Appendix E

Data Collected from School Districts

The panel obtained data on school districts from a variety of sources, including both administrative data sources described in Chapter 3—the Common Core of Data (CCD) and form FNS-742. We also collected data directly from school food authority directors in selected districts. This appendix describes the panel’s collaboration with these directors and the data that they provided.

In collaboration with the school food authority directors of our five case study districts, we collected school level data concerning enrollment, certification and meals served by year for those districts. The case study data collection plan is described in Chapter 4 of the panel’s interim report (National Research Council, 2010); additional detail on the data is provided in Chapter 3 of this report. Part 1 of this appendix presents detail on the data collected and comparisons of data provided by the school districts and data from national administrative databases.

On March 3 and 4, 2010, the panel held a workshop with school food authority directors from the case study districts, as well as other directors who had experience with Provision 2 or 3. One representative of child nutrition services from a state education department also participated in the workshop. The agenda for the workshop appears as Part 2 of this appendix. The discussions at the workshop are summarized in Chapter 3.

The panel also conducted an informal survey of districts that reported operating under Provision 2 or 3 on the FNS-742 form. Results of that survey are discussed in Chapter 3. Part 3 of this appendix describes the survey methodology. The results of the survey represent only the views



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 286
Appendix E Data Collected from School Districts T he panel obtained data on school districts from a variety of sources, including both administrative data sources described in Chapter 3--the Common Core of Data (CCD) and form FNS-742. We also collected data directly from school food authority directors in selected districts. This appendix describes the panel's collaboration with these directors and the data that they provided. In collaboration with the school food authority directors of our five case study districts, we collected school level data concerning enrollment, certification and meals served by year for those districts. The case study data collection plan is described in Chapter 4 of the panel's interim report (National Research Council, 2010); additional detail on the data is pro- vided in Chapter 3 of this report. Part 1 of this appendix presents detail on the data collected and comparisons of data provided by the school districts and data from national administrative databases. On March 3 and 4, 2010, the panel held a workshop with school food authority directors from the case study districts, as well as other directors who had experience with Provision 2 or 3. One representative of child nutrition services from a state education department also participated in the workshop. The agenda for the workshop appears as Part 2 of this appendix. The discussions at the workshop are summarized in Chapter 3. The panel also conducted an informal survey of districts that reported operating under Provision 2 or 3 on the FNS-742 form. Results of that survey are discussed in Chapter 3. Part 3 of this appendix describes the survey methodology. The results of the survey represent only the views 286

OCR for page 286
APPENDIX E 287 of the respondents. The sample frame was incomplete, and the response rate was very low, so the results are not generalizable. PART 1: CASE STUDY DISTRICTS--DATA COLLECTION AND COMPARISON OF ENROLLMENT AND CERTIFICATION The panel contacted six school districts in the United States, invit- ing them to serve as case study districts for this study. Chapter 4 of the panel's interim report (National Research Council, 2010) describes how potential case study districts were identified and how they were recruited. Case study districts were needed so the panel could investigate how a new provision might work in individual schools or groups of schools. The districts provided digitized attendance area boundaries and detailed information on program operations. The ability to provide digitized atten- dance boundaries is a key requirement of this potential new provision if a district wishes to operate at a subdistrict level. The Census Bureau pro- vided American Community Survey (ACS) estimates of eligibility for all the schools with boundaries in the case study districts. The panel evalu- ated the quality of the ACS-based estimates in terms of sampling error and other properties that affect fitness for use. This analysis is described in Appendix F. The panel also used data from the case study districts on the percentages of meals served by category (free, reduced price, and full price). The participation analysis is described in Chapter 4. This part of the appendix documents the collection of data from the case study districts, provides a brief description of each district, and compares school-level data provided by the districts with data available through national administrative sources. It first describes the districts in terms of percentage of students that are likely to attend schools with digi- tized boundaries. This discussion paves the way for analysis of the impact of open enrollment, charter, and magnet schools in Appendix G. Also included is a discussion of individual schools identified by the district and by administrative sources. Finally, this part of the appendix presents comparisons of data on enrollment, number of students certified for free meals, and number of students certified for reduced-price meals for each public school listed by the district and by the National Center for Educa- tion Statistics' (NCES') CCD, a public source of information about public schools and public school districts in the United States. The five school districts listed in Table E-1--Austin, Texas; Chatham County, Georgia; Norfolk, Virginia; Omaha, Nebraska; and Pajaro Valley, California--agreed to serve as case study districts for this study. The panel would like to express its appreciation for the vast amounts of data they provided, the help they offered while we compiled and analyzed the

