OMB Justification Memo

National Direct-to-Consumer Advertising Survey

OMB CONTROL NUMBER: 0910-0828

EXPIRATION DATE: 01/31/2020

Change Request (83-C)

April 17, 2017

As part of this OMB-approved data collection, we conducted a pilot study to test the data collection process. We estimated a 35% response rate, based on recent work on similar studies. Table 1 shows the initial burden estimates.

Table 1. Initial Estimated Annual Reporting Burden for the Pilot Study

Activity	No. of Respondents	No. of Responses per Respondent	Total Annual Responses	Average Burden per Response	Total Hours
Pilot Study
Survey invitation letter	100	1	100	.08 (5 min.)	8
Reminder postcard	100	1	100	.03 (2 min.)	3
Non-response letter	82	1	82	.08 (5 min.)	7
Non-response questionnaire letter	81	1	81	.08 (5 min.)	7
Second postcard	60	1	60	.03 (2 min.)	2
Survey	35	1	35	.33 (20 min.)	12
Total					39

Pilot Study Actual Response

The pilot study resulted in 20 completes for an 18.2% response rate. Table 2 shows the actual burden for the pilot study.

Pilot Study Experiment

We plan to continue the pilot study data collection with two changes: revising the survey title and testing the inclusion of the paper survey in the first contact. Table 2 shows the expected changes to the burden for the pilot study.

In order to achieve a response rate that is higher overall and more representative of the general population the title of the survey will be changed. We will change the title of the survey to “National Survey of Health Information and Communication” to avoid any unintentional association with advertising.

Responses for the paper survey were also lower than expected. We initially sent the paper survey to nonrespondents in the fourth mailing. To increase the prominence of the paper version of the survey, a paper survey will be sent to a portion of the experimental sample on the first contact with the invitation letter. It is hypothesized that showing the full content of the survey at first contact may also help to dissociate the survey from advertising.

A sample of 100 addresses will be drawn. Half the sample (n = 50) will receive the invitation letter and half (n = 50) will receive the invitation letter and a paper version of the survey. As the purpose of the experiment is to determine how response rates can be improved from the outset, all participants will only receive the first contact mailing. Response rates from the experiment will be compared to the response rate from the initial mailing in the already fielded pilot study. An improved response rate on the initial mailing will provide evidence for increased return overall. All other procedures specified for the pilot study will remain the same.

Table 2. Revised Estimated Pilot Study Burden

Activity	No. of Respondents	No. of Responses per Respondent	Total Annual Responses	Average Burden per Response	Total Hours
Pilot Study Actual Response
Survey invitation letter	118	1	118	.08 (5 min.)	9
Reminder postcard	118	1	118	.03 (2 min.)	4
Non-response letter	116	1	116	.08 (5 min.)	9
Non-response questionnaire letter	109	1	109	.08 (5 min.)	9
Second postcard	106	1	106	.03 (2 min.)	3
Survey	20	1	20	.33 (20 min.)	7
Pilot Study Experiment
Survey invitation letter	100	1	100	.08 (5 min.)	8
Survey	10	1	10	.33 (20 min.)	3
Total					52

3