ATTACHMENT_I.pdf

ATTACHMENT I:

PERFORMANCE WORK STATEMENT

USING DATA MATCHING IN THE SCHOOL MEALS

ELIGIBILITY AND VERIFICATION PROCESS

TABLEOF CONTENTS

1 GENERAL OVERVIEW ................................................................................................................... …2

2 BACKGROUND

2.1 The National School Lunch Program (NSLP) and School Breakfast Program (SBP)

2.2 Computer Matching in Nutrition Assistance Programs

2.2.1 Computer Matching in the NSLP………………………………………………..…3

2.2.2 Computer Matching in the Special Nutrition Assistance Program (SNAP)……

2.2.3 Computer Matching in the Special Supplemental Nutrition Program for Women, Infants, and Children

(WIC)…………………………………………………………..................................…….5

3 ISSUES AND CHALLENGES FOR COMPUTER MATCHING IN THE NSLP

3.1 Potential of Computer Matching

3.2 Obstacles to Computer Matching

4 SCOPE OF WORK

4.1 Study Task, Deliverables, and Research Questions……………………… ...................... …6

4.2 Task 1……………………………………………………………… ............................ ……6

5 STUDY DESIGN AND METHODOLOGICAL ISSUES……………………… .......................... …11

5.1 Design Overview……………………………………………………………

5.2 Design Parameters and Constraints…………………………………………

PERFORMANCE WORK STATEMENT

USING DATA MATCHING IN THE SCHOOL MEALS

ELIGIBILITY AND VERIFICATION PROCESS

1 GENERAL OVERVIEW

A recent Government Accountability Office report1 recommended that USDA explore electronically matching household-application information to other data sources—such as State income databases or public-assistance databases– to verify the accuracy and improve the certification process. This project would update and expand previous USDA work in this area to determine if data systems and datasets (e.g., Medicaid or Unemployment Insurance) can be linked to application information in a manner that supports timely and accessible certifications and used as the basis for verification for cause and other error-reduction strategies. Promising approaches identified through this review, if any, may be piloted under a separate contract in a limited number of States and Local Educational Authorities (LEAs) to test their feasibility, as well as their impact on program participation and improper payments

2 BACKGROUND

2.1 The National School Lunch Program (NSLP) and School Breakfast Program (SBP)

The NSLP is a federally assisted meal program operating in almost 100,000 public and non‐profit private schools and residential child care institutions. FNS administers the program at the Federal level. At the State level, the NSLP is usually administered by State education agencies, which operate the program through agreements with school food authorities (SFAs). School districts that participate in NSLP receive cash subsidies and donated commodities (USDA Foods) from USDA for each meal they serve. The cash subsidies are much larger for those meals served to children certified for free or reduced-price meals.

Any child enrolled in a participating school may purchase a meal through the NSLP. Children from families with incomes at or below 130 percent of the poverty level are eligible for free meals. Children from families with incomes between 130 percent and 185 percent of poverty are eligible for reduced-price meals (no more than 40 cents). For the period July 1, 2014, through June 30, 2015, 130 percent of the poverty level is $31,005 annually for a family of four; 185 percent is $44,123 annually. Children from families with incomes over 185 percent of poverty pay for full-price (paid) meals at rates established by local school food authorities. In 2014, more than 28 million children each day received a nutritionally balanced meal. More than 70 percent of these meals were served to low-income children certified to receive free or reduced-price meals.

The School Breakfast Program (SBP) is a federally assisted meal program administered by the FNS that is operating in almost 90,000 public and nonprofit private schools and residential child care institutions. It began as a pilot project in 1966, and was made permanent in 1975. In 2014, more than 12 million children received a school breakfast each day. The SBP uses SNLP certification to determine meal eligibility

1 Government Accountability Office, School Meals Programs, USDA Has Enhanced Controls, but Additional Verification Could Help Ensure Legitimate Program Access, May 2014.

(free/reduced-price/paid). More than 80 percent of these meals were served to low-income children certified for free or reduced-price meals.

