Supporting Statement_2010A2

SUPPORTING STATEMENT

U.S. Department of Commerce

U.S. Census Bureau

2010 Census Coverage Measurement Person Interview, Person Interview Reinterview Operations, and Recall Bias Panel Study

OMB Control # 0607-XXXX

A. Justification

1. Necessity of the Information Collection

The U.S. Census Bureau requests authorization from the Office of Management and Budget to conduct the Census Coverage Measurement (CCM) Person Interview (PI) and Person Interview Reinterview (PIRI) operations as part of the 2010 Census. This request is being submitted for a second time. The clearance request was withdrawn and is now being resubmitted because the Census Bureau has made changes to the request. The sample size has been lowered, the reinterview rate has been increased and a new Census Coverage Measurement Recall Bias Panel Study has been added. Two previous notices were published in the Federal Register announcing plans to submit this request (June 19, 2009 on page 29166 and Nov. 24, 2009 on page 61329). Neither of the previous notices included information about these changes. The CCM program will provide estimates of net coverage error and components of census coverage (including omissions and erroneous enumerations) for housing units and persons in housing units (see Definition of Terms). The data collection and matching methodologies for previous coverage measurement programs were designed only to measure net coverage error, the difference between the estimate of the true population count and the actual census count.

The 2010 CCM will be composed of two samples selected to measure census coverage of housing units and the household population: the population sample (P sample) and the enumeration sample (E sample). The primary sampling unit is a block cluster, which consists of one or more contiguous census blocks. The P sample is a sample of housing units and persons obtained independently from the census for a sample of block clusters. The E sample is a sample of census housing units and enumerations in the same block clusters as the P sample. The results of the housing unit matching operations will be used to determine which CCM and Census addresses will be eligible to go to the CCM Person Interview (PI) Operation. The PI Operations will contain approximately 205,000 sample addresses. The Person Interview Reinterview Operation will be a sample of those cases with an estimated 30,750 sample addresses.

The automated PI instrument will be used to collect the following information for persons in housing units only:

1. Roster of people living at the housing unit at the time of the CCM PI Interview.

2. Census Day (April 1, 2010) address information from people who moved into the sample address since Census Day.

3. Other addresses where a person may have been counted on Census Day.

4. Other information to help us determine where a person should have been counted as of Census Day (relative to Census residence rules). For example, enumerators will probe for persons who might have been left off the household roster; ask additional questions about persons who moved from another address on Census Day to the sample address; collect additional information for persons with multiple addresses; and collect information on the addresses of other potential residences for household members.

5. Demographic information for each person in the household on Interview Day or Census Day, including name, date of birth, sex, race, Hispanic Origin, and relationship.

6. Name and above information for any person who has moved out of the sample address since Census Day (if known).

We also will conduct a quality control operation - PI Reinterview (PIRI) on 15 percent of the PI cases. The purpose of the operation is to confirm that the PI enumerator conducted a PI interview with an actual household member or a valid proxy respondent and conduct a full person interview when falsification is suspected. If PIRI results indicate falsified information by the original enumerator, all cases worked by the original enumerator are reworked by reassigning the cases to a different PI enumerator.

In addition to the CCM PI and PIRI Operation, the 2010 CCM Recall Bias Panel Study will be conducted, as part of the initiatives approved by the Census Bureau Director to reduce nonsampling error in the CCM Program. The study is a telephone study to examine recall bias in the CCM with respect to residence during the 2010 Census cycle. One of the recurring questions regarding the 2010 CCM is whether conducting the CCM Person Interview (PI) and Person Followup (PFU) operations later than in previous post-enumeration surveys will cause degradation on the data collection of respondent moves since Census Day (April 1, 2010) and the information on alternate addresses for the residents. The main goal of the study is to provide initial insight into the issue of recall bias for the CCM PI and PFU. The plan is to design and implement further studies of this issue for CCM in the 2020 Decennial testing life cycle.

Starting in May 2010, telephone interviews for four panels with 10,000 cases each will be conducted. Each panel will contain a random digit dialing (RDD) sample (containing both telephone lines and cell phones), and a sample of addresses from the Master Address File (MAF) with indicators from the National Change of Address (NCOA) File. These addresses contain known movers from the March and April 2010 timeframe. Interviews will be conducted during May, June, and September 2010, and February 2011. The May panel will be used as the control; June was the timeframe of the Person Interview (PI) in the 2000 Accuracy and Coverage Evaluation Survey; September is the midpoint of 2010 CCM PI; and February is the midpoint of the CCM Person Followup (PFU). We will roster the contacted units and ask them questions similar to the CCM PI/PFU series of questions. After the data are collected, we will compare each panel's answers to the questions about moving and Census Day residence. Because we have a representative RDD sample in each panel, we expect the same overall proportion of moves during April in each panel, and we expect the same proportion of other residences to be reported for that time frame as well. If there are changes in the proportions of moves within that month across the panels, we can conclude that the data degraded. We can then measure the magnitude of the degradation over time. Using the NCOA cases we will be able to get the same measures as the RDD sample, but will be able to determine the effect of timing of the interview on recall bias for known movers.

Title 13, United States Code, Section 141 authorizes the Secretary of Commerce to conduct a decennial census of the population, and Section 193 authorizes the Secretary to conduct tests to gather supplementary information related to the census.

2. Needs and Uses

The CCM program will provide estimates of net coverage error and components of census coverage (including omissions and erroneous enumerations) for housing units and persons in housing units (see Definition of Terms). Person interview will collect the unit status and independent roster of people who lived at the sample address either at Census Day or Interview Day. These people will be used for person matching.

