Supporting Statement B - AABS

Supporting Statement B: Annual Arts Benchmarking Survey
B.

COLLECTION OF INFORMATION EMPLOYING STATISTICAL
METHODS

B.1. Universe and Respondent Selection
The Census Bureau will conduct the AABS as a February supplement in conjunction with
the CPS. The universe is the more than 100 million housing units in the 3,141 counties
and independent cities that make up the United States. From this universe, the Bureau
will select a sample of approximately 72,000 households each month, of which
approximately 60,000 households are eligible to be interviewed (most ineligible housing
units are non-residential or unoccupied). About 55,500 households will be interviewed
each month. The Bureau will administer the AABS supplement to a random one-quarter
of the 60,000 eligible households, or the approximately 15,000 households who are in
two of the CPS months in sample rotations. In larger households, a second supplemental
AABS interview will be administered. The Census Bureau estimated that based on the
2008 SPPA we should expect that there will be approximately 16,000 adults eligible for
the AABS supplements. Since the AABS allows proxy responses for spouses or partners
we expect that data will be collected on approximately 18,000 adults. This is what
occurred in the 2008 SPPA using the same sampling procedures and also using two CPS
months in sample rotations. The exact number of interviews will ultimately depend on
the supplemental response rate. However, based on prior experience and the current
design that permits some proxy responses, we expect the supplement response rate to be
close to the overall CPS response rate, which is approximately 81.6 percent.

B.2. Procedures for Collecting Information
The main data for the Annual Arts Benchmarking Arts Attendance and Arts Learning
survey will be collected as a supplemental survey associated with the February 2013
CPS. The main data for the Annual Arts Benchmarking Arts Creation and Performance
survey will be collected as a supplemental survey associated with the February 2014
CPS. Attachment K gives an overview of the CPS sample design and weighting.

B.3.

Methods for Maximizing Response Rate

One of the main reasons that the NEA has chosen to have the AABS as a supplement to
the CPS is the fact that the CPS has historically achieved higher response rates than what
would otherwise be possible. To avoid breakoffs we have kept the number of
supplemental questions to a minimum. We also have plans (described in the next
section) to test each question to ensure that the questions are easy to answer and will not
lead to any significant item nonresponse. As with the SPPA, to avoid lower response
rates we have given the Census Bureau permission to collect the information via proxy
from a knowledgeable adult in the event that the target respondent is unavailable during
1

the CPS field period. Finally, the Census Bureau’s final AABS supplement survey
weights include a nonresponse adjustment which reduces potential nonresponse bias.

B.4.

Test of Procedures and Methods

Prior to the Census Bureau’s fielding of the AABS supplement, the questions will be
pilot-tested by Silber & Associates. Specifically, the Arts Attendance and Arts Learning
questions (Attachment D) will be pilot-tested by trained telephone interviewers using a
computer-assisted telephone instrument with randomly selected adults from 300
households in fall 2011. We do not have plans to conduct separate cognitive interviews
because these questions were cognitively tested for the SPPA study. The sample for the
pilot test will be a national random-digit dial probability sample that will include both
landline telephone and cell phone exchanges. The telephone interviewers will be
instructed to record all comments and concerns that respondents may raise during the
interview. At the end of the interview, respondents will be asked if they had difficulty
answering any of the questions. The pilot test will include a few demographic questions
(age, sex, race, education) that will be used to determine whether any particular group
had difficulty answering the questions.
Given that most of the AABS questions are taken directly from the SPPA1, we expect that
the pilot study will likely lead to only small revisions to wording or perhaps changes in
question order, but will unlikely lead to any significant changes. However, we will
provide OMB with the final version after completing the pilot study and will revise the
overall OMB package if requested. The same pilot-testing procedures described above
will be used in fall 2012 to test the Arts Creation and Performance questions.

B.5.

Consultations on Statistical Aspects of the Design

The Census Bureau will issue an official source and accuracy statement2 that documents
the estimation procedures for the AABS supplement including a detailed discussion on
how to properly report accuracy of the estimates. In addition, NEA will either provide
data users with a user’s guide or modify the current SPPA user’s guide (attachment J) to
include information on how to use the AABS data.
The following is contact information for the firms that have been involved in the
statistical design of the AABS studies and will likely assist NEA in creating a user’s
guide as well as help data users with their statistical analyses:

Bohne G. Silber, Ph.D.
Silber & Associates
1

Attachments H, I, and J contain additional information on the SPPA, including the 2012 SPPA pilot study
results, 2008 CPS-SPPA Source and Accuracy Statement, and the 2008 SPPA User’s Guide.
2 We expect the source and accuracy statement to be similar to the SPPA version shown in attachment I.

2

13067 Twelve Hills Road, Suite B
Clarksville, Maryland 21029-1144
(410) 531-2121, ext. 11
[email protected]
Timothy Triplett, Senior Survey Statistician
The Urban Institute
2100 M Street, NW
Washington, DC 20037
(202) 261-5579
Attachments
A.

Title 13, Section 182, United States Code

B.

First Federal Register Notice, Vol. 76, No. 109, p. 32992

C.

Second Federal Register Notice, Vol. 76, No. 171, pp. 54807-54807

D.

AABS Questionnaire: Arts Attendance and Arts Learning

E.

AABS Questionnaire: Arts Creation and Performance

F.

CPS Advance Letter

G.

Confidentiality Brochure

H.

2012 SPPA Pilot Results

I.

Source and Accuracy Statement (2008 CPS/SPPA)

J.

2008 SPPA User’s Guide: http://www.nea.gov/research/SPPA/users-guide.pdf

K.

Overview of CPS Sample Design and Methodology

3