UNITED STATES DISTRICT COURT DISTRICT OF CONNECTICUT LASHAWN ROBINSON et al., CIVIL ACTION NO. (SRU) Plaintiffs, v. DIANNA WENTZELL, in her of?cial . Capacity as Commissioner of Education, APRIL 27, 2018 Connecticut State Department of Education, et al., Defendants. AFFIDAVIT OF GLEN PETERSON Glen Peterson, being duly sworn, does hereby depose and say: 1. I am over the age of eighteen (18) and I understand and appreciate the obligations of an oath. 2. I am an Education Division Director for the Connecticut State Department of Education CSDE hired me in September of 2013 and I have held the same title for the duration of my employment at CSDE. 3. Prior to my employment with CSDE, I worked as an educator, including ten years as principal of the Glastonbury-East Hartford Magnet School in Glastonbury, Connecticut from 2003 through 2013. GEHMS is a magnet school operated by the Capitol Region Education Council Prior to my employment with CREC, I worked as a principal in the Winchester School District and as a teacher in Concord, Massachusetts. 4. In my current position as Education Division Director at CSDE, I oversee the Sheff Of?ce, Regional School Choice Of?ce School Choice Office, and various other CSDE programs. The ?rst two of?ces the Sheff Of?ce and RSCO were created by the CSDE pursuant to the Phase II Stipulation and Order (?Phase II Stipulation?), dated April 8, 2008, in the .5716ij v. ?Neill, 238 Conn. 1 (1996), desegregation case (?Sheff?). The Phase II Stipulation was approved by the General Assembly and became and an Order of the Court on June 11, 2008. The Phase II Stipulation sets forth the responsibilities of the Sheff Of?ce and RSCO. . Prior to expiration of the Phase II Stipulation on June 30, 2013, the parties extended the Agreement for an additional year through June 30, 2014, by Stipulation and Order, dated April 30, 2013. On December 13, 2013, the parties executed a one year Phase Stipulation, as approved by the Court, through June 30, 2015 to continue voluntary efforts to reduce racial, ethnic and economic isolation for Hartford-resident minority students. The Phase Stipulation continued operation of RSCO to support the state?s efforts in meeting its obligation pursuant to the Stipulation and She??v. ?Neill litigation. The parties extended the Phase Stipulation for two consecutive one?year periods, ending respectively on June 30, 2016 and June 30, 2017. By order of the Connecticut Superior Court, dated August 7, 2017, the current application and lottery processes and management and administration provisions of the Phase Stipulation, as extended, continue in effect until further order of the court. . In accordance with the Phase II Stipulation, the CSDE created the Sheff Of?ce as the central authority in the planning, development, implementation, and support of all educational programming designed to reduce the racial, ethnic, and economic isolation of Hartford?resident minority students in the Greater Hartford Region. CSDE also created RSCO pursuant to the Phase II Stipulation as a collaborative between and among the Sheff Of?ce, CSDE, and Sheff stakeholders, including CREC and the Hartford Public Schools, to support the various Sheff initiatives and programming to reduce the racial, ethnic, and economic isolation of Hartfordnresident minority students in the Greater Hartford Region. . Among numerous other responsibilities, the Phase II Stipulation tasks RSCO with the development and implementation of a common application for all interdistrict magnet schools and Open Choice schools in the Greater Hartford Region that are included Within the Region?s Sheff School Choice programming to reduce racial, ethnic and economic isolation of Hartford?resident minority students. CREC, Hartford and the other operators of Sheff interdistrict magnet schools participate in the development and implementation of the common application. RSCO makes the application available to families from some time in October or November through the closing date in late January or February of the same school year. For the 2017-18 school year and the upcoming 2018?19 school year, the on?time application period ran from November 1St to February 28th for placement in a school/program the following school year. . Applicants complete the RSCO application online at For families that do not have access to a computer, RSCO makes computers available at Parent Information Center in Hartford, at various branches of the Hartford Public Library, at satellite sites in Hartford, and at various libraries in the suburban communities. Families may submit only one RSCO lottery application per student. . Families submit an application for a seat in a Sheff School Choice program in the school year prior to enrollment. Students entering pie?kindergarten through grade 12, and who reside in Connecticut at the time of application, are eligible to apply. The largest number of available seats tends to be in transition grades, including prekindergarten, kindergarten, 6th grade and 9th grade. 