Links to fact sheets about these various school-based child nutrition programs are provided below:

NSLP: http://www.fns.usda.gov/cnd/Lunch/AboutLunch/NSLPFactSheet.pdf

SBP: http://www.fns.usda.gov/sites/default/files/SBPfactsheet.pdf

Currently there are two main methods by which students are certified for NSLP/SBP free or reduced-price meals: by application or direct certification. Households can submit applications for meal benefits based on participation in the Special Nutrition Assistance Program (SNAP), the Temporary Assistance to Needy Families (TANF), or the Food Distribution Program o Indian Reservations (FDPIR) (categorical eligibility) or based on household income and household size (income eligibility). Households are required to self-report a case number reflecting participation in SNAP, TANF, or FDPIR or information on household size and household income. While most households submit applications at the beginning of the school year for annual meal benefits, SFAs continue to accept applications throughout the year.

Direct certification is mandated for students who are categorically eligible for free meals based on their enrollment in SNAP. States and local agencies have discretion to also directly certify students who are categorically eligible for free meals based on their enrollment in TANF or FDPIR. Children who are directly certified for free meals do not have to complete an application and are not subject to income verification.

2.2 Computer Matching in Nutrition Assistance Programs

Computer matching refers to the computerized comparison of records for the purpose of establishing or verifying eligibility for a federal benefit program or for recouping payments or delinquent debts under such programs. The Computer Matching and Privacy Protection Act of 1988 (P.L. 100-503) which became effective on December 31, 1989 amended the Privacy Act to add several new provisions to regulate the use of computer matching by federal agencies. These provisions added procedural requirements for agencies to follow when engaging in computer-matching activities; provides matching subjects with opportunities to receive notice and to refute adverse information before having a benefit denied or terminated; and requires that agencies engaged in matching activities establish Data Protection Boards to oversee those activities. Generally, matching programs involving federal records must be conducted under a matching agreement between the source and recipient agencies. The matching agreement describes the purpose and procedures of the matching and establishes protections for matching records. The agreement is subject to review and approval by a Data Integrity Board.

Computer matching is currently used in several of the largest nutrition assistance programs, including the NSLP. Uses for computer matching include verification of income, assets, and other eligibility criteria, obtaining health information about participants, identifying duplicate program participation in different States, and verifying participation in other assistance programs. The most common use of computer matching in the NSLP is direct certification in which school children are determined to also participate in either SNAP or TANF.

http://www.fns.usda.gov/cnd/Lunch/AboutLunch/NSLPFactSheet.pdf http://www.fns.usda.gov/sites/default/files/SBPfactsheet.pdf

2.2.1 Computer Matching in the NSLP

Direct certification and direct verification provide two opportunities to use computer matching to streamline NSLP operations. Direct certification increases access to school meals for eligible students while decreasing the administrative burden on families and school district staff by limiting the amount of paperwork they must complete and process. State Agencies conduct direct certification by matching school enrollment data with data from SNAP or other programs that confer categorical eligibility to students. Direct certification can be accomplished by State-level computer matching or district-level matching. Under State-level matching, the State agency matches school enrollment data with SNAP, TANF, and/or other means-tested programs’ participation data. Under district-level matching, the State agency provides the local school district with records of SNAP/TANF children residing in the school districts geographic area and the district uses computer matching or manual matching to identify children in SNAP/TANF households who are enrolled in the district.

The Healthy, Hunger-Free Kids Act of 2010 requires that States meet specific performance targets in their direct certification systems. In school year (SY) 2012-13, 80 percent of the school-age children receiving SNAP benefits were expected to be directly certified. State agencies that did not meet the performance targets are required to develop and implement a Continuous Improvement Plan. Beginning in SY 2013- 14, the direct certification performance targets increased to 95 percent for all States. FNS annually reports to Congress on the State-specific rates of direct certification using SNAP data. A link to the latest annual report is provided here:

(http://www.fns.usda.gov/sites/default/files/ops/NSLPDirectCertification2014.pdf ).

NSLP regulations require Local Education Agencies (LEAs) to verify a sample of NSLP applications.