Information quality is an integral part of the pre-dissemination review of data by the Census Bureau (fully described in the Census Bureau’s Information Quality Guidelines located at http://www.census.gov/quality/). Information quality also is integral to the information collections conducted by the Census Bureau and is incorporated into the clearance process required by the Paperwork Reduction Act.

3. Use of Information Technology

Information technology will be employed during the PI and PIRI operations. PI and PIRI are Computer Assisted Data Collection instruments and interviews are conducted by personal visits or telephone (see Attachments B and C). It will also be used during the Recall Bias Panel Study. The Recall Bias Panel Study is also a Computer Assisted Data Collection instrument and interviews are conducted by telephone. (see Attachment E)

4. Efforts to Identify Duplication

CCM operations are designed to collect information to evaluate the methods used in the 2010 Census and possibly identify new approaches to more accurately measure and prevent (as possible) coverage error (net and for components) for housing units and persons in future censuses. This includes estimates of the number of people and housing units duplicated. This effort does not duplicate information collected by any other agency.

5. Minimizing Burden

The proposed information collection consists of collecting demographic and address information about persons rostered in the CCM Person Interview and Person Interview Reinterview operations and the Recall Bias Panel Study. Interviews will be conducted at the smallest number of housing units required to adequately estimate net coverage error and the components of census coverage or measure differences in mover reporting. Respondents will be asked the minimum number of questions to identify Census Day and Interview Day residence of the sample address and other potential addresses where they may have been counted in the census.

6. Consequences of Less Frequent Collection

Improving census coverage is one of the major goals of reengineering the 2010 Decennial Census Program. If these activities were not conducted, it would adversely affect our ability to measure the coverage of the population in the 2010 Census.

7. Special Circumstances

No special circumstances exist.

8. Consultations Outside the Agency

The Census Bureau published a notice in the Federal Register on June 19, 2009 (Vol. 74, No. 117, PG 29166), inviting public comment on our plans to submit this information collection.

One comment was received from the Brookings Institution (see Attachment D). The letter agrees with the need for conducting both CCM PI and CCM PIRI for the Census Bureau to judge its performance as well as provide essential information. It also made one minor suggestion deemed not relevant for this CCM operation.

The first submission of this notice received one comment from the Brookings Institution as well before being withdrawn. It agrees with the need for both CCM PI And PIRI again and addresses its previous comment with a request for evaluation in the 2020 design phase (see Attachment F).

9. Paying Respondents

Respondents participating in this survey will not receive any form of compensation for their participation.

10. Assurances of Confidentiality

All CCM PI and PI RI respondents will be informed that participation is mandatory, that their responses are confidential, and that all information that could identify individuals will be held in the strictest confidence under applicable Federal statutes. This survey complies with the requirements of the Privacy Act of 1974 and the Paperwork Reduction Act. (See Attachment A – Introductory Letter for CCM PI and PIRI)

For the Recall Bias Panel Study, all respondents will be informed that participation is voluntary, that their responses are confidential, and that all information that could identify individuals will be held in the strictest confidence under applicable Federal statutes. This survey complies with the requirements of the Privacy Act of 1974 and the Paperwork Reduction Act. (See Attachment E – Specification for Recall Bias Study)

11. Justification for Sensitive Questions

The Census Bureau perceives no question in the CCM PI, PIRI, or Recall Bias Panel Study questionnaires as being sensitive.

12. Estimate of Hour Burden

A sample of approximately 205,000 housing units will be selected spread evenly among the 50 states, District of Columbia, and Puerto Rico for the PI operation. A sample of approximately 30,750 housing units will be selected for the PIRI operation in the same areas. A sample of approximately 40,000 respondents will be selected for the Recall Bias Panel Study. To calculate the burden hours, we assumed a theoretical 100 percent response rate and an approximate completion time of 15 minutes per respondent (based on the results from the 2009 Person Interview Operational Test). The estimated total annual respondent burden for the PI and PIRI operations and Recall Bias Panel Study is approximately 68,938 hours.

13. Estimate of Cost Burden

There is no cost to respondents, except for the time it takes to respond to the questions.

14. Cost to the Federal Government

The cost incurred by the Census Bureau to conduct the Person Interview field operation is estimated to be about $42,143,247¹. An inter-divisional Census Bureau team developed the data collection methodologies. The Census Bureau’s Decennial Management Division allocated the resources for the effective and efficient management of the information.

15. Justification for Change in Burden

This collection is being submitted as new.

16. Project Schedule

Activity	Schedule
	Planned Start	Planned Finish
Develop and Test CCM PI Instrument and Systems	08/01/09	06/10/10
Develop and Test CCM PIRI Instrument and Systems	08/01/09	06/10/10
Develop and Test CCM Recall Bias Panel Study Instrument and Systems	12/15/09	3/15/10
Conduct Person Interview	08/14/10	10/02/10
Conduct Person Interview Reinterview	08/21/10	10/14/10
Conduct Recall Bias Study – Panel 1	5/7/10	5/24/10
Conduct Recall Bias Study – Panel 2	6/11/10	6/28/10
Conduct Recall Bias Study – Panel 3	9/10/10	9/27/10
Conduct Recall Bias Study – Panel 4	2/4/11	2/21/11

17. Request to Not Display Expiration Date

No exemption is requested. Expiration Date will be read to the respondent by the enumerator during the interview.

18. Exceptions to the Certification

There are no exemptions.

1 This is the original cost estimate before the sample reduction. The Census Bureau is currently updating the cost for the new sample size.