10. On?time applications include those complete applications submitted during the application 11. period. RSCO accepts late applications through a speci?ed date which is published on the RSCO website and RSCO catalog but those applicants do not receive a placement until on?time applications for the particular school or Open Choice program are exhausted. For the 2017?1 8 lottery, families are able to submit an online late application from April 2, 2018 through July 31, 2018. Late applications are considered in the order received. The common application allows families to apply for enrollment in a Sheff interdistrict magnet school or an Open Choice program or both. An applicant may select up to five interdistrict magnet schools, in order of preference, and/ or up to ?ve Open Choice schools. For Hartford residents applying to an Open Choice option, applicants may rank up to five suburban districts within the applicant?s residential zone in order of preference. As an option, Hartford-resident applicants may select ?All Districts? to be considered for an available seat in any participating Open Choice suburban district. Suburban residents may also apply for a seat in the Open Choice Program in a public school in Hartford. Suburban applicants may select up to five available schools, in order of preference. On the magnet side, applicants may select an ?All Magnets? option as an additional selection beyond their specific magnet choices (up to five) as part of their on?tirne lottery application. If an interdistrict magnet school exhausts its waitlist, RSCO will offer available placement opportunities at that school to applicants who selected the ?All Magnets? option. 12. In addition to designing and implementing the common application, the Phase II Stipulation tasks RSCO with the design, implementation, and operation of a uniform school choice lottery for placing applicants in available seats in interdistrict magnets schools and Open Choice programs within the Sheff portfolio. RSCO must operate a lottery for student l3. 14. 15. placements since the number of applicants for most schools and programs exceed the number of seats available. The RSCO lottery is the computer?based method that places students who have submitted a complete and on?time application in available RS CO schools and programs. The CSDE designs and operates the computerized lottery system in?house based on protocols and seat declarations provided by the various operators of the interdistrict magnet schools and Open Choice Programs within the Sheff School Choice options. Although the CSDE operates the lottery, the operators and Open Choice districts manage their respective educational environments and, therefore, declare the number of seats available at each grade level during each lottery round. For the interdistrict magnet schools, the operators designate the number of seats they want to ?ll on a school?by~school basis, broken down by grade and by Hartford?resident students and suburban-resident students. if the school maintains a partnership agreement with two or more towns, the operator designates the number of seats allocated to students from those towns for each applicable grade level based on the terms of the partnership agreements. In declaring seats for the interdistrict magnet schools, CREC and Goodwin College, by statute, must declare seats in such a way as to ensure that at least 50% of their incoming class at each magnet school are Hartford?resident students. Unless they receive a waiver from the CSDE for a speci?c reason, CREC and Goodwin College incur a ?scal penalty for failing to meet the 50% minimum. This provision was added to the law to increase Hartford~resident enrollment at Goodwin College and CREC magnet schools. By statute, Hartford must meet the 50% minimum Hartford-resident enrollment requirement for Great Path Academy absent a waiver from the standard. 16. 17. Open Choice schools in suburban communities also declare seats by school and by grade in their respective towns for placement of Hartford?resident students participating in the RSCO lottery. Hartford Public Schools provides educational opportunities to suburban- resident students in certain Hartford neighborhood schools pursuant to the Open Choice Program and provides RSCO with the grade and school information for placement of suburbanuresident students participating in the RSCO lottery. In addition to providing RSCO with the seat declarations for interdistrict magnet schools and Open Choice Programs, operators provide RS CO with what we refer to as ?protocols? for pro gramming the lottery algorithm. Essentially, protocols provide the instructions