LEAs must verify a sample of error-prone applications equal to 3 percent of all approved applications (up to a maximum of 3,000 applications). Error-prone applications are income applications with monthly income within $100 of the free or reduced-price eligibility limit. LEAs typically request that sampled households provide documentation to support the current NSLP eligibility. Failure to respond to the verification request or providing documentation of income in excess of NSLP eligibility limits results in a reduction in meal benefits. LEAs may also directly verify sampled applications by utilizing income and program participation information from a public agency administering SNAP, TANF, FDPIR, State Medicaid program, or other means-tested programs. The Child Nutrition and WIC Reauthorization Act of 2004 (P.L. 108-265) authorized direct verification with Medicaid and State Children’s Health Insurance Program (SCHIP) data. Direct verification of SNAP case numbers reported on applications is already done by many LEAs. Direct verification of income applications uses income information collected by other means-tested programs to verify NSLP income eligibility. Electronic wage records are not currently used to verify income applications.

2.2.2 Computer Matching in the Special Nutrition Assistance Program (SNAP)

Computer matching is used to improve integrity in the SNAP. Indeed, Federal law requires SNAP eligibility workers to conduct computer matching with certain data files. An FNS-sponsored study by Mathematica Policy Research, Inc., An Assessment of Computer Matching in the Food Stamp Program, found that most States use computer matches of SNAP caseload data against more than 10 different systems. National and State data bases used by the majority of States include the following:

http://www.fns.usda.gov/sites/default/files/ops/NSLPDirectCertification2014.pdf

• State Wage Information Collection Agency,

• State Data Exchange System,

• Unemployment Insurance,

• Social Security’s Beneficiary Data Exchange,

• Social Security Administration Prisoner Verification System,

• Social Security Administration death records,

• Department of Motor Vehicles,

• the Internal Revenue Service Beneficiary Earnings Exchange Reports System,

• the Federal Disqualified Recipient System,

• Social Security quarterly earnings file,

• the Internal Revenue Service Systematic Alien Verification System, and

• State databases on recent hires.

The study showed that technological advances, particularly the growth in communications networks, have greatly improved State capabilities for running computer matches. These advances have led to much more rapid responses from external databases, allowing SNAP caseworkers to initiate queries in real time rather than wait for routine batch match results. However, States did indicate that certain data files were not helpful because data retrieved was outdated or erroneous.

2.2.3 Computer Matching in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC)

WIC, like SNAP, typically maintains statewide computer systems that are updated in real-time. However, there are significant differences between SNAP’s and WIC’s computer systems because of differences in program characteristics and regulatory requirements. WIC enrolls individuals and provides various services to them, so most WIC systems rely on relational databases to link individual certification records to other information such as food package prescriptions, voucher data, and nutritional education appointments. A link to a fact sheet about WIC is provided here:

http://www.fns.usda.gov/sites/default/files/wic/WIC-Fact-Sheet.pdf.

WIC also uses computer matches to detect dual participation in both WIC and the Commodity Supplemental Food Program or in more than one State WIC program. These computer matches may be conducted in batch format. Finally, computer matches between WIC files and vital record information is often used to conduct research on WIC.

Features of many State WIC data files preclude routine use of most computer matches. Specifically, because there is no regulatory requirement to collect Social Security Numbers (SSNs) for WIC participants, many WIC programs do not collect this information. There is little or no data integration between SNAP, WIC, and NSLP. While many States include master client indexes linking records of social service programs, including SNAP, only a few States include WIC in these indexes. FNS is partnering with a Midwest State to test the feasibility of matching WIC applicant records against the State’s Medicaid database at the time of WIC certification. This exploratory project takes advantage of that state’s use of a common participant ID across several public assistance databases.

http://www.fns.usda.gov/sites/default/files/wic/WIC-Fact-Sheet.pdf

3 ISSUES AND CHALLENGES FOR COMPUTER MATCHING IN THE NSLP

3.1 Potential of Computer Matching

Computer matching has the potential to improve program integrity without lowering the probability that eligible households are approved for free or reduced-price school meals in three ways:

1. Computer matching has the potential to prevent ineligible households from being approved by detecting their ineligibility at the time of application. Computer matching could be used to screen household applications to, in part or in whole, assess the relationship between information provided on the self-reported application and other salient data about the eligibility of applicant households.

