MEMORANDUM FOR Dr. Nancy
Alan R. Tupek
SPD99: Proposal to Provide Incentives in the Upcoming
I. Executive Summary
In this document, we suggest an alternative proposal concerning incentive use in the upcoming data collection of the 1999 Survey of Program Dynamics (SPD). The original proposal referred to in this document was an attachment to a letter addressed to you from Chester E. Bowie, Chief of Demographic Surveys Division, dated February 5, 1999. The revised proposal reflects our understanding of the outcome of the discussion we had with you and Katherine Wallman on the use of incentives in the 1999 SPD. An overview of the changes from the original proposal is given in the following table. The detailed revised proposal is given below.
Overview of 1999 SPD Incentive Proposal Changes
This document also provides the more detailed information you requested on the operational plan and field procedures that would be used to give out the incentives. Finally, enclosed is the requested detailed table of the sample loss from the early waves of the Survey of Income and Program Participation (SIPP) through to the 1998 data collection of the SPD.
II. The Revised Proposal
In our continued attempt to slow the rate of attrition for the SPD, we request permission to offer a prepaid incentive of $40 to all Type A noninterview households returning from the 1998 data collection as part of the Type A Follow-up conversion procedure. The Type A noninterviews would receive the incentive in the advance letter, by priority mail, prior to interviewing.1 Type A noninterviews occur when no one is home after repeated attempts, household members are temporarily absent (for example, they are away on vacation for the entire interview period), the household refuses to participate in the survey or other.
We also plan to offer a special $40 incentive to households that could potentially become a Type A noninterview for the first time at the 1999 interview period. These households were not Type As in the 1998 data collection and thus, would not be receiving an incentive in Type A follow-up conversion mentioned above. Households already receiving incentives through the Type A follow-up conversion would not be eligible to receive this special incentive. Details for identifying a household as eligible for this incentive are given in the next section.
It is important to us to prevent any additional noninterview from occurring due to the already sizeable sample loss suffered by this important and unique longitudinal survey.
III. Operations and Field Procedures
A. Incentives for Type A Follow-up Conversion
Type A noninterviews from the 1998 SPD data collection
will be identified for this
If the respondent states they did not receive the
advanced letter with the prepayment, the Field
B. Incentives for Potential Type As
Households that were not Type As in the 1998 data
collection, but could potentially become a
IV. SPD Sample Loss
There are many reasons why we feel the use of incentives are justified for SPD. These reasons are given in detail in the original proposal document referred to the Executive Summary. One of these is the sizeable household sample loss suffered in SPD. The SPD inherited a 26.6 percent household sample loss rate from the Survey of Income and Program Participation sample. After two waves of SPD, the household sample loss rate is 50 percent. Table 1. offers some of the detailed sample loss information you requested. Similar tables by poverty and nonpoverty will be forthcoming. Previous studies on SIPP sample loss have shown that the sample loss is not uniform. Persons in and near poverty attrit at a higher rate than other persons. Since poverty households are a key target population in the study of welfare reform, there is some concern about nonresponse bias.
Table 1. Overall Household Sample
Loss-An Average of the 1992 and 1993 SIPP Panels and SPD
* A 10th interview was only conducted for the 1992 panel for only 3/4 of the sample.
** Only those HHs interviewed in the last wave of the 1992 or 1993 panels were initially assigned to field for the SPD Bridge.
*** Only those HHs interviewed in the Bridge and selected during the subsampling were initially assigned to field for the 1998 SPD.
We propose giving a $40 incentive to all Type A 1998 SPD cases and all potential Type A 1999 SPD cases for the 1999 SPD data collection which would be in the field at the end of April 1999. We need OMB approval by March 15, 1999 to be able to implement the procedure successfully.
1 Note that all SPD households will be receiving their advance letter by priority
mail. There is evidence that using priority mail (in the SIPP) increases
cooperation among reluctant respondents.