Supporting StatementB revised Aug 2012

Supporting Statement B

U.S. Department of Commerce

U.S. Census Bureau

BC-1294, Field Representative Exit Questionnaire

BC-1294(D), Decennial Field Staff Exit Questionnaire

OMB Control No. 0607-0404

2. Collection of Information Employing Statistical Methods

1. Universe and Respondent Selection

The BC-1294 universe will be composed of all current survey interviewers who have voluntarily resigned during a given sampling period. The sample does not include interviewers that were terminated for cause (i.e. fired). A random sample of one-half of all former current survey interviewers - stratified by regional office - will be drawn. The sample size will vary since it is dependent on the universe size, which can vary from sampling period to sampling period. However, we estimate that we will not sample more than 500 former field representatives or current survey interviewers annually.

Based on the current 2020 Census inter-decade testing plan, we do not anticipate the conduct of decennial site tests or Coverage Measurement activities.

2. Procedure for Collecting Information

A random sample, stratified by regional office, will be drawn from a list of former current survey interviewers. The method used is systematic and employs a computer-based random number generator. Means and frequencies are the primary (descriptive) statistics used. Categorical measures of association (e.g., chi-square based techniques) are the primary analytical and inferential measures used. The proposed sample sizes are large enough to detect statistically significant differences (between regions) with 90% confidence, and to calculate sample estimates with a statistical precision of +/- 4 percentage points. Attitude and demographic trends analyses have been established at a significance level of alpha = 0.10, which is adequate to identify trends associated with turnover and thereby assist in making management decisions. Sampling is periodic (i.e., approximately quarterly) in a given year. A telephone methodology is used for the actual interviewing.

2. Methods to Maximize Response Rates

A telephone methodology is employed, which has historically yielded response rates greater than those obtained from similar mail out/mail back strategies, especially when the collection interval is relatively short and the audience is former employees. In addition, using a telephone methodology allows repeated attempts to be easily and inexpensively made to follow up on non-responding cases. In the past couple of years, we have achieved an average response rate of 66 percent.

2. Testing of Procedures

The BC-1294(D) was tested in 2000, 2004 and 2006. The BC-1294 was tested between 2000 and 2007. Current changes to the BC-1294 resulted from testing conducted in August-October, 2011. Feedback showed that we needed to include the realignment questions and/or categories presented in Attachment 1.

2. Contacts for Statistical Aspects of Data Collection

The sampling plan was prepared by Darlene Moul (301-763-1935). She may also be contacted about questions regarding data collection.

List of Attachments:

1. BC-1294, Field Representative Exit Questionnaire and proposed changes since the last request for clearance.

2. BC-1294(D), 2010 Census Decennial Field Staff Exit Questionnaire

2