OCR for page 286
288 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS TABLE E-1 Case Study Districts Number of Number of Students in Schools Participating Students Without Boundaries School District Schools (in thousands) (percentage of enrolled)* Austin, Texas 114 83 3.0 Chatham, Georgia 46 35 5.4 Norfolk, Virginia 56 36 10.0 Omaha, Nebraska 86 47 4.6 Pajaro Valley, California 32 19 7.4 *Omaha and Chatham are also open enrollment districts. In open enrollment districts, many schools have geographic boundaries, but students are not required to attend neighborhood schools. SOURCE: Prepared by the panel. data, and their observations on how the school meals programs work in practice. Charter/Magnet/Open Enrollment Boundary Issues Austin has no charter schools and no magnet schools. The school dis- trict provided the panel with digitized boundaries for 106 schools, but the Census Bureau did not provide ACS data for 2 of these schools (Brooke Elementary and Ridgetop Elementary) because they did not pass disclo- sure review.1 Boundaries were provided for schools that were in operation during 2009-2010 (a year for which the panel did not collect detailed data from the district), including two schools that were new in 2009-2010-- Gorzycki Middle School and Green Tech High School. ACS estimates were provided for these schools, but the panel is not sure to which schools in prior years the data apply. There are 18 schools with no boundaries. Of these, 8 are nontraditional schools that do not participate in the school meals programs. Of the remaining 10, 5 are alternative schools, and the others appear to be traditional schools: 1 middle, 2 high, and 2 elementary schools. The percentage of students attending participating schools that do not have boundaries is only 3 percent. Chatham provided digitized boundaries for 45 schools. There are 4 schools with no boundaries, 3 of which are charters that do not partici- pate in the school meals programs. One is a vocational school that does participate in the school meals programs and accounts for 5.4 percent of 1The panel did not receive an attendance boundary for the Read Pre-K Demonstration project; however, School Attendance Boundary Information System (SABINS) analysts deter mined that it does have boundaries. The attendance zone for Read includes the zones for Cook, McBee, and Wooldridge.

OCR for page 286
APPENDIX E 289 students in participating schools. According to the CCD, Chatham also has 13 magnet schools with 32 percent of district enrollment. Chatham is an open enrollment district. Norfolk provided digitized boundaries for 46 schools. There are 9 schools with no boundaries enrolling 10 percent of Norfolk's students, including 4 nontraditional schools and 5 others: Early Childhood Center at Berkley/Compostella, Easton Preschool, Ghent Elementary, Rosemond Middle, and the School of International Studies. All of these schools par- ticipate in the school meals programs. The district said it has no charter or magnet schools but has 2 open enrollment schools: Dreamkeeper Acad- emy and Ghent Elementary. The CCD indicates that there are no charter schools but lists 4 magnets: Blair Middle School, Dreamkeeper Academy, Maury High, and Norview High.2 Omaha is an open enrollment district. It provided digitized bound aries for 79 schools, but the Census Bureau did not provide ACS data for Franklin Elementary school because it did not pass disclosure review. The district also provided a matrix for 2009-2010 showing the number of enrolled students and numbers eligible for free and reduced-price meals by home school crossed by school attended. The district has no charter schools, but it has alternative schools and magnet schools that partici- pate in the school meals programs. None of the 8 alternative schools has boundaries (1.6 percent of enrollment). There are 4 main magnet schools (elementary and middle combined, and the elementary parts do not have boundaries). The elementary parts account for 3.1 percent of enrollment; the middle school parts account for 4.6 percent. There are also 13 other schools that have "magnet" as part of the school name (21.1 percent of students enrolled in participating schools). The CCD has no reported charter schools or magnet schools in Omaha. Pajaro Valley provided boundary information for 25 schools.3 It has charters but no magnets. None of the 5 charter schools has boundary information. Two other schools did not have boundary information: New School Community Day and Renaissance High Continuation. The charter 2There are two funding sources for Berkley/Campostella, and for this reason there are two catchment areas for the school. For 3-year-old children, the catchment area is the Compostella Elementary School boundary, while for 4-year-old children, the catchment area is the entire district. This is an example of a school having different boundaries for different grades. 3SABINS and Census Bureau analysts established that the district boundary for Pajaro Valley in the Census Bureau's Topologically Integrated Geographic Encoding and Refer- encing (TIGER) databases is larger than the geographic extent of the school areas reported by the district to SABINS. It was established that the latter is accurate and that the state of California did not provide updated boundaries for Pajaro Valley as part of semiannual boundary updates.

OCR for page 286
290 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS schools that participate in the school meals programs account for 6.7 per- cent of the enrollment of participating schools. The remaining two schools with no boundary information account for 1.4 percent of enrollment. For all five case study districts, the ACS school district enrollment estimates are somewhat larger than the sum of school catchment area esti- mates and estimates for enrollment in schools without boundaries. Census Bureau and School Attendance Boundary Information System (SABINS) analysts think the difference may be due to handling of pre kindergarten students. SABINS school boundaries are grade specific, beginning at kinder garten, while census districts go down to prekindergarten. If a dis- trict had students in prekindergarten, they would be included in Census Bureau school district estimates, but they would not be included in the school's counts if SABINS did not specify their inclusion. Data Provided The protocol used by the panel to request data from the case study districts appears as Attachment C to Chapter 4 of the panel's interim report (National Research Council, 2010). Table E-2 shows the data that were provided by the case study districts. The panel requested data for each school year from 2003-2004 through 2008-2009 so as to have data to compare with the 5-year 2005-2009 ACS estimates. Chatham did not provide enrollment data but did provide average daily attendance and meal count information. Chatham does not have data for 2003-2004 or for 2004-2005. Norfolk provided data on direct certification and applications for 2009-2010 but does not have these data for any preceding years. For Pajaro Valley, no data are available for 2003- 2004, and meal counts are not available for 2005-2006. Omaha does not have numbers of students certified for free and reduced-price meals in 2003-2004 or 2004-2005. TABLE E-2 Data Received from Case Study Districts Data Austin Chatham Norfolk Omaha Pajaro Valley Enrollment Attendance Certification Direct Certification NA Applications o NA Meal Counts NOTE: = data received at school level; o = data received at district level. NA= data not available for years requested SOURCE: Prepared by the panel.

