The U.S. Department of Agriculture, Food and Nutrition Service (FNS) intends to issue a Blanket Purchase Agreement (BPA) in 2011 for the procurement of analytic projects and is seeking statements of capability from potential offerors. Preference will be given to small businesses with experience and staffing to perform the requirement as demonstrated in the capability statement. Study Overview FNS seeks the services of qualified contractors to assist with analytic projects, short studies, research and programming tasks, and other work as needed. The contractor(s) will provide programmatic and technical expertise to FNS to support the agency's evaluation of the Supplemental Nutrition Assistance Program (SNAP), and the agency's assessment of current and emerging policy issues affecting the program. FNS expects that several contractors will be awarded the BPA and then the contractors can choose to bid competitively on each task. Each year approximately 3 to 6 tasks will be awarded based on the priorities of the agency. The parameters of the tasks are fixed, with stated objectives and a limited set of well-defined deliverables. FNS expects that most of the tasks under this BPA will be completed with previously compiled public datasets, although some tasks may require small scale data collections. Work to be performed under this contract may include, but is not limited to, statistical analysis of existing data, econometric modeling, authoring reports, independent review of reports including those produced under other FNS contracts, creation of analytic files from existing primary source datasets, literature searches and reviews, and related research and analysis tasks. Examples of the tasks that may be awarded each year are as follows: Task 1: Research Review of the Extent and Nature of Under Reporting of Nutrition Assistance Program Participation in National Surveys This task examines under reporting of SNAP participation in national Census surveys. These surveys contain counts of SNAP participants that tend to be lower than what FNS finds in its own Quality Control (QC) data and in State administrative records. The undercounts differ across surveys, and potentially exist for different reasons. Examples include, but are not limited to the following: •The American Community Survey (ACS) only has one question that addresses Food Stamp/SNAP receipt in its questionnaire. Moreover, this survey is intended to be a rolling replacement for the long-form Decennial Census and is not necessarily designed to extract detailed information about household income and/or participation in government assistance programs. •The March Current Population Survey (CPS) Annual Social and Economic Supplement (ASEC), while more detailed than the ACS regarding household income and employment, still has shortcomings regarding its ability to capture SNAP participation. CPS collects income sources on an annual basis, rather than a monthly basis used in the SNAP eligibility determination and does not collect information on assets and deductible expenses which affect SNAP eligibility. Moreover the CPS also underreports public assistance benefits, like Temporary Assistance to Needy Families (TANF) that would be counted as income in SNAP. •The Survey of Income and Program Participation (SIPP) tends to have the most detailed survey with regards to household income and program participation. However, SIPP (like CPS) also underreports TANF; has high attrition rates across the length of its survey panels, and has a long time lag (over two years) between data collection, processing and release that may affect its SNAP undercount. •The National Health and Nutrition Examination Survey (NHANES) has a series of questions that ask about previous and current SNAP receipt, including benefit amount. SNAP participation is underreported. In addition to the possible reasons for misreporting across the surveys as posited above; households could potentially misreport for reasons related to household characteristics, program participation characteristics (i.e. the length - or shortness- of the participation spell, the size of the monthly benefit), or due to reasons inherent in the design of the survey questionnaire(s). Overall, this task seeks to assess under reporting of SNAP participation in four national surveys with the goal of identifying potential remedies to the problem. More specific objectives of this task are to: •Review and summarize research conducted to examine the extent and nature of under reporting participation in SNAP, other nutrition assistance programs or other major assistance programs targeted to low-income persons. •Document and compare the magnitude of under reporting that occurs for SNAP participation in the CPS, SIPP, NHANES and ACS. •Identify and compare any variation in SNAP under reporting by household characteristics, SNAP participation patterns, time, question construction or administration, and other potentially relevant factors as observed in the CPS, SIPP, NHANES and ACS. •Identify possible next steps for addressing the issue of under reporting SNAP participation - e.g. additional analyses to assess problem, question/questionnaire changes, analytic adjustments to currently available participation