OMB AHS13 Justification_PartB

SUPPORTING STATEMENT

B. Collection of Information Employing Statistical Methods

1. Universe and Respondent Selection

The 2013 AHS sample will consist of about 167,500 sample units. The sample is comprised of four components.

• 63,600 sample units selected primarily from the 1980 census sample files by variables, such as owner-occupied housing units, renter occupied housing units, vacant units for rent, and mobile homes. This sample includes 63,000 returning and 500 new housing units.

• 5,100 subsidized housing units selected from the 2009 Tenant Rental Assistance Certification System (TRACS) and the 2009 Public and Indian Housing Information Center (PIC) files provided by HUD.

• 65,300 large metro area supplemental sample units selected from areas interviewed in previous AHS-MS surveys between 1998 and 2009. This number includes approximately 2,500 cases in these areas that were never interviewed.

• 33,500 large metro area supplemental sample units selected from areas that have never been interviewed for AHS-MS.

We designed the survey using the following reliability requirement: A two-year change of 10% in the median monthly costs for 5% subgroups will have a standard error of 5%.

Please refer to the attached 2013 AHS Sample Design and Weighting document for detailed statistical methodology.

2. Procedures for Collecting Information

We will attempt to collect the data primarily through telephone interviews. Prior to 2011, telephone interviews were reserved for returning, occupied housing units with at least one household member remaining from the previous survey who had given permission to conduct further interviews by telephone. However, due to the large increase of new units in the 2011 sample, it was not feasible to perform personal visits at all of these units. The current policy is that all units will initially be contacted by telephone. Respondents who prefer not to be interviewed via phone will receive personal visits.

3. Methods to Maximize Response

Based on the 91 percent weighted response rate of the 2011 AHS (89 percent unweighted), the Census Bureau expects the 2013 AHS response rate to be between 87 and 90 percent. If an occupant is reluctant to participate, the FR informs the regional office staff, who sends a follow-up letter explaining the survey in greater detail and urging the occupant's cooperation. A Census Bureau FR or his/her supervisor will contact the occupant again.

4. Testing Procedures

The Census Bureau conducted a test of the interview forms and procedures in September 1983 and again in 1994 and 1995. Based on the results of the tests, HUD made modifications and changes.

In 2004, HUD and the Census Bureau conducted a major review of the survey questions for neighborhood quality, income, utility costs, and renter subsidies, making changes to each to improve the quality of this information. In October 2004, we tested these changes in a live hot house test held in Pittsburgh, Pennsylvania.

In 2010, questions in the updated Mortgage module were cognitively tested by the Center for Survey Measurement (CSM) at the Census Bureau. We determined that more work needed to be done on the module, so we reverted to the 2009 AHS version and flow of questions where the 2011 version fell short.

In 2012, questions in the new Public Transportation and Pedestrian Accessibility, Disaster Planning, and Doubled-Up Households modules were cognitively tested by the Center for Survey Measurement (CSM) at the Census Bureau before implementation in the 2013 questionnaire.

The Neighborhood Observation module, consisting primarily of questions from the 2009 AHS, was also cognitively tested. The Census Bureau and HUD reviewed and discussed all of CSM’s recommendations, electing to follow most of the recommendations. In the cases where CSM could not make a definitive recommendation, HUD made the final decision.

Ten questions in the Neighborhood Social Capital module were previously cleared by OMB for HUD’s Choice Neighborhood Demonstration -- baseline survey. The other 11 questions were drawn from existing surveys that have been field-tested for numerous years and implements with NORC at the University of Chicago.

1. Cognitive Testing of the Public Transportation and Pedestrian Accessibility Module

Nineteen questions were tested during the first round of cognitive testing. Following a review and discussion of the first round cognitive testing results between the Census Bureau and HUD, HUD modified seven questions as follows:

1. PTRINTRO: The phrase “this neighborhood” was added to the introduction.

2. PUBTYP: The “Other” category is not to be read out loud.

3. PTGROCER: The word “can” was changed to “could” to clarify that we are not asking about physical capabilities.

4. PTWORSHIP: The text “if you have one” was added for respondents who do not attend church.

5. WBGROCER: The word “can” was changed to “could” to clarify that we are not asking about physical capabilities.

6. WBWORSHIP: The text “if you have one” was added for respondents who do not attend church.

7. WBWKSCHL1: The text “if you have one” was added for respondents who do not work or attend school.

Nineteen questions were tested during the second round of cognitive testing. Following a review and discussion of the second round cognitive testing results between the Census Bureau and HUD, HUD made revisions to three questions as follows:

1. PTWORSHIP: This question was deleted, as asking about places of worship was found to be too sensitive.

2. BWWORSHIP: This question was deleted, as asking about places of worship was found to be too sensitive.

3. WBWKSCHL1: HUD added a “Not Applicable” category for people who do not work or attend school

2. Cognitive Testing of the Neighborhood Observation Module

Twenty questions were tested during the first round of cognitive testing. Following a review and discussion of the first round cognitive testing results between the Census Bureau and HUD, HUD modified two questions as follows:

1. EWATER: Added the phrase “For this survey, a half block is about 300 feet in length, or the length of a football field.”

2. EKRPG: An FR instruction was added, “Do not include residential parking lots.”

Twenty questions were tested during the second round of cognitive testing. Following a review and discussion of the second round cognitive testing results between the Census Bureau and HUD, no revisions were made.

3. Cognitive Testing of the Disaster Planning Module

Fourteen questions were tested during the first round of cognitive testing. Following a review and discussion of the first round cognitive testing results between the Census Bureau and HUD, HUD modified two questions as follows:

1. EVKIT: Removed the word “necessary.”

2. LTHOPT: This question was removed after determining that responses would likely be unreliable.

There was not a second round of cognitive testing of these questions.

4. Cognitive Testing of the Doubled-Up Households Module

Forty-six questions were tested during the first round of cognitive testing. Following a review and discussion of the first round cognitive testing results between the Census Bureau and HUD, HUD modified six questions and added two questions as follows:

1. MOVEOUT: Added the phrase, “I have listed [names of household roster members].”

2. New Question: “Has [name] lived in this household for less than 12 months?”

3. New Question: “When did [he/she] move in?”

4. XCOST: Reworded question to ask “When [name(s)] moved here, did the amount [he/she/they/you] paid for utilities and mortgage or rent increase, decrease, or stay the same as [his/her/their/your] previous residence?”

5. WHYLEFT: Added the phrase, “among household members” to answer category that previously read “there was too much conflict.”

6. WHYLEV: Added the phrase, “among household members” to answer category that previously read “there was too much conflict.”

7. Initiative: Reworded question to ask, “Did [name(s)] move out on [date] of their own initiative, at the request of others in the household, or for some other reason?”

8. UNPAIDTX: added the phrase, “or some other reason?”

5. Contacts for Statistical Aspects and Data Collection

HUD consulted the following individuals on the statistical data collection and analysis operation:

Tamara Cole

Associate Directorate for Demographic Programs

U.S. Census Bureau

(301) 763-4665

Statistical Design, Analysis, and Collection

Aref Dajani

Demographic Statistical Methods Division

U.S. Census Bureau

(301) 763-1797

Statistical Design

Kenya Tyndle

Demographic Surveys Division

U.S. Census Bureau

(301) 763-5388

Collection

Attachments