011613 Supplemental Responses

Supplemental Information for: Study of the Effectiveness of Efforts to Improve Supplemental Nutrition Assistance Program Access Among Medicare's Extra Help Population Pilot Projects

Q1: The supporting statements acknowledge that some part of the population of interest may not have convenient access to a telephone and therefore will be offered an additional $10 incentive (on top of the $25 incentive) to participate.  I may have missed this, but will those respondents be sent a paper copy of the survey to fill out?  If so, would you please ask the Program to provide a copy of what that instrument will look like?  The IC in the system now is for the CATI survey and includes instructions for the programming of skip patterns etc. (which is fine for the CATI instrument, but I am guessing that a paper version will be a bit more user-friendly).

The respondents Mathematica has a difficult time contacting will be sent a letter that includes a contact information sheet (shown in Appendices I and N) with $10 asking them to send back contact information and a list of best times and days for the interviewer to contact them. If they return the contact information, we will contact them to conduct the interview; we will not be sending hard copies of the instrument.

Q2: Also, the incentive structure mentioned above ($10 + $25), is there literature to support this structure or, alternatively, has FNS conducted studies where this approach was used and found to be successful in retaining respondents?

In testing the effectiveness, Markesich and Kovak (2003) found that households receiving a prepayment of $2, and a post-completion payment of $33 reached the desired response rate in a shorter time period and with fewer contacts than households receiving a post-completion payment of $18 (Markesich & Kovac 2003). In 2012, in a study for the Department of Education, it was found that a $5 prepayment followed by a $30 post-payment boosted the response rate over an offer of a $5 prepayment and a $15 post-payment.

Additional studies document the effect of the two aspects of the approach—increasing incentives and delaying them. For example, Gunn & Rhodes (1981) found that as the incentive increased, so did the response rate, while Kulka (1994) and Abreu and Winters (1999) found that delaying the use of incentives can be used effectively to overcome nonresponse.

References

Markesich, J. and Kovac, J. (2003). The Effects of Differential Incentives on Completion Rates:

A Telephone Survey Experiment with Low-Income Respondents. Paper presented at the Annual Conference Of The American Association Of Public Opinion Research Nashville, TN, May 16.

Gunn, W. & Rhodes, I. (1981). Physician Response Rates to a Telephone Survey: Effects of

Monetary Incentive Level. Public Opinion Quarterly, vol. 45, pp. 109-115.

Kulka, R.A. (1994) "The Use of Incentives to Survey 'Hard to Reach' Respondents: A Brief

Review of Empirical Research and Current Practice", Paper prepared for the seminar on

"New Directions in Statistical Methodology" Sponsored by Council of Professional

Associations on Federal Statistics (COPAFS), Bethesda, MD, May 25-26.

Abreu, D.A., and Winters, F. (1999) "Using Monetary Incentives to Reduce Attrition in the Survey of Income and Program Participation," Proceedings of the Survey Research Methods

Section of the American Statistical Association.

Q3: Finally, does FNS plan to draw conclusions from this study about the effectiveness of improving SNAP participation among Extra Help pilots in general, or just specific to the 3 sites (WA, NM, and PA) participating in this collection? If the plan is to say something about the Extra Help pilots in general, it would be helpful to update Part A with more information on Extra Help (eligibility criteria, goals, etc.) and why results from these 3 states would be generalizable to other Extra Help projects.

FNS does not plan to draw general conclusions from this study about the effectiveness of improving SNAP participation among Extra Help pilots. The authorizing legislation charged FNS “…to test the efficacy of the pilot projects at increasing SNAP participation…” and FNS is considering this a test. Only three States proposed acceptable pilot designs and each design fit a different model permitted in the FNS pilot request. The evaluation will provide some limited information on outreach success through Extra Help. If the results are encouraging FNS will expand testing to determine if the results are unique to the initial pilot sites or if they are generalizable.

2