Skip to Content


Proposed Information Collection; Comment Request; 2015 Optimizing Self-Response and Census Tests

Document Details

Information about this document as published in the Federal Register.

Published Document

This document has been published in the Federal Register. Use the PDF linked in the document sidebar for the official electronic format.

Start Preamble Start Printed Page 51948


U.S. Census Bureau.




The Department of Commerce, as part of its continuing effort to reduce paperwork and respondent burden, invites the general public and other Federal agencies to take this opportunity to comment on proposed and/or continuing information collections, as required by the Paperwork Reduction Act of 1995, Public Law 104-13 (44 U.S.C. 3506(c)(2)(A)).


To ensure consideration, written comments must be submitted on or before November 3, 2014.


Direct all written comments to Jennifer Jessup, Departmental Paperwork Clearance Officer, Department of Commerce, Room 6616, 14th and Constitution Avenue NW., Washington, DC 20230 (or via the Internet at

Start Further Info


Requests for additional information or copies of the information collection instrument(s) and instructions should be directed to Erin Love, Census Bureau, HQ-3H468E, Washington, DC 20233; (301) 763-2034 (or via email at

End Further Info End Preamble Start Supplemental Information


I. Abstract

During the years preceding the 2020 Census, the Census Bureau will pursue its commitment to reduce the costs of conducting a decennial census, while maintaining our commitment to quality. A primary decennial census cost driver is the collection of data from members of the public for which the Census Bureau received no reply via initially offered response options. Improving our methods for increasing the number of people who take advantage of self-response options will help increase the efficiency and effectiveness of Census operations. Additionally, improving our methods for enumerating people who do not initially respond can contribute to a less costly census while maintaining high-quality results.

In order to help achieve these two improvements, the Census Bureau will conduct two tests in early 2015: The 2015 Optimizing Self-Response (OSR) Test, and the 2015 Census Test. The 2015 OSR Test will allow the Census Bureau to, on a small scale, employ a variety of new methods and advanced technologies that are under consideration for the 2020 Census. In particular, for the 2020 Census the Census Bureau plans to allow response via the Internet. We studied this during the 2014 Census Test, but as described below, we want to study other aspects of this further in the 2015 OSR Test. Also as described below, the 2015 Census Test will be used to study use of automation and available real-time data to transform the efficiency and effectiveness of field data collection operations.

2015 OSR Test—To improve Internet self-response, the Census Bureau plans to continue testing multiple contact and notification strategies. For example, the 2015 OSR Test will include a “Notify Me” campaign, which allows respondents to pre-register their email address and cell phone number and provide their preference for future contacts, by email or text message. The Test will also include a communications component to increase awareness and encourage on-line participation by potential respondents. These outreach efforts will include some methods not previously employed for a decennial census, such as targeted digital marketing for demographic groups that we know to be hard-to-reach from past censuses and surveys. We will use our planning database to identify hard-to-count groups at the block or tract level and place ads with targeted messaging on digital sites frequented by these groups.

The 2015 OSR Test will also continue Census Bureau efforts to increase self-response via the Internet and make it easier for respondents by allowing them to respond without providing a pre-assigned User identification (ID) number associated with their address. Building on the work from the 2014 Census Test, we will test our ability to do real-time processing of responses lacking a pre-assigned User ID. Thus, for this test, while respondents are completing their census form online, we will attempt to search for their address in our Master Address File (MAF) to determine if it matches an existing MAF record or could be added as a new address. If the initial attempt to match is unsuccessful, and we are able to determine if further respondent input could assist us, the internet response instrument will prompt the respondent accordingly. In any case where a match cannot be derived, an automated process will attempt to assign the respondent address to a census block, and then the respondent will be asked to confirm or correct that location via a map interface integrated with the Census questionnaire. Ultimately, each response that lacks preassigned ID will either be matched to an address in the MAF or assigned to a census block. Matching a respondent's provided address information to our MAF permits the removal of the address from the non-response universe, thereby reducing the non-response follow-up effort required. This has the potential to produce significant cost savings.

2015 Census Test—In the decennial census, no matter how many response options we provide, and no matter how much we encourage self response, there will be households that do not respond, and there will be vacant units to which a form is delivered. Therefore, there will be a need for personal visit followup visits to addresses. In the 2015 Census Test, we will study strategies to most effectively and efficiently collect information from those households. In the 2015 Census Test, we will be testing an enhanced operational control system that will optimize the case assignments and routes for the enumerators. The use of this automation will also test a new structure for managing the work and the field staff. We will examine the effect on cost and data quality of reducing the total number of contacts made to a household during the Nonresponse Follow-up (NRFU) operation, as well as adapting the number and type of contacts made to a household based on information we already have about that household from administrative records. Administrative records can include information from federal, state or third-party sources. Examples of administrative records include Internal Revenue Service (IRS) Individual Income returns, Center for Medicare and Medicaid Services (CMS) Medicare Enrollment information, and information from the United States Postal Service (USPS) Undeliverable as addressed (UAA) file. This Test will help us determine to what extent using administrative records information to remove non-responding cases from the field workload can reduce the costs associated with NRFU operations.

