From previous large- and small-scale research supported by USDA, much already has been learned. Here, we offer a brief guide so that applicants can focus on potential research that is new and innovative.
The Request for Proposals has some of the most important information. As it explains, all applicants first submit a concept paper, and then applicants who pass the first review may be invited to submit a full proposal. The RFP provides an overview of USDA research interests in three areas:
- Food Choices: Economic Determinants and Consequences
- Economic Incentives in Food Assistance Programs
- Food Assistance as an Economic Safety Net
The RFP also provides a list of questions that will be used by external reviewers in evaluating proposals under four headings:
- Research Merit of the Proposal
- Overall Approach
- Workplan, Budget, and Cost-Effectiveness
- Key Personnel
Because the concept papers in the first stage are short for a reason – to reduce applicant burden – it is good to keep the background short. The concept paper can explain the importance of the problem to be addressed just very briefly, assuming the reader already is familiar with the nutrition assistance programs. The emphasis should be on the research merit (explaining what is new and interesting) and the overall approach (reassuring the reviewer that the applicants have decided the research approach, that this approach will work, and that the project can be completed as intended).
A research study may make a useful contribution through an appropriate combination of (a) bringing new information to light and (b) using a strong research design with promise for answering an important question. For example, studies may measure the effects of nutrition assistance programs or the determinants of program participation.
When the purpose of a study is principally descriptive – bringing new information to light – applicants can consider whether the proposed data source is the best source, and whether it is representative, providing a representative sample of a well-defined population of program participants or low-income non-participants.
When the purpose of a study is analytic – measuring program effects or the determinants of program participation – there are a number of useful sources for thinking about research design. In seeking to measure the effects of nutrition assistance programs, a major longstanding challenge has been “self-selection” into program participation. Because program participation is voluntary, it is possible that participants are systematically different from non-participants in several respects. This possibility makes it difficult to use participant/non-participant comparisons to assess program impacts. In a classic USDA report, Hamilton and Rossi (2002) summarize strengths and weaknesses of a wide variety of research designs. Wilde (2007) and Wilde (2013) give examples of multiple approaches to measuring the effect of SNAP on household food security. More recently, for SNAP, Andreyeva, Tripp, and Schwartz (2015) summarize a large literature about effects on dietary quality. Many of the studies cited in these reviews illustrate contemporary methods for addressing this challenge.