OCR for page 286
APPENDIX E 291 All five districts reported that applications are reviewed and verifica tions are conducted centrally, by the district. Norfolk said that appli cations are mailed to all households for students registered at the end of July, and that applications also are available at orientation programs for new students prior to the start of the school year, online, in all school offices, and in all cafeterias. According to the Food and Nutrition Service (FNS), this approach apparently is typical, although Chatham said it is trying to get families to use the online form. Pajaro Valley reported that it is operating under Provision 2 for breakfast in 14 schools. Omaha is using Provision 2 for breakfast in all schools. Norfolk uses a policy of no fee for reduced-price lunches and is considering Provision 2 for breakfast. Chatham began offering univer- sal free breakfast in schools with at least 80 percent of students eligible for free and reduced-price meals in 2009-2010. Other responses provided by the case study districts to the panel's telephone survey are summarized in Annex E-1 at the end of this appendix. Each district provided multiple lists of schools: one list with atten- dance, one with certification information, one with meal counts by category, sometimes another for enrollment, one with school addresses, and so on. In some cases, schools have different names on different lists. For each school district, the panel entered the school-level data into spreadsheets with one tab for each school year. In the spreadsheets, one row contains all information about a given school, including information for that school from the CCD. A number of calculated variables also are included on the spreadsheet; examples include participation rates (meals served in a cat- egory divided by students certified in that category) and number of days on which meals were served in October (average daily participation divided by the total number of meals served). These variables provided information with which to check data entry, as well as identify questions for the district. The CCD conducts five surveys annually to collect fiscal and non fiscal data on all public schools, public school districts, and state education agencies in the United States. For purposes of this study, the most rel- evant information from the CCD is school characteristics and school-level counts of enrolled students, students certified for free meals, and students certified for reduced-price meals. Enrollment and certification data are as of October 1 (or the closest school day to October 1) of the school year for all grade levels (prekindergarten, kindergarten, and grades 1 through 12) and ungraded students. State officials are instructed to include students enrolled in the school who reside in the attendance area of a different agency. This can occur, for example, when a school district "tuitions out" a student to receive some services the district cannot provide. In this case, the receiving public school and agency include the student in their membership counts. Also, if the student tuitions out to a public school

OCR for page 286
292 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS district in a different state, the student is counted where he/she receives education services. However, students tuitioned out to private schools are not included in the CCD. FNS collects data on verification activities on the School Food Author- ity Verification Summary Report, Form FNS-742.4 With few exceptions, each school district that operates the National School Lunch Program (NSLP) or School Breakfast Program (SBP) must report the information on this form annually. Section I of the form solicits information as of the last operating day in October. Among the data items included are the number of schools operating the NSLP or SBP and the number of enrolled students with access to the NSLP or SBP. The total numbers of students certified for free and reduced-price meals under the school food authority (SFA) also are reported. Data on number of schools, enrollment, and percentages of free- and reduced-price-eligible students can be compared with aggre- gates of school level data from the district and with school district-level data from the CCD. For each of the five case study districts, the panel's analysis of enroll- ment and certification started with the district's list of schools that provided meals under the NSLP for 2008-2009 (data as of October 31, 2008). For these schools, we compared enrollment, number certified for free meals, and number certified for reduced-price meals with the equivalent data from the CCD for 2008-2009 (data as of October 1, 2008). Note that the dates of the data are different, and this is one reason we do not expect the numbers to agree perfectly. In the notes below, we describe the matching of the schools identified by the district and participating in the school meals programs with the schools listed in the district in the CCD. In most districts, there are schools (generally charter schools or nontraditional schools of some kind) that do not participate in the NSLP but for which enrollment and the numbers of students certified for free and for reduced-price meals are available. Occasionally, schools are combined for reporting purposes. All districts identified schools with a range of grades, from pre kindergarten through grade 12. School districts that operate p rekindergarten programs can claim meals under either the NSLP/SBP or the Child and Adult Care Food Program.5 4The form is available at http://www.fns.usda.gov/cnd/Governance/Forms/SFA_Direct Verification_Summary.pdf. 5The Child Nutrition Act of 1966 provides the following: PRESCHOOL PROGRAMS SEC. 12. 42 U.S.C. 1781. "The Secretary may extend the benefits of all school feeding programs conducted and supervised by the Department of Agriculture to include preschool programs operated as part of the school system." Policy memorandums regarding Head Start and Even Start Programs are at the follow- ing links: http://www.fns.usda.gov/cnd/governance/Policy-Memos/2008/SP_23-2008. pdf and http://www.fns.usda.gov/cnd/governance/Policy-Memos/2008/SP_34-2008.pdf.