2. Computer matching has the potential to enhance or replace the current income verification process, including verification for cause, and lower the rate of continued certification among ineligible households.

3. Computer matching has the potential to increase the rate of free and reduced-price approval among eligible children by enhancing the coverage rate of the direct certification process using information from additional means-tested programs.

3.2 Obstacles to Computer Matching

With computer matching, a portion of the burden of verification is shifted from the household to the parties conducting the computer matches. However, a number of obstacles that limit the feasibility and impact of computer matching must be addressed.

1. There may not be a common and unique identifying variable for data linking. Section 301 of the Healthy, Hunger-Free Kids Act of 2010 amended the Richard B. Russell National School Lunch Program Act to only require the last four digits for one adult member of the household on the NSLP free and reduced price meal application. Moreover, some households may not provide a Social Security number for the adult completing the application, but rather indicate that they have no Social Security number. In some cases they may provide a Social Security number for a child in the household, a Social Security number belonging to a non-household member, or even a fraudulent Social Security number. This is a particular concern among families headed by undocumented immigrants.

2. Data matching may not produce accurate and usable data. Many types of earnings will not show up in wage data files. For example, in verifying household income, many households earn money off the books doing services such as carpentry and construction, childcare, housekeeping, or restaurant work.

Moreover, earnings data on these files is frequently outdated. According to SIPP data, low-income families move in and out of free or reduced-price school meal eligibility frequently. This poses a challenge to the usefulness of six- or even three-month old wage information on the data files if the data is suppose to represent their household income at the time of application.

3. Privacy concerns must be respected. It is uncertain that FNS or LEAs even have legal authority to conduct computer matches. For direct verification, the National School Lunch Act allows LEAs to use information from programs other than SNAP, FDPIR, TANF, and Medicaid (Sec. 9(b)(3)(F)).

However, the Federal/State Agencies administering those programs may not allow it. The language on direct certification seems more restrictive. For example, is it sufficient to add language on the free and reduced price application form authorizing use of the data for computer matches?

4 SCOPE OF WORK

The purpose of this project is to update and expand previous USDA work in the area of data matching in the NSLP to determine if data systems and datasets (e.g., Medicaid or Unemployment Insurance) can be linked to application information in a manner that supports timely and accessible certifications and used as the basis for verification for cause and other error-reduction strategies. Promising approaches identified through this review, if any, may be piloted under a separate contract in a limited number of States and Local Educational Authorities (LEAs) to test their feasibility, as well as their impact on program participation and improper payments.

4.1 Tasks /Deliverables

The contractor shall furnish all necessary personnel, materials, services, facilities, and otherwise do all things necessary to execute the scope of work for this project. In performance of this contract, the contractor shall complete the tasks set out below. While the Government has established the major tasks to be delivered, the Contractor shall propose intermediary tasks for each major task to allow for collaboration and FNS input. The information gathering for these tasks may take many forms, including but not limited to, a literature review including an examination of data matching implementation at other Federal Agencies, input from a panel of experts, focus groups/ exploratory interviews or case studies with a small number of State Agencies and/or LEAs. In the performance of this contract, USDA expects to have opportunities for full engagement in the review and comment on all important and significant information gathering activities as well as the review and comment on the various iterations (intermediate tasks).

Through this project, FNS hopes to better understand the feasibility for States and school districts to use computer matching of wage records, means-tested program information, and other data sources as a tool for determining and verifying the eligibility of households with school age children for free and reduced-price school meals. FNS is seeking information on current data matching practices at the State and district levels, to delineate and understand the issues that make data matching in the NSLP challenging (including, but not limited to, those listed above). We seek to identify potential promising approaches, if any, through this review, that could be piloted, under a separate contract, in a limited number of States and Local Educational Authorities (LEAs) to test their feasibility, as well as their impact on program participation and improper payments.

