represents Area A and B based on both income and minority status. Area A is defined as below the district average in median household income (census 2000) AND above the district average in % minority (2011 JCPS student file)
NORTON CHANCEY

ZACHARY TAYLOR DUNN PORTLAND CHENOWETH BOWEN WILDER

TEL EV Y O S RR RO PE

ATKINSON

KING BYCK COLERIDGETAYLOR

BRECKINRIDGE/ FRANKLIN FIELD MIDDLETOWN

STOPHER

FOSTER WHEATLEY MAUPIN KENNEDY MCFERRAN

ENGELHARD ST MATTHEWS BLOOM LOWE

SHELBY

HITE

COCHRAN

Areas
CRUMS LANE CANE RUN MILL CREEK FRAYSER

HAWTHORNE KLONDIKE COCHRANE

No Data

CAMP TAYLOR

GOLDSMITH

A B

WELLINGTON

JACOB

SEMPLE WATTERSON TULLY JEFFERSONTOWN

GUTERMUTH

GILMORE LANE

PRICE

KERRICK HAZELWOOD SHACKLETTE

RUTHERFORD INDIAN TRAIL FERN CREEK WHEELER RANGELAND SLAUGHTER

LUHR HARTSTERN FARMER

SANDERS

GREENWOOD

TRUNNELL

KENWOOD EISENHOWER

AUBURNDALE

MINORS LANE

OKOLONA

SMYRNA

WILKERSON JOHNSONTOWN LAYNE DIXIE

BLAKE BATES

WILT

FAIRDALE STONESTREET

CORAL RIDGE

BLUE LICK

LAUKHUF MEDORA

WATSON LANE

The boundaries displayed here are approximations and should be used only as a general reference.

Copyright (c) 2010, LOUISVILLE AND JEFFERSON COUNTY METROPOLITAN SEWER DISTRICT (MSD), LOUISVILLE WATER COMPANY (LWC), LOUISVILLE METRO GOVERNMENT, and JEFFERSON COUNTY PROPERTY VALUATION ADMINISTRATOR (PVA). All Rights Reserved.

8 Miles

Page 34

PT:RJR:RS:DD:rmc 10/06/2010 Accountability, Research & Planning

JEFFERSON COUNTY PUBLIC SCHOOLS 1011 Elementary Resides School by Track/Block AB Using District Minority and Income (Excludes ECE-SC/K) Resides School CANE RUN CRUMS LANE DIXIE EISENHOWER FOSTER GREENWOOD GUTERMUTH JOHNSONTOWN RD KENNEDY KERRICK SANDERS SHACKLETTE WATSON LANE WELLINGTON WILKERSON Cluster 1 Total Resides School AUBURNDALE CORAL RIDGE FAIRDALE HAZELWOOD JACOB KENWOOD LAYNE MEDORA MILL CREEK RUTHERFORD SEMPLE STONESTREET TRUNNELL Cluster 2 Total Resides School BLAKE BLUE LICK CAMP TAYLOR COCHRAN FRAYSER GILMORE LANE HARTSTERN LAUKHUF MAUPIN MCFERRAN MINORS LANE OKOLONA SMYRNA WILT Cluster 3 Total A/B A A B B A B B B A B B B B A B Cluster 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 A 329 312 B 28 154 374 298 387 296 274 206 597 377 425 187 293 3896 B 607 292 345 263 45 405 315 282 36 97 398 346 504 3935 B 561 425 378 89 112 165 635 1 40 169 528 355 3458 Total 357 466 374 298 823 387 508 274 237 399 597 378 425 399 293 6215 Total 647 302 345 499 593 464 315 282 487 743 561 346 504 6088 Total 561 425 378 300 562 188 376 635 372 424 93 234 528 355 5431 %A 92.2% 67.0% 0.0% 0.0% 100.0% 0.0% 41.7% 0.0% 100.0% 48.4% 0.0% 0.3% 0.0% 53.1% 0.0% 37.3% %A 6.2% 3.3% 0.0% 47.3% 92.4% 12.7% 0.0% 0.0% 92.6% 86.9% 29.1% 0.0% 0.0% 35.4% %A 0.0% 0.0% 0.0% 70.3% 100.0% 40.4% 56.1% 0.0% 99.7% 90.6% 100.0% 27.8% 0.0% 0.0% 36.3% % FRL 86.3% 83.1% 76.2% 58.7% 91.4% 65.1% 71.1% 61.8% 87.8% 74.9% 70.7% 54.2% 80.2% 76.4% 68.6% 75.2% % FRL 78.5% 82.1% 64.8% 89.6% 93.6% 73.3% 61.4% 57.1% 85.9% 86.8% 86.1% 62.4% 63.6% 78.2% % FRL 73.2% 75.1% 56.6% 79.7% 92.9% 86.7% 62.5% 39.8% 94.9% 95.1% 94.6% 75.2% 48.5% 61.8% 70.9% % Minority 88.5% 71.7% 19.3% 12.4% 97.3% 26.1% 56.1% 12.4% 99.2% 44.4% 30.9% 32.3% 14.8% 56.9% 17.1% 48.9% % Minority 45.1% 39.7% 12.1% 60.9% 69.5% 43.0% 23.1% 10.6% 83.8% 65.9% 40.6% 22.5% 33.7% 46.7% % Minority 33.6% 29.2% 21.2% 59.0% 79.2% 45.7% 51.6% 26.5% 97.8% 92.5% 66.7% 41.0% 34.1% 31.7% 49.1%