OCR for page 286
APPENDIX E 293 Austin In 2008-2009, there were 114 schools participating in the school meals programs in Austin. All of these schools are also listed in the CCD. Data for these schools are included in the comparisons provided below. The FNS-742 form also reports that there are 114 schools participating in the school meals programs in Austin. Ten of these schools (with enrollment of 3,427, or 4.1 percent of district students) have no school-level geographic boundaries and are open enrollment schools. Both the district and the CCD have enrollment and certification data for five nontraditional schools that do not participate in the school meals programs: the Austin State Hospital, a residential facility for students with mental health issues; the Gardner-Betts Juvenile Justice Center's Leadership Academy, a halfway house for students who have been adju- dicated and incarcerated (the program is not housed in one of Austin Independent School District's [ISD's] schools); Travis County Juvenile Justice Alternative Education Program, a temporary nonresidential school for students in the county that have been adjudicated (this countywide program is hosted on one of Austin ISD's campuses, and Austin receives verification from surrounding school districts as to each students' status for meal benefits); Phoenix Academy, a residential school for students with drug addiction (not hosted on an Austin ISD campus); and Travis County Juvenile Detention Center, also a countywide program not hosted in an Austin ISD school. According to both data sets, these are small schools, with a total of 218 students in 2008-2009. Two schools listed by the district are not shown as participating in the school meals programs and are not in the CCD--the Infant Development Center and Even Start (babies). The district says Infant Development is for the babies of students; in some cases, the district serves meals, depending on the age of the baby. The students in Even Start (babies) are categori- cally eligible, but the program changes location annually. The district did not provide enrollment data but did provide certification data. For each year, it also provided the list of schools where Even Start was housed. One additional school--the Travis County Day School, with 16 students--is in some Austin district records for 2008-2009 but with no meals served. Chatham Chatham has 46 schools that participate in the school meals programs. For one of these schools, Woodville Tompkins Tech and Career Institute, a vocational school listed in the CCD beginning in 2007-2008, there are neither enrollment nor certification data from the CCD or the district. The Internet shows this is to be a vocational high school that serves the school

OCR for page 286
294 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS district. It is the only school serving meals that does not have geographic boundaries, and is specifically open to all district children. The other 45 schools are listed in the CCD, which contains both enrollment and certification data for these schools. The summary below is based on the 45 schools for which data are available in both systems. There are three schools--Coastal Empire Montessori Charter (new in 2008-2009), Savannah Arts, and Oglethorp Charter--for which the district provided certification data but does not provide meals. These schools are also listed in the CCD, which contains both enrollment and certification data for them. Three schools were not listed by the district: the Bethesda Home for Boys, the Savannah Gateway to College (new in 2008-2009), and the Universal Health Services of Savannah Coastal Harbor Treatment Center. The CCD has enrollment data for these schools but not certification data and terms them "regular" schools. The Bethesda Home for Boys collects tuition and therefore is not a public school. The Savannah Gateway to College is a charter high school. Savannah Coastal Harbor Treatment Center works with children who have failed in other residential settings; it provides 24-hour nursing in a locked and secure environment. The CCD lists a total of 51 schools with students in Chatham (note that this excludes Woodville Tompkins). The FNS-742 form reports 55 schools for Chatham. Norfolk Norfolk has 56 schools that participate in the school meals pro- grams. The district provided enrollment and certification data for all of these schools. The district also provided geographic boundaries for all schools except the 4 nontraditional schools noted below and 5 regu- lar schools: Early Childhood Center at Berkley/Campostella, Easton Pre- school, Ghent Elementary, Rosemont, and School of International Studies at Meadowbrook. Together these schools have 3,494 students, which is about 10 percent of the district's enrolled students. In 2008-2009, the CCD listed only "regular" schools. Hence it did not list the four nontraditional schools in the district: Madison Career Center, Norfolk Marine Institute, Norfolk Re-Ed School (South Eastern Coopera- tive Education Programs at Tucker), and Norfolk Technical Center. The district reported that none of these is a traditional school, and all draw from the whole city. Norfolk Re-Ed School includes students from other cities as well, mainly emotionally disturbed students. Madison Career Center draws students from the whole city, primarily those with disci- pline problems, and attendance may be recorded at the home school. Norfolk Marine Academy is the same as Norfolk Marine Institute. Cur-

OCR for page 286
APPENDIX E 295 rently, students attend every other day, and attendance may be recorded at the home school. Norfolk Technical Center students alternate days there and at their home school. Students are tracked by the day and do not have access to meals at both schools on the same day. Attendance records at Norfolk Technical Center are shared electronically with the home school. (The CCD reported Norfolk Technical Center as a vocational school in 2003-2004, 2004-2005, and 2005-2006. It listed Madison Career Center as other/alternative in the same years.) Little Creek Elementary School and Little Creek Elementary School Annex are another issue. Little Creek is among the 56 schools for which the district provided data. Although most data items were reported sepa- rately for the two schools, enrollment and attendance data provided by the district combine the two. This combined enrollment number also is reported by the CCD (the Annex is not reported separately). However, the numbers of students eligible for free and reduced-price meals as reported by the CCD match the district's data for Little Creek Elementary only. For comparison with the CCD, the panel combined the data for the Annex with the data for Little Creek Elementary, yielding a total of 51 schools. The district did not provide a separate boundary for Little Creek Elemen- tary Annex. The FNS-742 form lists 60 schools for Norfolk. Omaha Omaha reported 86 schools participating in the school meals pro- grams. Of these, 78 are traditional schools,6 and 4 (Alice Buffett, King Science/Technology, Maars, and Morton) are magnet schools with both elementary and middle school programs. There are 8 alternative pro- grams: Blackburn, Early Childhood Center, ESL Teen Literacy, Integrated Learning Program, JP Lord, Parrish, Transitions at PVA, and Wilson. The district also lists the University of Nebraska at Omaha (UNO)7 as partici- pating in the school meals programs. Three of the alternative programs (Integrated Learning Program, Lord, and Parrish) have separate elemen- tary and secondary programs. On some lists (such as enrollment from the district), the elementary and secondary programs are counted separately; counting them separately would yield a total of 93 schools participating in 6"Traditional" means the district did not list the school as "alternative." The district's lists include 7 high schools, 7 middle schools, 60 elementary schools, and 4 elementary/middle school magnets that are listed separately as elementary and middle on some lists and are combined on others. The total number of traditional schools is either 78 or 82, depending on how the 4 magnets are treated. These schools are on all lists provided by the district. They are also included in the CCD. 7The CCD lists JP Lord, Jackson Elementary, Transitions at PVA, and Yates as "special education schools"; all the rest are "regular."

