SOP Data Collection Changes Narrative 020813

Changes to the Street Outreach Program (SOP) Data Collection Project
Type of Change:
Non-substantive
Justification of Proposed Changes:
Computer assisted personal interviews were being recorded in order to capture answers to open-ended
questions that might be too long for interviewers to type. The answers to the open-ended questions
collected during the pilot study ranged from one word to two sentences; interviewers reported being
able to easily type the answers into the computer tablets. The open-ended questions require the
interviewer to read a script about audio recording the answers, which proved cumbersome and
awkward during the pilot study. Therefore, in order to simplify the data collection interview protocol, it
is proposed that the audio recordings be discontinued.
Description of Proposed Changes:
Consent forms. Information about audio recording computer assisted personal interviews has been
removed from the consent forms.
Focus group questions. One question was removed because it was confusing to respondents. One
question was added to better understand the services respondents might seek to obtain permanent
housing.
Since the open-ended questions did not yield detailed responses, a fourth focus group is proposed for
each site for a total of 4 four (2 female, 2 male) per each site.
Interview questions. Information about audio recording computer assisted personal interviews has
been removed from the interview protocol.
Based on the pilot test in the Omaha site, 5 questions about the respondent’s current homeless
situation, events that led to becoming homeless, and main reasons for becoming homeless were
removed from the computer assisted personal interview. In addition, one question asking about gift card
preference was removed. Five questions about satisfaction with the SOP agency were added to the
computer assisted personal interview. Given the elimination of the 6 questions, the 5 new questions will
not add to the response burden.
Based on the feedback obtained from community agencies during the data collection training that took place
with all 11 grantees, we have: 1) changed the wording of some questions to increase clarity; 2) eliminated 3
questions; 3) added 8 new questions; and 3) added 16 followup questions to existing questions. Based on the
practice interviews conducted during 3 days of training with the 11 grantees, the additional questions and
response categories add an additional 1 minute and 40 seconds to the interview. Since interviews conducted
during the pilot test averaged 45 minutes, the addition of these questions will not increase response burden.
The 8 new questions ask about new topics for which information is lacking but deemed important by the
community agencies to inform their work. The questions are on the following topics: overdose (2 questions);
law enforcement (2 questions), weapon use (3 questions), and ethnicity (1 question).