823 212 237 193 1 212 2319

A/B B B B A A B B B A A B B B

Cluster 2 2 2 2 2 2 2 2 2 2 2 2 2

A 40 10 236 548 59

451 646 163

2153 A/B B B B B A B B B A A B B B B Cluster 3 3 3 3 3 3 3 3 3 3 3 3 3 3 A

211 562 76 211 371 384 93 65

1973

RJR.dd.rmc
Page 35

12/15/10

JEFFERSON COUNTY PUBLIC SCHOOLS 1011 Elementary Resides School by Track/Block AB Using District Minority and Income (Excludes ECE-SC/K) Resides School BATES FARMER FERN CREEK GOLDSMITH INDIAN TRAIL JEFFERSONTOWN LUHR PRICE RANGELAND SLAUGHTER WATTERSON WHEELER Cluster 4 Total Resides School BLOOM BYCK COCHRANE ENGELHARD HAWTHORNE HITE KLONDIKE LOWE MIDDLETOWN SHELBY ST MATTHEWS STOPHER TULLY WHEATLEY Cluster 5 Total Resides School ATKINSON BOWEN BRECK/FRANKLIN CHANCEY CHENOWETH COLERIDGE-TAYLOR DUNN FIELD KING NORTON PORTLAND ROOSEVELT-PERRY WILDER ZACHARY TAYLOR Cluster 6 Total Grand Total A/B B B B B A B B A A A B B Cluster 4 4 4 4 4 4 4 4 4 4 4 4 A B 583 411 403 112 56 651 320 4 150 465 540 3695 B 377 26 229 3 258 379 528 509 490 183 459 607 544 9 4601 B 190 621 186 928 297 447 152 437 96 58 219 395 4026 23611 Total 583 411 661 647 421 651 511 312 601 443 636 540 6417 Total 377 572 369 328 271 379 724 543 490 612 555 607 544 363 6734 Total 459 621 342 928 397 259 447 236 317 437 274 396 334 454 5901 36786 %A 0.0% 0.0% 39.0% 82.7% 86.7% 0.0% 37.4% 98.7% 100.0% 66.1% 26.9% 0.0% 42.4% %A 0.0% 95.5% 37.9% 99.1% 4.8% 0.0% 27.1% 6.3% 0.0% 70.1% 17.3% 0.0% 0.0% 97.5% 31.7% %A 58.6% 0.0% 45.6% 0.0% 25.2% 100.0% 0.0% 35.6% 100.0% 0.0% 65.0% 85.4% 34.4% 13.0% 31.8% 35.8% % FRL 32.4% 30.2% 65.4% 77.1% 83.8% 36.9% 60.1% 83.3% 91.7% 86.0% 48.6% 40.2% 60.2% % FRL 21.8% 95.1% 62.5% 95.4% 25.5% 28.3% 60.5% 30.3% 31.4% 88.1% 31.1% 8.6% 21.4% 95.9% 49.5% % FRL 94.3% 28.5% 70.5% 34.7% 47.9% 98.8% 11.4% 41.4% 92.7% 7.1% 92.0% 95.2% 39.5% 33.9% 50.9% 63.8% % Minority 23.8% 24.6% 52.2% 64.8% 70.5% 28.7% 52.1% 79.2% 87.0% 79.9% 52.8% 32.2% 52.8% % Minority 15.6% 88.0% 46.4% 84.1% 14.0% 23.9% 48.1% 35.3% 32.2% 67.0% 27.0% 23.4% 23.3% 91.5% 44.5% % Minority 64.1% 33.4% 52.0% 50.2% 28.0% 98.8% 20.8% 27.2% 98.4% 16.9% 67.9% 72.2% 35.0% 41.4% 48.0% 48.3%

