Request for Nonsubstantive Change

The American Community Survey

0607-0810

Planned Nonsubstantive Changes to Data Collection Materials for Use in 2010

During 2010, households who receive the American Community Survey (ACS) may not understand that they are required to participate in both the ACS and the 2010 Census. An interdivisional workgroup examined all the ACS data collection materials and determined which materials should be revised and what new materials should be created to convey to ACS households the importance of participating in the both the ACS and the 2010 Census. A change request to these materials will be sent to OMB for approval in the Fall of 2009.

The following lists the ACS data collection materials that will be revised or are new for 2010:

• the prenotice/advance letter and the envelope in which it is mailed

• the questionnaire mailing package letters (initial and replacement) and the envelope in which the questionnaire is mailed

• new pre-recorded scripts about the ACS and the 2010 Census were added to the ACS toll free Telephone Questionnaire Assistance (TQA) phone line (the Interactive Voice Recognition (IVR) system)

• new Frequently Asked Questions (FAQs) were developed for the telephone and personal visit interviewers to use to explain the difference between the ACS and the 2010 Census

• a new flyer was created for use by personal visit interviewers to help to explain to respondents the difference between the ACS and the 2010 Census

The revised letters include new wording to inform respondents that they will receive both the ACS questionnaire and the 2010 Census questionnaire and that participation in both is required by law. The Group Quarters (GQ) letters will contain similar revisions.

The revised prenotice letter wording includes the following two paragraphs:

“The American Community Survey is a separate part of the 2010 Census program. Some households, including yours, will receive both the American Community Survey and the 2010 Census this year. Your response to both is required by U.S. law.

In their own way, the American Community Survey and the 2010 Census provide an important picture of America. They show not only the number of people who live here, but how we live as a Nation – our education, housing, jobs and more. The information you provide helps community leaders decide where schools, highways, hospitals, and other services are needed.”

The revised initial questionnaire package letter wording includes the following paragraph:

“Some households, including yours, will receive both the American Community Survey and the 2010 Census this year. Your response to both is required by U.S. law. The Census Bureau is required by U.S. law to keep your answers confidential.”

The revised replacement questionnaire package letter wording includes the following paragraph:

“We realize that you may have recently answered the 2010 Census. Some households, including yours, will receive both the 2010 Census and the American Community Survey this year. Answering both is important and is required by U.S. law. Your response is so important that a Census Bureau field representative may attempt to contact you by telephone or personal visit if we do not receive your questionnaire. The Census Bureau is required by U.S. law to keep your answers confidential.”

The ACS envelopes currently are printed on white paper with black text, with a window through which the sample address is visible. In addition, on the questionnaire envelope to the left of the address window is a text box with the survey name and the statement “Your response is required by law.” The test envelopes were revised to include the survey name and “U.S. Census Bureau” prominently in an outlined text box above the address window of each envelope. This is to distinguish the ACS mail pieces from the 2010 Census mail pieces (which have the 2010 logo above the address window of the envelope). In addition, a small sample of households will receive revised envelopes that have a green colored background in each of the text boxes on the envelopes. The envelopes and letters were all cognitively tested.

The FAQs for use by the ACS interviewers and also contained in the ACS IVR were developed in conjunction with 2010 Census staff. In addition, the flyer created for use by personal visit interviewers contains FAQs about the ACS and the 2010 Census on one side and illustrations of the cover page of the ACS form and a 2010 Census form on the opposite side. This will enable the interviewers to visually show respondents the difference between the two forms. This flyer was also cognitively tested.

To evaluate the effect of the revised letters and the revised envelope designs before, during and after the major 2010 Census activities, the 2010 ACS sample will be divided into four treatments for the January - November 2010 production panels. The chart below shows the treatments and the sample sizes for each month.

Treatments - ACS 2010	Jan-Feb	Mar-May	Jun-Nov	Dec.
Revised 2010 letters Revised 2010 envelopes with new text and green color	15,000	15,000	10,000
Revised 2010 letters Revised 2010 envelopes with new text, no color	15,000	Balance of sample	Balance of sample
2009 production letters Revised 2010 envelopes with new text, no color	10,000	10,000	10,000
2009 production letters 2009 production envelopes	Balance of sample	15,000	10,000	Entire sample

The 2010 ACS letter and envelope messaging study was designed to look at the impact of the new letters/envelopes on mail response in three time periods: Pre-census (January - February 2010), Census (March - May 2010), and Post-Census (June - November 2010). The analysis strategy is to examine the effect of color on the envelope first by comparing two of the treatments, and then comparing the treatment with the highest mail response rate to the current ACS production mailing. The minimum detectable difference for the mail response comparisons varies based on the time period in question as well as the specific treatments being compared. Assuming a 50% mail response rate, the minimum detectable difference ranges from .59 to 1.15 percentage points, using 80% power and 90% confidence level, depending on these factors.

ACS response rates in 2010 will be monitored closely. By testing a small panel with the 2009 production letters and the revised envelopes without color, we will be able to assess the impact of potentially retaining the revised envelope without color into 2011 and beyond. A final report on the 2010 treatments will be available in 2011.

Additionally, the USDA requested an immediate change to the Food Stamp question due to the new name (SNAP) on the questionnaire. There is also a minor change to the administrative portion of the Group Quarters (GQ) questionnaire.

There is no change to the burden.

Contact: Susan Schechter, Chief, American Community Survey Office, 301-763-8950 or [email protected].