on how to order and run the lottery; those instructions identify preferences for each school and Open Choice Program, which direct the order of placement of applicants in each respective school or program. Preferences are applied for a number of reasons, including, for example at interdistrict magnet schools: a sibling preference for applicants with a sibling(s) at the selected school; a neighborhood or zone preference for applicants that live within a certain distance from the selected school; a staff preference for applicants who have a parent(s) working at the selected school; a pathway preference for applicants transitioning from the end grade of a Sheff interdistrict magnet school and applying to the entry grade of the selected school; and a partner district preference for applicants that live in the town that has a partnership agreement with a Sheff magnet school to send speci?c numbers of students to that school. or the Open Choice Program, preferences include the following: an applicant sibling preference for an applicant with a sibling or siblings who also applied to Open Choice in the same district; an enrolled sibling preference for applicants with a sibling(s) who is already enrolled in Open Choice; and 18. 19. 20. an Open Choice Only preference for applicants who select only Open Choice on the RSCO lottery application. Preferences do not guarantee placement in a selected school/district but provide the applicant with an increased likelihood of placement over an applicant without a preference. Operators provide the preferences on a school?by?school and grade?by- grade basis separately for Hartford?resident applicants and suburbannresident applicants, and determine the weight a preference will have at each school in terms of ordering the applicant pool for placement. The preferences for Hartford applicants may differ from the preferences for suburban applicants; for example, the neighborhood preference for applicants who live within a certain radius of a Hartford school would only apply to Hartford?resident applicants. In addition to the aforementioned preferences, CSDE programs the lottery to sort the suburban applicant pool for magnet schools based on certain characteristics of their resident town. In the past, the two sorting methods were limited to a sorting preference based on the sending town?s participation rate in school choice options (hereinafter referred to as ?Participation Rate Sort?), or the sending town?s percentage of applicants who identify as a race and ethnicity that meets the reduceduisolation definition provided in the Phase Stipulation, as extended, and the Commissioner of Education?s interdistrict magnet schools Reduced?Isolation standards for purposes of Section 10?264], as amended by PA 17?172 (hereinafter referred to as ?Town Reduced?Isolation Sort?). A reduced- isolation student is a student who identifies as Native American, Asian, Alaska Native, Native Hawaiian, Other Paci?c Islander, and/or White, and does not identify as Black/African American, or Hispanic. For the 2017-18 RSCO lottery (for placement of 21. 22. students in the 2018?1 9 school year), RSCO added a third sorting method that divides sending towns into United States Census block groups and sorts the block groups by the percentage of applicants who identify as a race and ethnicity that meets the reduced? isolation de?nition (hereinafter referred to as ?Block Group Reduced?Isolation Sort?). Operators identify the preferred sorting as part of their protocols for each school in their portfolio. RSCO collaborates with the operators to ensure the selected sorting method is appropriate to achieve the required reducedwisolation compliance standard for Sheff interdistrict magnet schools. However, before any sorts or preferences are applied to the applicant pool, the computer assigns each applicant a randomized number. The randomized number determines the order of offers in ascending order once preferences and protocols are applied. The Participation Rate Sort is the sorting protocol that we have used since implementing the uniform lottery for all Sheff interdistrict magnet schools and Open Choice placements in the Greater Hartford Region. The Participation Rate Sort sorts applicants who live in towns other than Hartford by school, by grade, and by magnet choice according to their town?s rate of participation in school choice options, from lowest percentage to highest. The Participation Rate Sort (as well as the Town Reduced?Isolation Sort and the Block Group Reduced-Isolation Sort) preserves applicants? magnet choice order so the first choice applicants to each school are ordered above the second choice applicants to a school and so on through the fifth choice; but the applicants are ordered within each choice, by grade, according to the town participation rate, from lowest to highest. Within each town grouping of applicants in order of participation rate, students receive offers in ascending order based on their randomized number. 