258 535 365 191 308 601 293 171 2722

A/B B A B A B B B B B A B B B A

Cluster 5 5 5 5 5 5 5 5 5 5 5 5 5 5

A 546 140 325 13 196 34 429 96

354 2133 A 269 156 100 259 84 317 178 338 115 59 1875 13175

A/B A B B B B A B B A B A A B B

Cluster 6 6 6 6 6 6 6 6 6 6 6 6 6 6

RJR.dd.rmc
Page 36

12/15/10

Attachment E

Middle School Boundaries for 2011-12, with new A/B configuration (and boundary tweaks in yellow) with changes to Moore, and a second round of changes for Ramsey-Newburg, Myers New A/B % 2011-12 Boundary 1 Conway 14% 2 Farnsley 32% 3 Frost 37% 4 Lassiter 38% 5 Olmsted N 62% 6 Olmsted S 47% 53% 7 Stuart 24% 35% 8 9 10 11 12 13 14 Carrithers Knight Moore Newburg Noe Ramsey T. Jefferson 36% 17% 39% 44% 41% 31% 55% 40% 23% 38% 28% 39% 48% 28% 33% New data for proposed changes to 2011-12 Boundaries Resides 990 946 641 724 668 1070 1738 1135 6174 646 617 835 972 1233 894 987 6184 1249 1103 1079 1141 927 1003 6502 #A 266 306 237 282 406 421 827 431 2349 233 194 246 340 518 258 461 2250 332 432 305 441 401 267 2178 %A 27% 32% 37% 39% F/R 762 622 529 605 595 891 1486 930 4934 395 481 589 551 601 506 763 3886 549 597 487 602 694 612 3541 %F/R 77% 66% 83% 84% 89% 83% 86% 82% 80% 61% 78% 71% 57% 49% 57% 77% 63% 44% 54% 45% 53% 75% 61% 54% MIN 478 354 293 310 439 559 998 516 2949 319 259 408 534 543 417 527 3007 537 486 504 587 568 437 3119 %MIN 48% 37% 46% 43% 66% 52% 57% 45% 48% 49% 42% 49% 55% 44% 47% 53% 49% 43% 44% 47% 51% 61% 44% 48% Student Impact 171 43