4.2 Task 1: Prepare a White Paper Examining the Current Information Technology Environment in State Child Nutrition Agencies and School Districts to Improve Program Integrity and Oversight of the School Meals Programs through Data Matching

The last FNS comprehensive study of data matching in the National School Lunch Program occurred a decade ago. It is not the intent of this study to reproduce results of the Data Matching in the National School Lunch Program: 2005 (http://www.fns.usda.gov/data-matching-national-school-lunch-program- 2005) by conducting a census of State Child Nutrition Agencies. Utilizing a variety of sources, the contactor shall collect information to determine if the technology environments for State Agencies and LEAs have changed since the last comprehensive study. The contractor may wish to draw on experiences of State staff from State Agencies other than the one housing the Child Nutrition Division (e.g., State Department of Health/Labor, etc.) The White Paper shall synthesize the findings related to the operational feasibility of computer matching in the NSLP. The White Paper shall identify promising approaches to data matching for both eligibility determination and eligibility verification in the NSLP, if any, and identify viable candidates for pilot testing the feasibility of these approaches.

Task/ Deliverable 1.1 Draft White Paper

Task/Deliverable 1.2 Revised White Paper

Task/Deliverable 1.3 Final White Paper

Research Questions

• How are State Child Nutrition Agencies/LEAs currently using computer matches for the NSLP?

What databases are being used? For what specific purposes are computer matches being used? Are they being used for direct certification and/or direct verification? Is computer matching being used as the basis for verification for cause and other error-reduction strategies? Are categorical applications being verified through data matching? What are the challenges that States experience? What are the benefits that these computer matches bring? How effective/cost-effective are State computer matching efforts (by type of database)? Have States developed or tested matching processes that they concluded were ineffective or cost-inefficient and no longer use? What future uses of data matching are being envisioned by State Agencies and what databases will be used?

• Why do States and LEAs not use computer matching more extensively? Why don’t those that do matching use more databases, such as wage and earnings files? What are the obstacles to successful computer matching? Are there state polices with regard to computer matching and data sharing which inhibit computer matching in NSLP? Are there steps that FNS, States, or LEAs can take to make computer matching more feasible?

• Do computer matches provide a real, meaningful potential to improve certification and/or verification accuracy in the NSLP without increasing household burden or reducing participation among eligible children? Can meaningful gains in certification and/or verification accuracy through additional data matching be achieved at reasonable cost?

http://www.fns.usda.gov/data-matching-national-school-lunch-program-2005 http://www.fns.usda.gov/data-matching-national-school-lunch-program-2005

• Are there other means-tested programs (besides SNAP, TANF, FDPIR, and Medicaid) that could be used to directly certify children for free meals? Could these other means-tested programs be used to establish that household income is within key ranges to qualify for reduced-price meals? Are there specific States that the feasibility of using these other means-tested programs could be examined?

• Is matching with wage or earning data feasible? How complete, accurate, and timely are the data?

What data elements are needed to conduct a reliable wage match? How much follow-up is needed for the income verification process?

• What legal and privacy issues must be addressed in the development of computer matching systems?

After review of the White Paper, FNS may wish to determine the feasibility of any promising data matching approaches. Under a separate contract, the feasibility of implementing these promising approaches would be examined and the impact of these approaches on eligibility determination and verification results would be evaluated. To assist in framing the parameters of the White Paper, FNS is offering the type of research questions that may be asked of the data matching pilots.

For promising computer matching approaches designed to identify students eligible for free or reduced-price meals:

• What challenges were encountered in implementing the data match? How were each of these challenges resolved?

• How many individual students (number and percentage) were matched at the State level? At the local school district level? At both? How many of the individual matched students (number and percentage) were already directly certified for free meals using SNAP/TANF/FDPIR data? How often did this matching process occur?

• How does the quality/certainty of the matches compare to matches using SNAP/TANF/FDPIR? How is that quality/certainty assessed? Does the new matching process generate relatively more questionable matches than matching against SNAP/TANF/FDPIR datasets?