The 2015 Census Test will also include an Evaluation Follow-up interview for a subset of households to help reconcile and understand differences observed between the administrative records and the NRFU interview results from self-responses and proxy respondents. Additionally, the Test will use focus groups to analyze reactions to the contact method and administrative record use, as well as any privacy or confidentiality concerns.Start Printed Page 51949

II. Method of Collection

The 2015 Optimizing Self-Response Test— The Census Bureau will conduct this Test in a location that offers a medium-sized media market for advertising and outreach that also provides diversity in demographics, address types, and internet penetration and usage. To the extent practicable, the selected market should be self-contained to limit bleed-over of advertising efforts into neighboring markets. The site will provide a community to engage “together” about the Census test with outreach, promotion, advertising, and the contribution of social media to create the civic spirit to participate in the Test. The Test is comprised of four parts, namely Communication, “Notify Me,” Enumeration, and follow-up via Focus Groups.

Communication—The Census Bureau wants to learn about expanding the use of a variety of communication strategies and methods aimed at increasing the use of self-response options in a decennial census. In the 2015 OSR Test, these strategies could include outreach, promotion, advertising, partnerships, social media, email, and postal contacts. In addition to traditional advertising strategies (e.g., television and newspaper advertisements), this test will also use targeted digital (e.g., cable companies directing specific ads by address) and online advertising.

“Notify Me”—By deploying a variety of awareness, advertising, direct contact, and partnership strategies, this test hopes to reach and engage respondents to “pre-register” for the census test. This provides an opportunity for respondents to have early engagement in the census process and to select their preferred mode for future invitations and reminders (i.e, how to “Notify Me” when it is time to complete the census form—email or text message). The ability to engage respondents to pre-register is an important objective of the Test to continue research in optimizing the respondents' use of the Internet as a self-response mode.

Enumeration—The 2015 OSR Test will collect Internet self-response enumeration data from both a set of households contacted directly as well as respondents within the Test site who become aware of the Test only via outreach, promotion, and advertising. The Census Bureau will directly contact up to 300,000 housing units to notify them of the survey. A subsample of these notifications will provide a User ID, and the remaining sample will not. Additionally, respondents who become aware of the Test, but have not been directly contacted, can self-respond via the Internet without the need for a User ID. The two sets of non-ID respondents will allow us to further test our non-ID processing methodology, which compares responses without a pre-assigned ID to our Census address and geographic database. In the 2015 OSR Test, we will also test our ability to conduct real-time non-ID processing so that we are able to prompt a respondent (while they are still on line filling out the form) for additional address and location information if the respondent's address cannot be matched or geocoded. A non-ID respondent whose address cannot be matched to our address database will be prompted during his or her Internet self-response session to confirm the address information they provided while filling out the form, or to indicate the location of their address on an on-screen map, and no subsequent contact will occur. Additionally, we plan to test a mechanism for validating all non-ID respondents. We will also be testing optimal strategies for delivering mail materials, including paper questionnaires, to households who do not or cannot respond online.

Focus Groups—The OSR test will also be used to obtain some qualitative data from respondents and non-respondents regarding the pre-registration strategy. This will be collected via eight focus groups, comprised of various categories of respondents and non-respondents. Through the focus groups, we will ask about perspectives on burden; whether they thought that pre-registration was the actual participation or response to the Census Test; and whether they have preference to wait for Census Day without registering a contact preference. We also want to learn about respondent opinions and perspectives on broader objectives for the OSR testing, so we will try to ascertain and discuss the outreach, promotion, media/mode or method that informed the respondent about the pre-registration option, and/or the Census Test. We will also ask if the ability to respond without having to provide a preassigned User ID made the respondent more likely to participate.

The 2015 Census Test— The Census Bureau will conduct this test in one county or two contiguous counties (location to be determined). We expect the location will have a combined population over one million people and we want to have an area with high concentrations of Hispanic population, vacant housing units and mobile populations. This will allow us to study the impacts of the usage of administrative records on the Hispanic population, vacant housing units and areas with more mobile populations. We will select approximately 170,000 housing units to be contacted, including an initial self-response phase that is followed by a NRFU phase of no more than 80,000 non-responding housing units.

For the self-response phase, households within the test site will receive an initial invitation to go to the 2015 Census Test Web site and complete their census response online. For households that have not responded within an allotted time period, the Census Bureau will attempt to contact them additional times, which will include postcard reminders and a final reminder along with a paper questionnaire that they can complete and return by mail.

If a household ultimately does not respond by a certain date, it will be included in the universe for the NRFU portion of the test. There are two new NRFU strategies (described below) being tested in the 2015 Census Test. In addition, the test will include, as a control group, a strategy similar to the 2010 Census field procedures from which to compare results.