131 167

48% 38% 38% 36% 31% 29% 35% 42% 29% 47% 36% 27% 39% 28% 39% 43% 27% 33%

49 107

182

15 16 17 18 19 20

Crosby Highland Kammerer Meyzeek Myers Westport

47

18860 6777 36% 12361 66% 9075 48% 897 Note: 80 of the 897 students impacted are returned to their pre-2011-12 resides 1/12/11 Column Header Descriptors New A/B 2011-12 Boundary - The percentage of A geography students (under current A geography) residing in boundaries already adopted for 2011-12 data columns in box use proposed 'A within B' configuration: Resides - the total # of students residing in the boundary adjusted fcr proposed changes #A - the # of geography 'A' students residing in the boundary adjusted for proposed changes %A - the % of geography 'A' students residing in the boundary adjusted for proposed changes #F/R - the # of Free/Reduced Lunch qualifiers residing in boundary adjusted for proposed changes %F/R - the % of Free/Reduced Lunch qualifiers residing in boundary adjusted for proposed changes #MIN - the # of minority students residing in boundary adjusted for proposed changes %MIN - the % minority students residing in boundary adjusted for proposed changes Student Impact - the # of students reassigned under proposed adjustements
Page 37

Potential Changes to Middle School 2011-12 Boundaries


Legend
MID_newAB_Dis
<all other values>
S I LAN D

BOUND_DIFF
KAMMERER
S I LAN D

( ! WESTPORT
R VE R I

PO R LA N T D

CARRITHERS CONWAY FARNSLEY KNIGHT


M U PI N A

( !
AN C O R G E H A

WESTERN
KI N G

S I LAN D

( ! BROWN
R O SE VE LT- P ER R O Y STO P H R E

( ! MEYZEEK ( !
LO WE

CROSBY ( !

MYERS NEWBURG OLMSTED NORTH RAMSEY

M FE R R N C A

NOE
M FE R R N C A

( !

HIGHLAND ( !
H WTH O R E A N

M FE R R N C A

MYERS ( !
JAC O B

STUART

FARNSLEY ( ! ( ! CONWAY ( !

OLMSTED NORTH ( ! OLMSTED SOUTH


R TH ER FO R D U

NEWBURG ( !
PR I E C

CARRITHERS ( !

AI R O R P T

THOMAS JEFFERSON ( !
R N E LAN D A G LU R H

WH EL ER E

RAMSEY ( !
O O LO N K A

MOORE ( !

EI SE N O WE R H

AU BU R N AL E D

LASSITER ( ! ( ! ( ! STUART ( ! ( ! KNIGHT T.T.

FROST ( !

Page 38

High School Data with new A/B configuration New A/B% 2012-12 BOUNDARY 33% 29% 51% 20% 23% 53% With Proposed Boundary Changes Resides 998 1015 1234 1739 1132 990 7108 1432 1624 1251 1687 1646 7640 #A 333 315 587 462 330 487 2514 490 587 491 707 543 2818 %A 33% 31% 48% 27% 29% 49% 35% 34% 36% 39% 42% 33% 37% F/R %F/R 671 67% 740 73% 942 76% 1055 61% 816 72% 719 73% 4943 70% 787 893 768 963 1117 4528 55% 55% 61% 57% 68% 59% MIN %MIN 482 48% 327 32% 659 53% 685 39% 421 37% 584 59% 3158 44% 693 855 589 750 717 3604 48% 53% 47% 44% 44% 47% Student Impact

Doss Fairdale Iroquois PRP Valley Western

65 137 92 77

Fern Creek Jeffersontown Moore Seneca Southern

40% 36% 39% 50% 29%

247 203 103

Atherton Ballard Eastern Acad@Shawnee Waggener

33% 18% 24% 60% 39%

1087 359 33% 471 43% 368 34% 1644 369 22% 545 33% 718 44% 99 2130 435 20% 676 32% 761 36% 753 404 54% 607 81% 405 54% 76 1384 554 40% 791 57% 780 56% 6998 2121 30% 3090 44% 3032 43% 1099 Note: 207 of the 1099 impacted return to pre-2012-13 resides 21746 7453 34% 12561 58% 9794 45%