• How do the definitions of household and household income compare with those used in determining eligibility for NSLP? What was the gap between the date of household size and household income determination in other means-tested programs and the date of determination of eligibility for free/reduced-price meals? How did this gap differ among different States and LEAs?

• How much staff time was required by State and LEA employees to complete the match? How did the staff time differ by States and LEAs? What lead to particularly large and small staff time burdens?

Are these staff time burdens likely to decrease over time if the new match process is adopted permanently?

• How did success in matching vary:

o By State and school district characteristics (including, but not limited to, urban/rural, higher vs. lower percentage of free/reduced-price students, State and local data systems, levels of SNAP direct certification as a percentage of free certification? or o By recipient characteristics (including but not limited to race, ethnicity, family/household size and composition, name differences of members of the family/household)?

For promising computer matching approaches designed to improve the direct verification process:

• What challenges were encountered in implementing the data match? How were each of these challenges resolved?

• Were procedures implemented to verify the eligibility of categorically eligible applicants? If so, how was this computer match conducted? What percentage of categorically eligible students could not be matched to SNAP case numbers? To TANF case numbers? When was this match conducted? During certification determination?

• Were procedures implemented to identify potential ineligible households by matching household income to wage and earnings data bases? How was this match conducted? What data bases were used? What identifying information was used in the matching process? How much follow-up was required? How many individual students (number and percentage) of the verification sample could be matched to wage and earnings data? How many individual students (number and percentage) of the verification sample were determined to be ineligible for the meal benefits they were receiving?

• How much staff time was required by State and LEA employees to complete the match? How did the staff time differ by States and LEAs? What lead to particularly large and small staff time burdens?

• How did success in matching vary:

o By State and school district characteristics (including, but not limited to, urban/rural, higher vs. lower percentage of free/reduced-price students, State and local data systems, levels of SNAP direct certification as a percentage of free certification? or o By recipient characteristics (including but not limited to race, ethnicity, family/household size and composition, name differences of members of the family/household)?

References

Cole, Nancy (2003) Feasibility and Accuracy of Record Linkage to Estimate Multiple Program Participation: Volume 1: Record Linkage Issues and Results of the Survey of Food Assistance Information Systems. (E-FAN-03-008-1) Washington, DC: USDA, Economic Research Service.

Cole, Nancy and Chris Logan (2006). Data Matching in the National School Lunch Program: 2005.

Volume 1: Final Report. Project Officer Jenny Laster Genser. USDA, Food and Nutrition Service, Alexandria, VA.

Cole, Nancy and Chris Logan (2006). Data Matching in the National School Lunch Program: 2005.

Volume 2: Select Case Studies. Project Officer Jenny Laster Genser. USDA, Food and Nutrition Service, Alexandria, VA.

General Accounting Office (GAO) ((2014). School Meals Programs: USDA Has Enhanced Controls, but Additional Verification Could Help Ensure Legitimate Program Access. Washington, DC: U.S.

General Accounting Offioce, GAO-14-262

Gleason, Philip, Tania Tasse, Kenneth Jackson, and Patricia Nemeth (2003). Direct Certification in the National School Lunch Program: Impacts on Program Access and Integrity, Final Report. Washington, DC: USDA, Economic Research Service, EFAN-3-009.

Gotho, A., Moore, Q, Conway, K., Kyler, B. (2014). National School Lunch Program Direct Certification Improvement Study: State Practices and Performance Report. Project Officer: Joseph F. Robare. USDA, Food and Nutrition Service, Alexandria, VA.

Hulsey, L., Gordon, A., Leftin, J., et. al (2015). Evaluation of Demonstrations of National School Lunch Program and School Breakfast Program Direct Certification of Children Receiving Medicaid Benefits: Access Evaluation Report. Project Officer: Allison Magness. USDA, Food and Nutrition Service, Alexandria, VA.

Jackson, Kenneth, Philip Gleason, John Hall, and Rhoda Strauss (2000). Study of Direct Certification in the National School Lunch Program. Project Officer, Matthew Sinn. USDA, Food and Nutrition Service, Alexandria, VA.

