HUD is currently funding all data collection activities related to hiring and training of interviewers, interviewer administrative tasks, respondent recruitment, some interviewing activities, quality assurance, data support, data deliverables and reporting as part of a consortium of funders. Our assumption for Wave 1 data collection was that Voucher sample would require approximately one-third the effort of population cases primarily because (a) the cases would be pre-consented by Housing Authorities; (b) be pre-screened for eligibility by the Housing Authorities; and (c) would have updated contact information allowing for telephone screening and appointment setting.
During the first quarter of the first wave of data collection, we found that the Voucher sample actually required more effort per complete than the population sample. This was primarily because (a) contact information was either outdated or missing; (b) voucher cases had a significantly lower eligibility rate than expected under pre-screening; and (c) voucher cases had resistance rates similar to or (in Dallas) in excess of the population resistance rates, which is the opposite of what we would expect from pre-consented sample.
In order to meet the overall goals of the study, we will need supplemental funding to offset the increased hours per interview during the first Wave of data collection. To obtain the increased hours needed, we will need to both add a supplemental interviewer training and extend the Wave 1 field data collection period by 4 months. We are also proposing adding respondent locating effort to the last four months of data collection to streamline the ability of interviewers to more quickly locate, contact, and interview voucher respondents.
Johns Hopkins University
12/01/2018 to 06/30/2020