Supporting Statement B (1220-0176)

SUPPORTING STATEMENT

B. COLLECTION OF INFORMATION EMPLOYING STATISTICAL METHODS

1. Universe and Respondent Burden

The September 2008 supplement will be conducted in conjunction with the CPS, for which the universe is 111 million households. From this universe, we select a sample of approximately 72,000 households each month, of which approximately 60,000 households are eligible for interviews. About 55,000 households are actually interviewed each month. The items in the supplement are asked, as appropriate, for members of the households.

2. Procedures for Collecting Information

This is a supplemental survey associated with the September 2008 CPS. The statistical properties of these supplemental items will fall within those associated with the CPS itself. (Attachment G)

3. Methods to Maximize Response

Response rates and data accuracy for the CPS are maintained at high levels through internal consistency edits in the computerized instrument, interviewer instructions, training, and close monitoring of these data. The supplement type A nonresponse rate was 12.2 in 2002, 13.5 percent in 2003, 13.1 percent in 2004, 12.6 percent in 2005, 13.9 percent in 2006, and 8.8 percent in 2007. This type A rate is set when the respondent refuses to continue prior to the supplement. After data processing, the criteria for non-interviews are more stringent - the first two questions must be equal to "refused" or "don't know." Under these more stringent criteria, the nonresponse rate for was 14.5 in 2006 and 9.8 in 2007.

4. Testing of Procedures

The new item proposed for the 2008 supplement was scheduled to be tested by the Census Bureau, but due to an error in the creation of the testing instrument an earlier draft of the question was used instead. This earlier draft of the question did not include the phrase “religious organizations” and asked about the actions of the respondent and anyone else in the household rather than just the respondent. Several issues were raised during the testing of this question , such as not being able to accurately respond for other members of the household and uncertainty about whether or not to include donations to religious organizations, that should be resolved by using the correctly worded question. Because of this error, we intend to behavior code the responses to the new question during implementation of the September 2008 supplement and analyze the results to determine data quality. Attachment H discusses the reliability of the estimates produced from the September 2007 Volunteer Supplement.

5. Contact Persons

The following individuals may be consulted concerning the statistical data collection and analysis operation:

Statistical Design:

Samson Adeshiyan

Demographic Statistical Methods Division

Bureau of the Census

Washington, D.C. 20233

(301) 763-5874

Statistical Analysis:

Diane Herz

Office of Employment and Unemployment Statistics

Bureau of Labor Statistics

Washington, D.C. 20212

(202) 691-6383

Data Collection/Survey Design:

Kathleen Creighton

Demographic Surveys Division

Bureau of the Census

Washington, D.C. 20233

(301) 763-3814

Attachments

A. Proposed Supplement Questions

B. Title 29, United States Code, Sections 1 through 9

C. News release from September 2007 survey

D. CPS Advance letter

E. Confidentiality Brochure

F. Title 13, United States Code

G. Overview of CPS Sample Design and Methodology

H. Source and Accuracy Statement for the September 2007

CPS Microdata file on Volunteers