OCR for page 286
296 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS the school meals programs. None of the 8 alternative schools has bound- aries, and the 4 magnet schools have boundaries only for middle school. In these schools, the elementary program does not have a home area. Omaha has a program whereby some students can take college courses at UNO and receive credit toward graduation, as well as credits toward a degree. Since eating at the university would be more costly for students, especially those certified for free or reduced-price meals, the district drops off meals at the university. The district lists UNO as par- ticipating in the school meals programs, but does not have enrollment or certification data for these students. Since the students belong to other schools but attend the university for one or two classes, the meals served at UNO probably are counted at the students' home schools. The CCD lists 98 schools for Omaha, 89 of which have nonzero or nonblank enrollment. The CCD includes all the traditional schools and counts the elementary and middle school programs of the 4 magnets separately. Enrollment and certification data are available for all of these schools. The CCD lists five of the alternative schools participating in the school meals programs--Blackburn Senior High Program, ESL Teen Lit- eracy (Career Center), Parrish, Transition Program at PVA, and Wilson-- but includes no enrollment or certification data. The CCD also lists Yates Alternative Center without providing enrollment or certification data. The district reported that there was a name change, and Yates is now called the Integrated Learning Program. UNO is not listed in the CCD. In sum- mary, the CCD has enrollment and certification data for all the traditional schools and 2 of the 9 alternative schools (Early Childhood Center and Lord), yielding a total of 80 schools (combining the magnets' elementary and middle school programs). The certification data from the district were combined for the elementary and middle school programs of 3 of the 4 magnets (Buffett, King Science and Technology, and Morton), so the comparison below is for 81 schools. The CCD includes Omaha Public Schools Homebased, which is also on the district lists but with no meals served. The CCD lists seven schools8 that are not on the district lists, each designated as an Early Child- hood Center: Blackburn, Blumfield, DC West, Educare, M ockingbird, Karen-Western, and Fern Williams. The CCD reports enrollment at three of these. The district reported that it does serve meals at these schools. Some are Head Start programs that are operated by different school districts/entities. The FNS-742 form indicates that 88 schools in Omaha participate in the school meals programs. 8These schools were reported as being in the Omaha Public School District by the CCD beginning in 2007-2008.

OCR for page 286
APPENDIX E 305 the City of Austin street centerline file and would prefer to provide this information as it is already completed. If this is not possible, however, the district would be willing to conduct a test run to determine the amount of staff time necessary to geocode 100 percent of all students using Topo- logically Integrated Geographic Encoding and Referencing (TIGER) files. Pajaro Valley said the last major change to its boundaries was in 2003. There have been minimal changes to add a street here or there, but noth- ing significant so as not to upset parents, students, and schools. The dis- trict respondent is not sure what would be involved in geocoding student addresses using TIGER line files, but expressed willingness to work with the panel on this. However, the district has a large migrant population and children move frequently, so addresses may be an issue. The migrant season starts in May, and by October, it is over. Omaha reported that school boundaries were digitized in a geo- graphic information system (GIS)--Esri's ArcView--using historical maps and written descriptions of the smallest unit of division in the district, the unit zone. The boundaries underwent verification in 2008, with digitized versions of city data being used to correct all boundaries for accuracy. Parcel files from both Douglas and Sarpy Counties were used to place all lots in the correct attendance area. Additionally, streets, waterways, and railroads were used to draw boundaries not located along property lines. Boundaries are updated with any changes in attendance areas or the addition of new schools. All boundaries use spheroid Geodetic Refer- ence System (GRS) 1980 and North American Datum 1983 (NAD83) in the coordinate system State Plane Nebraska. Omaha currently geocodes official membership databases every year to serve a number of purposes. If the release of these data were allowed under the Family Educational Rights and Privacy Act (FERPA), the dis- trict would consider providing the panel with a geocoded version of the official membership database. 8.Does your district use the data on numbers of children certified for free or reduced-price meals for other purposes? If so, please list programs, how much funding is involved, and the source of the funding (state, local, and other). Austin reported only that state and local funding was $35.5 million, and that Title I federal funding was $22 million. Chatham reported that it uses the data on numbers of children certi- fied for free or reduced-price meals for: Title I programs; National Assessment of Educational Progress (NAEP) programs;