Logan, Chris, Nancy Cole, and David Hoaglin. Direct Verification Pilot Study Final Report (2009).

Project Office: Sheku Kamara. USDA, Food and Nutrition Service, Alexandria, VA. CN-09-DV2.

USDA, Food and Nutrition Service (2002). An Assessment of Computer Matching in the Food Stamp Program: Volume I Summary of Survey Results. Final Report. By William S. Borden and Robbi Ruben- Urm, Alexandria, VA.

5 STUDY DESIGN AND METHODOLOGICAL ISSUES

5.1 Design Overview

The contract will be awarded in Fiscal Year 2015. Information gathering to determine if the technology environments for State Child Nutrition Agencies and LEAs have changed since the last comprehensive study examining computer matching in the NSLP shall occur in 2015/16. The White Paper that synthesizes the findings related to the operational feasibility of computer matching in the NSLP shall be completed within nine (9) months of contract award. Contractor shall include in their proposal a clear study plan detailing how they plan to gather data and synthesize the results in the White Paper to address the research questions in Task 1: Preparation of a White Paper. It is not the intent of this study to reproduce results found in the Data Matching in the National School Lunch Program: 2005 by conducting a census of State Child Nutrition Agencies. If promising approaches are identified, the contractor shall identify viable State Agencies/LEAs that the feasibility of these approaches could be pilot tested.

5.2 Design Parameters and Constraints

• To conserve limited study resources, the offeror shall use existing data collection instruments to the maximum extent feasible. If new instruments are deemed necessary, the contractor shall note that any data collection with more than nine (9) respondents shall require approval under terms of the Paperwork Reduction Act, described below. Based on time constraints, FNS does not anticipate there is adequate time for the approval process.

• If the contractor is proposing to collect information from more than nine (9) respondents, the contractor shall prepare the required publication notices and clearance package for submittal to the Office of Management and Budget (OMB) to obtain approval for all data collection activities in all components of the evaluation subject to the requirements of the Paperwork Reduction Act. The clearance package must provide an explicit, concise description of the direct links between the tasks, research questions, variables, instrument items, data analysis plans and desired products. The PRA data collection package shall contain copies of all final data collection instruments and a supporting statement as set forth in the revised Standard Form No. 83a, “Instructions for Requesting OMB Approval under the Federal Reports Act”, as Amended.

• Contractor is advised that obtaining approval for data collections under the Paperwork Reduction Act can be a lengthy process. The Offeror should be aware that the OMB data collection package will be reviewed by multiple groups within USDA. Revisions to the package may be required after each level of review and the package will not be considered complete until it receives official OMB clearance. Past experience has shown that it can take six to nine months from the first submission of a 60-day Federal Register notice and up to fifty-two weeks from the start of the process to final approval of the information collection. Vendors are advised that clearance is often the primary rate limiting factor in many studies. Preparation of a high-quality initial draft of instruments and justification statements, followed by timely responses to required revisions, will greatly facilitate the approval process. For additional information — including detailed guidance, a checklist for final collection request submissions, and estimated timelines, see the following link

(www.ocio.usda.gov/policy-directives-records-forms/information-collection). Offerors are advised that clearance package formats that have worked at other Federal agencies may be rejected by USDA.

• Full conformance with the OMB standards and guidelines for surveys is required. Contractor shall note that an 80 percent survey response rate is required; if response rates are less, a nonresponse bias analysis and correction if bias is found are mandatory. Contractor shall explain and justify their proposed procedures for ensuring that high response rates are obtained and demonstrate their understanding and acceptance of OMB standards and guidelines for calculating response rates and actions that are required if response rates are under 80 percent.

USDA is interested in research of high scientific rigor communicated in a policy relevant fashion. The Contractor shall propose an approach to document, communicate, and present the results of the evaluation and supporting analyses to USDA, the broader research and policy field.