23. 24. RSCO began using the Town Reduced?Isolation Sort in the 2015~16 RSCO Lottery for the 201647 school year for schools that were not meeting the standard for a Reduced?Isolation Setting (hereinafter as required by the Phase Stipulation, as extended. In accordance with the Phase Stipulation and the Commissioner of Education?s interdistrict magnet schools Reduced?Isolation standards for purposes of Section 10?264], as amended by PA 17?172 (hereinafter the Standard?), a school provides a RIS if the percentage of enrolled students who identify as any part Black/African American, or any part Hispanic, does not exceed 75% of the school?s total enrollment. Unless a school received a speci?c waiver by agreement of the parties, the Phase Stipulation, as extended, only counted schools that met the R18 for purposes of the annual benchmark goal of increasing the number of Hartford-resident minority students in reduced-isolation settings. The Order of the Superior Court, dated August 7, 2017, continued to set 75% as the standard for a Reduced?Isolation Setting until further order of the court. The Town Reduced?Isolation Sort sorts applicants who live in towns other than Hartford by school, by grade, and by magnet choice according to their town?s percentage of applicants who identify as a reduced?isolation student, from highest percentage to lowest. Like the Participation Rate Sort, the Town Reduced?Isolation Sort preserves applicants? magnet choice order so the ?rst choice applicants to each school are ordered above the second choice applicants to a school and so on through the fifth choice; but the applicants are ordered within each choice, by grade, and by town according to each town?s percentage of reduced?isolation applicants, from highest to lowest. Within each town grouping of applicants, applicants are randomized and receive offers in ascending order based on their randomized lottery number. Applicants are not sorted by race within their town grouping. 25. 26. 27. During the 2017?1 8 RSCO lottery cycle, we added a third sorting method for placing students for the 2018~l9 school year at schools where the percent enrollment of reduced? isolation students is 27% or less. This third protocol, the Block Group Reduced?Isolation Sort, divides sending towns into United States Census block groups and sorts applicants by school, by grade, by magnet choice according to their block group?s percentage of reduced? isolation students, from highest to lowest. A Block Group is a geographical unit used by the United States Census Bureau, which is between the Census Tract and the Census Block. The Block Group is the smallest geographical unit for which the bureau publishes sample data. Using census data from the Bureau?s American Community Survey, RSCO sorts applicants in the same manner as the Town Reduced?Isolation Sort except it uses block group rather than town. Within each block group of applicants, applicants are randomized and receive offers in ascending order based on their randomized lottery number. Applicants are not sorted by race within each block group. RSCO added this third sorting protocol for those schools below, at, or only above the RI Standard as a mechanism to improve compliance and meet the terms of the Order of the Superior Court, dated August 7, 2017. The three sorting methods identified above order the suburban applicant pool after each school?s pool is ordered by grade and by magnet choice (?rst through ?fth) and by operator preferences (such as sibling, staff, pathway). Applicants with preferences are ordered above applicants without preference, irrespective of the sorting protocol used at a school, but, as noted above, the operator selects the weight of each operator preference. Applicants . receive offers in ascending order based on their randomized number to fill the seats as declared by the operator. The Hartford applicant pool is also sorted by school, by grade, by 10 28. 29. magnet choice and by preference but not by the other sorting protocols since approximately 97% of the Hartford~resident student population identify as a race(s) and/or ethnicity Within the de?nition of a minority student as set forth in the Phase Stipulation, as extended. Like suburban applicants, Hartford applicants receive offers in ascending order based on their randomized number to ?ll the seats as declared by the operator. RS CO does not jump. over applicants but extends offers in ascending order based on the lottery run. Prior to running the actual lottery, RSCO cleans the data to make sure applications are complete and there are no duplicates, and then runs various lottery simulations to ensure protocols are