estimates. i Task 2: Examine the Feasibility of Using GIS Data to Catalog a Representative Sample of Food Deserts and Characteristics of the SNAP Households Residing There SNAP benefits are targeted to the purchase of food for home use and are delivered through normal channels of trade - currently about 200,000 authorized food retailers. Access to a variety of high quality and affordable foods are essential to meeting the Program's mission. There has been an ongoing concern that some SNAP household, especially those living in rural areas or densely populated, low-income communities have limited access to the stores providing nutritious and reasonably priced foods. Such environments are known as food deserts. Most often they are defined in terms of the distance between one's residence and the closest supermarket or large grocer Economic Research Service, 2009). EBT transaction data provide a clear picture of the types of stores SNAP recipients use to redeem their benefits. However, there is little information about the proximity and variety of store types to SNAP households. This project will examine the feasibility of using geographic information system (GIS) tools and extant data sets to identify and catalog a representative sample of urban and rural food deserts, along with the characteristics and shopping patterns of SNAP households living within them. General objectives of Task 2 are to: •Determine available sources of information (addresses, characteristics and food shopping patterns) on SNAP households and stores (addresses and characteristics) to use as inputs in identifying food deserts. •Determine available GIS tools capable of electronically linking relevant data sources. Determine also whether data sources can be linked through other means. •Determine the steps and/or resources are needed to access these data sources. •Identify the key features of these data and GIS tools. Compare these features in terms of format requirements, accuracy, detail, cost, and ease of updating. •Identify any other advantages, constraints or unknowns associated with any identified GIS tools and data sources. Task 3: Estimating the Racial and Ethnic Characteristics of SNAP Participants With the introduction of additional response categories for recording race and Hispanic/Latino background, there has been some concern about the Agency's ability to accurately identify these characteristics among SNAP participants. To date, there has been a substantial amount of missing data for these variables in the QC sample. The purpose of Task 3 is to determine if there are strategies to improve the quality of such information. Among the approaches to consider are: •Meeting with representatives from the Census Bureau and federal agencies with programs serving low-income populations to share information on the extent to which obtaining racial and ethnic data is problematic and on strategies to address the problem; and/or •Compare for the same time frame aggregate racial and ethnic data for SNAP participants recorded by States on the FNS-101 form and stored in the Agency's National Data Bank with information in the QC sample. This strategy would also involve obtaining more detailed information on how SNAP eligibility workers assign racial and ethnic codes that are captured in the FNS-101. Task 4: Create a Searchable Database for SNAP Nutrition Education Evaluations Because SNAP Nutrition Education (SNAP-Ed) takes a variety of forms and is targeted to diverse program subgroups both across and within States, a single comprehensive evaluation is not feasible. In addition, while States are required to conduct needs assessments and use evidence-based educational interventions, they are not required to conduct implementation or impact evaluations. The evaluations conducted may or may not be communicated beyond the implementing or State SNAP agencies. When shared, they may be found in SNAP-Ed reports to FNS, organizational web-sites or reports, or peer-reviewed journals. Typically the content varies substantially from one report to another. The purpose of Task 4 is to identify the body of SNAP-Ed evaluations completed in the last five years that meet a set of selection criteria established in coordination with FNS. These criteria shall include some indication of design strength for impact assessments as well as indicators of the completeness of information contained in the report. Key information from each evaluation shall be extracted and used to create a searchable database. The data base shall include links to obtaining the full report, be searchable with variety of keys, user friendly, and easily expanded to incorporate additional evaluations. Other Information All responses must be submitted to
[email protected] NO LATER THAN 4:00 p.m. on February 10, 2011. Responses will be considered solely for information purpose only. This is not a request for proposal (RFP) and responses may or may not be used for future RFP. The Government is under no obligation to pursue further action on this request. For questions concerning this notice, please contact Belal Hammad, Contract Specialist, at 703-305-2262. Please limit the length of the packages to 20 pages. Packages over 20 pages will not be assessed.