6 SCHEDULE OF DELIVERABLES

This contract shall be completed within nine (9) months of contract award. USDA also has a strong interest in obtaining results as soon as possible from the date contract award. The Contractor shall propose intermediary tasks and deliverables for each major task to allow for FNS review and comment.

Contractors should carefully review and revise the proposed draft timeline in accordance with the Performance Work Statement. FNS expects to receive a technical proposal with firm timelines that have been customized to reflect the contractor’s technical proposal approach while incorporating the required timeframes for FNS review of deliverables. FNS requires a minimum of two weeks to review major study deliverables and three weeks to review drafts of the final deliverable.

Task

Prepare a White Paper Examining the Current Information Technology Environment in State Child Nutrition Agencies and School

Districts to Improve Program Integrity and Oversight of the School Meals Programs through Data Matching

Deliverable Format/Quantities Due Date

Draft White Paper 1.1 MS Office Word document To be proposed by Contractor

Revised White Paper 1.2 MS Office Word document To be proposed by Contractor

Final White Paper 1.3 MS Office Word document To be proposed by Contractor http://www.ocio.usda.gov/policy-directives-records-forms/information-collection/information-collection-package

PERFORMANCE REQUIREMENTS SUMMARY

The Performance Requirements Summary (PRS) describes the requirements of this project, along with a performance standard and measure per outcome.

Performance Measure

Performance Standard

Performance Deliverable Monitored

Surveillance Method

Performance Rating Surveillance Frequency

Monitoring Performed

Draft White Paper

Written document is clear, comprehensive, well-organized, and error-free. Plan evidences a high level of technical expertise and demonstrates the use of a creative approach to achieve FNS tasks with available resources. Addresses all tasks, deliverables and research questions in section 4.2 and is edited as a final version.

All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100 percent inspection

Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables

Acceptance or Rejection of each deliverable is noted via a Government Transmittal Sheet attached to the deliverable

Revised White Paper

Written document is clear, comprehensive, well-organized, and error-free and addresses all comments and feedback from FNS

All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100 percent inspection

Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables

Acceptance or Rejection of each deliverable is noted via a Government Transmittal Sheet attached to the deliverable

Final White Paper

The Final White Paper is clear, comprehensible, well-organized, and error-free. The final paper shall reflect a thorough examination of and response to all tasks and research questions.

Comprehensively incorporates all feedback from FNS

All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100 percent inspection

Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables

Acceptance or Rejection of each deliverable is noted via a Government Transmittal Sheet attached to the deliverable

All Proposed Intermediate Tasks

Addresses all tasks, deliverables and research questions in section 4.2 and Addresses all comments and feedback from FNS

All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100 percent inspection

Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables

Acceptance or Rejection of each deliverable is noted via a Government Transmittal Sheet attached to the deliverable

Monthly Progress Reports

Timely delivery of progress reports that meet all requirements set forth in the PWS.

All deliverables as described in the PWS and specified in the approved Schedule of Deliverables

100 percent inspection

Exceptional Very Good Satisfactory Marginal Unsatisfactory

Per the contract’s approved schedule of deliverables

Ratings Definitions Exceptional Performance meets contractual requirements and exceeds many to the Government’s benefit. The contractual performance of the element or sub-element being assessed was accomplished with few minor problems for which corrective actions taken by the contractor was highly effective.

Very Good Performance meets contractual requirements and exceeds some to the Government’s benefit. The contractual performance of the element or sub-element being assessed was accomplished with some minor problems for which corrective actions taken by the contractor was effective.

Satisfactory Performance meets contractual requirements. The contractual performance of the element or sub-element contains some minor problems for which corrective actions taken by the contractor appear or were satisfactory.

Marginal Performance does not meet some contractual requirements. The contractual performance of the element or sub-element being assessed reflects a serious problem for which the contractor has not yet identified corrective actions.

The contractor’s proposed actions appear only marginally effective or were not fully implemented.

Unsatisfactor y

Performance does not meet most contractual requirements and recovery is not likely in a timely manner. The contractual performance of the element or sub-element contains a serious problem(s) for which the contractor’s corrective actions appear or were ineffective.