working properly and schools will meet the RI Standard. An operator will adjust protocols and/or seat declarations if a school is not projected to meet the RI Standard based on the simulation. Despite these efforts, however, there are a few schools that are not projected to meet the compliance standard in the simulations but will go through the actual lottery run with the expectation the school will not be compliant. In those instances, RSCO works with the operator to adjust the seat declarations to avoid any further slippage from the RI Standard. After RSCO completes the simulations and protocols are working properly, RSCO runs the actual lottery and applicants receive notice of the outcome through the online system. Applicants who receive an offer must accept or decline the offer within the timeframe specified in the notice. During the initial weeks of the process, RSCO affords applicants two weeks to respond to an offer. RSCO administratively declines students who do not respond by the time specified in the notice. The timeframe for a response shortens through the summer and towards the start of the school year in late August to maximize ?lling ll 30. 31. 32. seats. Applicants who do not receive an offer are so noti?ed and receive a form that they must complete to remain on the waitlist at their selected schools. During the 2016?17 lottery (for the 2017?18 school year), RSCO reapplied the Town Reduced?Isolation Sort to the suburban applicant waitlist after the ?rst run of the lottery. RSCO implemented this reshuf?ing of the suburban waitlist after the initial placement offers were pulled but before letters were released. By reapplying the sorting protocol to the remaining list of applicants on the waitlist after the first placements were determined, RSCO increased the number of reduced?isolation students at the top of the suburban applicant waitlist. Once the reshuf?e was complete, RSCO released the offer letters and sent out the waitlist notices. RSCO will implement this same reshuf?ing to the suburban applicant waitlist after the first run of the lottery using the Block Group Reduced-Isolation Sort in 2017?1 8 for the 2018-49 school year. Once the period ends for applicants to respond to a placement offer, operators provide new seat declarations to RS CO by school and by grade, for Hartford residents and suburban residents separately, to ?ll any remaining available seats. RSCO extends placement offers based on the waitlist order from the initial run of the lottery, or as resorted prior to issuance of the ?rst placement notices. Operators must carefully ?ll seats through the various rounds of the lottery to ensure educational benefit, resource efficiency, ?scal integrity, and compliance with the R1 Standard. Since RSCO does not skip over applicants but follows the order on the waitlist when extending placement offers, operators must consider the racial and ethnic demographics of applicants in determining whether to fill available seats and how many to ?ll. RSCO collaborates with operators on filling seats as it relates to school compliance? l2 33. 34. 35. and will restrict the number of seats ?lled, if any, if adding students from the waitlist will negatively affect the school?s compliance with the RI Standard. RSCO communicated to all Sheff magnet operators that placement offers for Sheff magnet schools and Open Choice programs must go through the RSCO lottery process. On those occasions when RSCO identifies students who are placed in a school/program outside of the lottery process, the operator does not receive the per pupil grant payment for that student. In 2017?18, RSCO received 5,975 applications from Hartford?resident students and 12,743 applications from suburbanuresident students. Of the 12,743 applicants from the suburban communities, 48.8% are reduced-isolation students. Of the 5,975 Hartford-?resident applicants, 7.8% are reduced?isolation students. These percentages are typical of the applicant pool. The percentages of reduced?isolation applicants at schools that are at or below the RI Standard are much lower than the percentages of reduced-isolation applicants at schools that are above the compliance requirement, making it more dif?cult to remain below the 75% standard at these schools. In general, in order to enroll at least 25% reducedwisolation students to get to the RI Standard, a school?s suburban applicant pool must have at least 50% reduced~isolation applicants since schools enroll as close as possible to a 50?50 split between Hartford?residents and suburban-residents each lottery cycle and rely predominately on the suburban applicant pool to meet the compliance standard. 13 Glen Patti/son 52-7?? Subscribed and sworn to before me this day of Aprll, 2018. @Mo?uak/D D?ren Kowalski Notary Public Connecticut My Commission Expires June 30, 2020 14