OCR for page 286
306 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS Criterion-Referenced Competency Tests (CRCTs); federal, state, and local grants; after-school and summer programs; the SAT; and summer school. The district does not know how much funding is involved, but said the sources of funding are federal, state, and local. The Norfolk school food authority director said there are many pro- grams locally that use these data. With a mother's permission, free- and reduced-price-eligible children can have free driver's education, behind- the-wheel training, band equipment, payment for field trips, and so on. She does not know what would happen if individually identifiable information were not available. She has no way to know about the funding for many programs. The state requires counts of free- and reduced-price- eligible students on its state testing forms. She provides these data quarterly. The Omaha school food authority director said Title I funding comes through the state from the federal government and is allocated to the dis- trict based on eligibility for free or reduced-price meals; for the 2008-2009 school year, this amount was $22,639,970. The district also uses eligibility for free or reduced-price lunch in the equalization-of-funds calculation, which redirects and equalizes funding at all schools based on these num- bers; the amount of district funds impacted in 2008-2009 was $8,571,778.61. The district uses eligibility for free or reduced-price lunch to determine student activity card fees; the amount of district funds impacted in 2008- 2009 was $474,052.50. Additionally, Omaha obtains parent-reported lunch eligibility status on transfer forms to determine eligibility in the district's open enrollment plan. School choice and transportation to schools outside a student's home attendance area are determined by reported eligibility status. The Pajaro Valley school food authority director stated that she gives the data to the testing department and the advanced placement counsel- ors. Those who are eligible for free or reduced-price meals incur no or reduced-price fees for tests. The data are also given on the consolidated application for Title I funding for the district. 9.Does your district have up-to-date information about the number of charter and magnet school students and their participation in the school meals programs? Do you have data about the num- ber of children in home-schooling? Do you have information about students attending schools outside the school attendance boundaries because of open enrollment or public school choice programs?

OCR for page 286
APPENDIX E 307 Austin reported that it has no information about charter and mag- net school students, children in home schooling, or students attending schools outside their attendance area because of open enrollment or pub- lic school choice programs. Norfolk reported that it has no charter or magnet schools. It has one elementary school and one K-8 school with open enrollment. In these schools, most students are still from the local neighborhood. The district knows which students from other neighborhoods attend those schools and the name of their catchment area school. It also has a program called "minority/majority" that allows students from schools with high counts of minorities to be bused to schools that are more racially balanced. This is the last year for the program because of the cost of busing. Omaha has had an open enrollment program since 1999 and updated the program for the 2010-2011 school year. Data were available on all students not attending their home school for the 2008-2009 school year. There were no charter schools in the district in 2010, but the district had information regarding magnet schools and home schooling. Pajaro Valley reported that it had charter schools, some of which participate in the school meals programs. The school food authority direc- tor was not sure whether there were other charters (outside her school district) in the area. She reported that she has participation data for any schools that are part of her program but does not have any data on stu- dents that may be home schooled or attending schools outside of the district. PART 2: AGENDA FOR WORKSHOP WITH SCHOOL FOOD SERVICE AUTHORITY DIRECTORS EIGHTH PANEL MEETING WORKSHOP MARCH 3-4, 2009 The Panel on Estimating Children Eligible for School Nutrition Pro- grams Using the ACS hosted a workshop in Washington, DC, on March 3-4, 2011. Participants included school food service authority directors from the five case study districts and selected other individuals from the school food community who could provide insights about Provision 2 and the school meals programs more generally. The purpose of the workshop was to help the panel better understand issues pertaining to a potential new provision of the school meals programs, as well as the information school districts need to determine whether to adopt this special provision. The workshop was held at the 20 F Street Conference Center, 20 F Street, NW, Washington, DC.

OCR for page 286
308 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS Agenda: March 3, 2011 9:00 AM Welcome to Workshop, Introduction of Participants 9:15 Welcome to the National Academy of Sciences Connie Citro, director, Committee on National Statistics, National Academy of Sciences 9:30 Introduction to the Panel on Estimating Children Eligible for School Nutrition Programs Using the ACS 10:00-10:15Break 10:15 AM- 12:00 PM The Policy and Program Context: Traditional Operating Procedures and Old and New Special Provisions Panelists: Lynn Harvey (North Carolina), Tammy Yarmon (Omaha), Leo Lesh (Denver), Nicole Meschi (Pajaro Valley) What value do you see in special provisions such as Pro- vision 2 or the potential new provision? What challenges do you see in them? Describe the administrative burdens associated with the first year of Provision 2 and with subsequent base years. What do districts need to know to help them decide to participate in Provision 2? What do you think they will need to know to help them decide to participate in the potential new provision? How would you decide between Provision 2 and the potential new provision? The panel has observed that many districts elect to use Provision 2 for breakfast only. Why might that be true? Would you consider using the new provision for break- fast only? In current practice, the panel observed that most Provi- sion 2 schools are elementary schools. Do you think that this is the case in general? Why or why not? What do you view as the most promising benefit of the new provision? What are your greatest concerns based on what you have heard about the potential new provision? How do you think these concerns could be addressed?

