Request for a Non-Substantive Change to the 2015 Optimizing Self-Resonse Test

Request for a Non-Substantive Change to the 2015 Optimizing Self-Response Test

OMB No. 0607-0981

Purpose

The Census Bureau is requesting approval to add a new postcard panel of 30,000 housing units to the 2015 Optimizing Self-Response (OSR) Test. The addition of the new postcard panel will allow us to learn more about the impact of minimal but direct mail contact with households compared to varying levels of advertising on self-response.

Background

The 2015 OSR Test was designed to increase Internet self-response by testing multiple contact and notification strategies. The results from this test are critical inputs to the Census Bureau’s 2020 Census Operational Plan, design decisions.

The foundation of the 2015 OSR Test design is a contact approach termed “Internet push.” This refers to a series of four mailed contacts to each housing unit, each designed to promote self-response via the Internet questionnaire. Contacts include (1) a letter of invitation to complete the questionnaire online; (2) a postcard reminder with the questionnaire URL; (3) a second postcard prompt also including the questionnaire URL; and (4) a hard copy questionnaire for those who do not respond via the Internet.

This Internet push strategy was supplemented with an outreach and communications campaign targeting the entire local media market. Non-sampled households who did not receive any of the Internet push mailings but were exposed to the advertising materials were also invited to participate in the 2015 OSR Test. Online test submissions from non-sampled households were completed without a Census ID and are referred to as “Non-ID” cases.

Proposed Non-ID Postcard Panel

Prior to the 2015 OSR Test, Non-ID submissions had never been tested in conjunction with large-scale advertising. The submission of questionnaires without an ID have been achieved in previous censuses, and the functionality of the process – in particular on the Internet – has been previously tested. However, the impact on response rates of offering this option to respondents indirectly through community-level outreach was unknown. Based on the results we are seeing in production, we propose to supplement the original design with the addition of a new postcard panel. This proposed panel falls directly within the scope and intent of the original Internet push strategy, expanded to include housing units not originally sampled. Having the flexibility to make adjustments while still in a testing environment is crucial if well-informed decisions about the 2020 Census design are to be made.

The proposed panel of 30,000 housing units will be mailed a postcard inviting them to complete the Census questionnaire online. These cases will have been exposed to varying levels of traditional, community-level, and digital advertising but will not have received prior invitations to participate in the 2015 OSR Test. The postcard text, shown below, was developed to be consistent with the messaging used in the community-level advertising campaign. Sampled housing units who choose to participate will do so without a Census ID. They will be provided a distinct test URL to allow for tracking of these cases apart from the general, community-level or communications campaign-based, Non-ID submissions.

Postcard Text

Dear Resident:

The U.S. Census Bureau is conducting the 2015 Census Test in your area. Results from the next census will be used to help each community get its fair share of federal funding for many programs that you and your neighbors need. It is not too late to respond.

Please complete the survey online at: www.census.gov/wecount

Help us conserve natural resources, process data more efficiently, and save taxpayer money by responding online. If you are unable to respond online, or need assistance, please call our toll-free number 1–866–226–2836.

* Note: the postcard text does not include any statement about responses required by law.

Sample Design

Maintaining the integrity of the original research design built into the 2015 OSR Test was a critical factor when designing the current postcard panel. The approved current plan includes three separate and distinct tests.

• Mail Panels. Three panels of 30,000 housing units each were selected to test the impact of varying contact strategies and community-level advertising on self-response. One panel received an additional invitation to pre-register and be notified when the test was available. Another panel excluded a Census-ID from the mailings, directing households to complete the Census test using the Non-ID functionality (and testing our ability in processing to link the responses with the selected ID). The third panel was a control panel of households who received a Census-ID.

• Influencer Telephone Calls. Autodialed telephone calls and voicemail messages are being made to 60,000 housing units in the 2015 OSR Test area. Three “voices” are being tested in 20,000 households each: the Mayor of Savannah; a local news anchor “celebrity;” and a generic voice not identifiable as being local to the area.

• Digital Advertising Panels. Five advertising panels (four treatment and one control) were selected to test the impact of targeted advertising and levels of spending on response to the 2015 OSR Test. Panels varied in size, ranging from approximately 48,000 to 100,000 households.

The proposed postcard panel was selected to exclude all 90,000 housing units sampled in the three original mail panels and all 60,000 households receiving the influencer telephone calls. Additionally, any household within the remaining test area that was not previously sampled but that completed the 2015 OSR Census Test will be excluded.

The new postcard panel will be selected from the remaining addresses, all of which are part of the digital advertising panels described above. A random sample of 30,000 housing units will be selected, proportionate to size in the five digital advertising panels.

Schedule

The 2015 OSR Test will end all advertising and Internet data collection on May 31, 2015. The proposed postcard panel mailing is scheduled for the first week of May 2015, to allow at least three weeks for response.

3