1/12/11 Column Header Descriptors New A/B 2011-12 Boundary - The percentage of A geography students (under current A geography) residing in boundaries already adopted for 2011-12 data columns in box use proposed 'A within B' configuration: Resides - the total # of students residing in the boundary adjusted fcr proposed changes #A - the # of geography 'A' students residing in the boundary adjusted for proposed changes %A - the % of geography 'A' students residing in the boundary adjusted for proposed changes #F/R - the # of Free/Reduced Lunch qualifiers residing in boundary adjusted for proposed changes %F/R - the % of Free/Reduced Lunch qualifiers residing in boundary adjusted for proposed changes #MIN - the # of minority students residing in boundary adjusted for proposed changes %MIN - the % minority students residing in boundary adjusted for proposed changes Student Impact - the # of students reassigned under proposed adjustements

Page 39

Legend
H_BOUN_CHANGE_Dis
<all other values>

Potential Changes to High School 2012-13 Boundaries

S I LAN D

Boun_Chnge
BALLARD DOSS FAIRDALE
KI N G PO R LA N T D

S I LAN D

BALLARD !
R VE R I

SHAWNEE MAGNET CAREER ACADEMY !


S I LAN D R O SE VE LT- P ER R O Y

AN C O R G E H A

IROQUOIS JEFFERSONTOWN PRP SENECA SHAWNEE SOUTHERN VALLEY WESTERN


!
JAC O B M U PI N A M FE R R N C A

CENTRAL MAGNET CAREER ACADEMY !

STO P H R E

WAGGENER MAGNET CAREER ACADEMY !


LO WE

EASTERN !

M FE R R N C A

DUPONT MANUAL ! ATHERTON !


H WTH O R E A N

M FE R R N C A

BUTLER TRADITIONAL

SENECA MAGNET CAREER ACADEMY ! JEFFERSONTOWN !

LOUISVILLE MALE TRADITIONAL ! WESTERN MST MAGNET ! IROQUOIS MAGNET CAREER ACADEMY !
PR I E C

AI R O R P T

R TH ER FO R D U

WH EL ER E

R N E LAN D A G

DOSS MAGNET CAREER ACADEMY ! PLEASURE RIDGE PARK MCA !


O O LO N K A EI SE N O WE R H AU BU R N AL E D

LU R H

FERN CREEK MAGNET CAREER ACADEMY ! MOORE HIGH SCHOOL MAGNET CAREER ACADEMY !

SOUTHERN MAGNET CAREER ACADEMY ! VALLEY MAGNET CAREER ACADEMY FAIRDALE MAGNET CAREER ACADEMY !

Page 40

JCPS Distances for Elementary Schools Original Non-Contiguous Plan (2008-09) School 1 School 2 Stopher Laukhuf Eisenhower Chenoweth Bowen Minors Lane Hite Byck Maupin Coral Ridge Tully Coleridge-Taylor Gilmore Lane Dixie King Johnsontown Road Cane Run Watson Lane Portland Sanders Kennedy Stonestreet Farmer Klondike

Attachment F

Miles 19.83 19.67 16.62 16.52 16.36 15.23 13.89 12.38 12.24 11.27 11.27 9.35

Cluster 10 12 11 1 7 3 8 6 2 4 5 9

Revised Non-Contiguous Plan School 1 School 2 Stopher Atkinson Foster J-Town Portland Hite Maupin Coral Ridge Johnsontown Camp Taylor Watson Lane Cane Run Farmer Indian Tr Laukhuf Fern Creek

Miles 21.55 19.86 17.53 16.36 16.02 12.23 11.11 8.26

Cluster 3 7 1 5 4 2 8 6

Contiguous Plan School 1 Stopher Wilt Chancey Coral Ridge Watson Lane Farmer

School 2 Shelby Maupin King Mill Creek Foster Indian Trail

Miles 21.2 20.78 20.25 15.52 15.2 10.44

Cluster 5 3 6 2 1 4

RJR.dd.dv
Page 41

9/10/10

You might also like