First, the Adaptive Design strategy will test a method of managing data collection more efficiently by adapting contact attempt strategies on a per case basis. For example, using alternative modes of contact, variable number of visits to households, and modeling the best day and time to contact a household. Second, the Administrative Records strategy will remove cases from the NRFU workloads at various stages of fieldwork in an effort to reduce costs associated with visiting non-responding households. In some areas, administrative records information will be used as the response for households (both unoccupied and occupied) before we make any visits to those households. In other areas, administrative records information will be used to enumerate only unoccupied units before we make any visits to those units. In these areas, we will conduct one in-person visit to the remaining households to enumerate the people in that household. If we are unsuccessful at contacting anyone in the household when we visit, administrative record information will be used to enumerate any occupied households remaining. For those households where administrative records are not available, we will continue to contact them based on the adaptive design strategies discussed above.

As part of the administrative records research, this Test will also include an Evaluation Follow-up interview to help reconcile and understand differences Start Printed Page 51950observed between the administrative records and the NRFU interview results from self-responses and responses we receive from neighbors or others who are knowledge about the households (referred to as proxy responses). We will attempt to conduct an interview with a sample of approximately 10,000 cases from various groups (one example comparison will be between cases that self-responded and cases that administrative records identified as vacant) to provide additional data points for which to analyze the effectiveness of using administrative records in NRFU.

The Census Bureau will conduct NRFU with a combination of enumerator-owned and government-owned, commercially provided mobile devices. The use of employee owned equipment/services is commonly referred to as “Bring Your Own Device” or BYOD. A sample of up to 5,000 households will be contacted at the end of the field operation using this methodology. The objectives of this component of the test are to:

  • Design and develop software solutions, deployment, and support processes that run on commercially available employee owned mobile devices (i.e., iPhone, Android);
  • Deploy and support secure software solutions that can be installed on commercially available employee owned mobile devices;
  • Conduct interviews of respondents using employee owned mobile devices; and
  • Capture lessons learned for future operations.

It is important to note that the Census data collection application, known as COMPASS, collects, stores, and securely transmits data for smart phones used by Census enumerators. This application requires a series of security measures to be met in order for the enumerators to collect, store, access, and transmit sensitive information.

Focus groups will be conducted in the geographic site of the test. Respondents will be recruited into groups with regard to their treatment and demographic characteristics (e.g., age, education). For example, respondents could be recruited into one of the groups of 8-12 participants by age and education as well as whether they were NRFU respondents or non-respondents. Focus groups would explore reactions to the contact method, administrative record use, any privacy or confidentiality concerns and how the Census Bureau might address these concerns through micro- or macro- messaging.

III. Data

OMB Control Number: None.

Form Number: Paper and electronic questionnaires with numbers as yet to be determined.

Type of Review: Regular submission.

Affected Public: Individuals or households.

Estimated Number of Respondents: 2015 OSR Test: “Notify Me”—100,000. Enumeration—305,000. Focus Groups—80. 2015 Census Test: Enumeration—220,000. Evaluation Follow-up—10,000. Focus Groups—80.

Estimated Time per Response: 2015 OSR Test: “Notify Me”—4 minutes per response. Enumeration—12 minutes per response. Focus Groups—2 hours per respondent. 2015 Census Test: Enumeration—10 minutes per response. Evaluation Follow-up—10 minutes per response. Focus Groups—2 hours per respondent.

Estimated Total Annual Burden Hours: 2015 OSR Test: “Notify Me”—6,667. Enumeration—61,000. Focus Groups—160. 2015 Census Test: Enumeration—36,667. Evaluation Follow-up—1,667. Focus Group—160.

Estimated Total Annual Cost: For the 2015 OSR Test, respondents who are contacted by text message per their election, may incur charges depending on their plan with their service provider. The Census Bureau estimates that the total cost to respondents will be no more than $20,000. There are no other costs to respondents other than their time to participate in this data collection. For the 2015 Census Test, there is no cost to the respondent other than the time to complete the information request.

Respondent's Obligation: Mandatory.

Legal Authority: Title 13, U.S.C. 141 and 193.

IV. Request for Comments

Comments are invited on: (a) Whether the proposed collection of information is necessary for the proper performance of the functions of the agency, including whether the information shall have practical utility; (b) the accuracy of the agency's estimate of the burden (including hours and cost) of the proposed collection of information; (c) ways to enhance the quality, utility, and clarity of the information to be collected; and (d) ways to minimize the burden of the collection of information on respondents, including through the use of automated collection techniques or other forms of information technology.

Comments submitted in response to this notice will be summarized and/or included in the request for OMB approval of this information collection; they also will become a matter of public record.

Start Signature

Dated: August 26, 2014

Glenna Mickelson

Management Analyst, Office of the Chief Information Officer.

End Signature End Supplemental Information

[FR Doc. 2014-20661 Filed 8-29-14; 8:45 am]