OCR for page 286
APPENDIX E 309 12:00-1:00 Working Luncheon 1:00-2:30 The Effects of Offering Universal Free Meals: Participation and Costs Panelists: Leo Lesh (Denver), Tim Cipriano (New Haven), Terry Mendez (Brownsville), Lyman Graham (Roswell), Mary Jo Tuckwell (inTEAM Associates) How would offering universal free meals affect the total number of meals served? How would it affect participa- tion by category? Have you seen any data on this? Are costs fixed, or are there economies of scale? If the number of meals served goes up, what is the marginal cost per additional meal? Texas has developed a table (to be provided at the work- shop) showing expected changes in participation under various situations for Provision 2. Are there other fea- tures that should be included in this kind of calculator? The panel has not found much information for estimat- ing cost savings from not having to do applications and verifications, direct certification, or counting meals by cat- egory. Do you have a rough estimate of the cost savings in your district? How important are these cost savings? Are there advantages to elimination of applications and verifications other than cost savings? Are there cost sav- ings associated with not doing direct certification? Is this different for a district that is entirely on Provision 2 versus one that is on for only a school or group of schools? If there are no applications, where would you get the information for other benefits, such as waivers of text- book or athletic fees for students eligible for free or reduced-price meals? 2:30-2:45Break 2:45-4:15 Dealing with Uncertainty and Variability Panelists: Lynn Harvey (North Carolina), Tammy Yarmon (Omaha), Tim Cipriano (New Haven), Terry Mendez (Brownsville) Provision 2 must be implemented at the beginning of a school year, unless the school has delayed implemen- tation; then it is implemented in the second claiming

OCR for page 286
310 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS period. Should the new provision also be implemented at the beginning of a school year? Is there any reason you can think of to allow for delayed implementation? By what date in a school year do you need to know your claiming rates for school meals? How much variation in claiming rates can you toler- ate from year to year? There are many ways to estimate claiming rates. Some estimators rely on more recent data and are more timely; that is, they better reflect more recent economic conditions. However, they may be sta- tistically less precise than other estimators that use data from not only the most recent year but also several pre- vious years. With either type of estimator, it would be possible to fix claiming rates for several years and update them only every few years. This would eliminate year- to-year changes in claiming rates--except in updating years--at the risk of the rates becoming "out of date." How do you assess the potential trade-off between vari- ability and timeliness? 4:15-5:30 Deciding to Implement Districtwide or at a Subdistrict Level (e.g., for groups of schools) Panelists: Tammy Yarmon (Omaha), Nicole Meschi (Pajaro Valley), Lyman Graham (Roswell), Onetha Bonaparte (Chatham), Mary Jo Tuckwell (inTEAM Associates) What are factors that influence the decision to implement for the entire district versus for a group of schools? How would you determine which schools should get universal free meals? American Community Survey estimates will be for stu- dents living in specified school attendance areas (as opposed to attending specified schools). Charter schools, magnet schools, and open enrollment draw students from neighborhood schools. -- How many nonneighborhood schools are in your dis- trict? What percentage of district students attend non- neighborhood schools? -- Do you have data concerning the number of students who attend nonneighborhood schools and the neighbor- hood schools to which they would have been assigned? With these data, the panel could compute free and reduced-price percentages for the open enroll-

OCR for page 286
APPENDIX E 311 ment schools based on an assumption that free and reduced-price students attend open enrollment schools and neighborhood schools in the same per- centages as those eligible only for full-price meals. Can you comment on the reasonableness of this assumption in your experience? --ACS estimates for charter/magnet/some open enroll- ment schools will not be available (unless estimated as above) because the schools do not have distinct attendance areas. Hence the new provision may not be applicable for them separately. Would this be an issue for your district? Agenda: March 4, 2011 8:45 AM Welcome to Workshop, Introduction of Participants 9:00-10:30Geography School Attendance Boundary Information System (SABINS) Collaborator: Sal Saporito Panelists: Onetha Boneparte (Chatham), Tim Cipriano (New Haven), Nicole Meschi (Pajaro Valley), Terry Mendez (Brownsville), Lyman Graham (Roswell) To implement the new provision at a subdistrict level (for a school or group of schools), the district will need to provide the Census Bureau with geographical boundaries for the school attendance areas. These digital boundaries must be in a format that the Census Bureau can use easily. Options for the district include: using (or working with) SABINS to obtain boundaries, using software provided by SABINS to obtain data, or using some other method to identify the unique census blocks that make up a school attendance area. This session will begin with introductory information from Mike and Sal. What problems do you foresee in providing the Census Bureau with boundary information? Which of the alternative methods of boundary definition would best fit with your district's operations? What forms of collaboration between districts and with SABINS could be of help to you?

OCR for page 286
312 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS How frequently would you need to update your informa- tion because of changes in school attendance boundaries? Can you comment on what might be reasonable costs to the districts if there are costs for obtaining boundaries or costs for obtaining tabulations from the Census Bureau? 10:30-10:45Break 10:45 AM- 12:30 PM The Process and Calculus of Decision Making: Evaluating the Attractiveness of a New Special Provision Panelists: Lynn Harvey (North Carolina), Leo Lesh (Denver), Onetha Boneparte (Chatham), Mary Jo Tuckwell (inTEAM Associates) How would your district/state make a decision whether to adopt a new special provision? What information do you need to make this decision? What information on variability in reimbursements would be most useful to you? For example, the panel can provide examples of blended reimbursement rates (average reimbursement per meal) for several years for different estimation methods. Would that be useful? Would estimates of statistical uncertainty--for example, the margin of error for your average reimbursement rate--be useful to you? What is your view on using eligibility rates as claiming rates? Do you think claiming rates based on eligibility should be adjusted to reflect participation? Do you have ideas about how this might be done? Eligibility rates from the ACS will be based on children living in normal housing. Examples of students who do not live in normal housing include the homeless and some migrants. The panel is considering whether it is possible to include these students using local data and an adjustment. Do you know of other examples of students in your dis- trict who do not live in normal housing? Do you know the number of students in your district who do not live in normal housing, and which schools they attend?

OCR for page 286
APPENDIX E 313 Can you comment on the National School Lunch Program (NSLP) and School Breakfast Program (SBP) eligibility status of these children? What do you think your colleagues would like to see in our report that would help them decide whether to use the potential new special provision? Invited Panelists Onetha Bonaparte, school meals program coordinator, Savannah- Chatham County Public School System, Georgia Tim Cipriano, executive director of food services, New Haven Public Schools, Connecticut Lyman Graham, foodservice director, Roswell Independent Public School District, New Mexico Mary Kate Harrison, general manager, Student Nutrition Services, Hillsborough County Public School District, Florida Lynn Harvey, section chief, child nutrition services, Division of School Support, North Carolina Department of Public Instruction, North Carolina Leo Lesh, executive director, Enterprise Management, Denver Public Schools, Colorado Terry Mendez, administrator for food and nutrition services, Brownsville Independent School District, Texas Nicole Meschi, director of food and nutrition services, Pajaro Valley Unified School District, California Mary Jo Tuckwell, senior consultant, Food Services Group, inTEAM Associates, Wisconsin Tammy Yarmon, director, Nutrition Services, Omaha Public Schools, Nebraska PART 3: SURVEY OF DISTRICTS OPERATING UNDER PROVISION 2 OR 3 The panel conducted a survey of school food authority directors in school districts that have participated in Provision 2 or 3. The purpose was to identify advantages and disadvantages of these provisions from their point of view and to determine whether the respondents had data they were willing to share that would help the panel identify changes in par- ticipation due to providing universal free meals. This part of the appendix provides detail on the pilot test and methodology for this survey.

OCR for page 286
314 USING ACS DATA TO EXPAND ACCESS TO THE SCHOOL MEALS PROGRAMS Pilot Test The first phase of the survey was a pilot test implemented via a telephone survey that was used to obtain preliminary information and test the questionnaire. The panel obtained the School Nutrition Associa- tion (SNA) profiles of participants at their 2010 School Nutrition Asso- ciation Legislative Action Conference. The profile included information about the school district and whether it participated in Provision 1, 2, or 3, as well as the school food authority director's contact information (excluding e-mail address). For the school districts on the SNA list, the panel added data concerning provision status for the past 5 years from the FNS-742 form to the SNA database. Only 16 of the 39 names on the SNA list were from school districts in the FNS-742 database that reported operating under Provision 2 or 3. The panel selected those school districts with FNS-742 provision status in no more than 4 of the 5 school years (hoping to capture school districts with a recent base year). This resulted in a list of 1211 potential candidates for the pilot data collection. (The 4 that were not selected reported operating under a provision [not in a base year] for each of the 5 years.) E-mail addresses for the selected school districts were found via an Internet search, and an e-mail invitation to participate in the telephone survey was sent. If the school food authority director responded positively, the telephone interview was scheduled, and the interview was conducted by a panel member. Ten of the 12 school food authority directors completed an interview. Of the 10, 1 had not implemented any special provisions, 3 were operating under Provision 2 for breakfast only, and 6 were operating under Provi- sion 2 for both lunch and breakfast (1 districtwide). Five directors stated that they had data they could provide (2 for breakfast only). The number of schools in these districts ranged from 10 to 140, with an average of 41. Enrollment ranged from 5,400 to 89,000, with an average of 30,000. Internet Survey The panel prepared the Provision Database, consisting of all school districts reporting on the FNS-742 that some of their schools operated under Provision 2 or 3 (not in a base year) for 1 to 4 of school years 2004- 2005, 2005-2006, 2006-2007, 2007-2008, and 2008-2009. These individual school year files were merged by district to form a longitudinal database of districts that reported participating in Provision 2 or 3 in any year. The panel decided to include only school districts with more than 500 students 11Of these 12 school districts, 4 reported that they were on provision status (not in a base year) for 4 years, 2 reported 3 years, 1 reported 2 years, and 4 reported only 1 year; 1 reported to SNA operating under Provisions 1, 2, and 3.

OCR for page 286
APPENDIX E 315 that reported participating in Provision 2 or 3 (not in a base year) during from 1 to 4 of the past 5 years. There were 287 districts that met these cri- teria. The panel worked with FNS to obtain contact information for these districts. Working with its regional offices, FNS provided e-mail addresses for 100 of these districts, each of which was sent a survey questionnaire via SurveyMonkey. Twenty-two districts completed the Internet survey. Of the 22 Internet survey respondents, 1 had not implemented a special provision and was out of scope. The number of schools in these districts ranged from 2 to 90, with an average of about 16. Enrollment ranged from 1,100 to 49,000, with an average of about 8,300. Most of these districts reported operating under Provision 2. One district reported operating under Provision 2 in the past, but could no longer afford to participate because of district finances. One school reported operat- ing under Provision 2 for breakfast only. The others all reported operating under Provision 2 for both breakfast and lunch. Eleven reported having implemented Provision 2 districtwide.