[Federal Register Volume 67, Number 110 (Friday, June 7, 2002)]
[Notices]
[Pages 39339-39341]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 02-14242]


-----------------------------------------------------------------------

DEPARTMENT OF COMMERCE

Census Bureau


Census 2003 Test; Proposed Collection; Comment Request

ACTION: Proposed collection; comment request.

-----------------------------------------------------------------------

SUMMARY: The Department of Commerce, as part of its continuing effort 
to reduce paperwork and respondent burden, invites the general public 
and other Federal agencies to take this opportunity to comment on 
proposed and/or continuing information collections, as required by the 
Paperwork Reduction Act of 1995, Public Law 104-13 
(44.U.S.C.3506(C)(2)(A)).

DATES: Written comments must be submitted on or before August 6, 2002.

ADDRESSES: Direct all written comments to Madeleine Clayton, 
Departmental Paperwork Clearance Officer, Room 6608, 14th and 
Constitution Avenue, NW. Washington, DC 20230 (or via the Internet at 
[email protected]).

FOR FURTHER INFORMATION CONTACT: Requests for additional information or 
copies of the information collection instruments and instructions 
should be directed to Suzanne Fratino, U.S. Census Bureau, Building 2, 
Room 2021, Washington, DC 20233-9200, 301-457-4134.

SUPPLEMENTARY INFORMATION

I. Abstract

    In Census 2000, the Census Bureau conducted four separate tests 
examining innovative ideas. One of these ``experiments'' was the 
Response Mode and Incentive Experiment (RMIE). RMIE attempted to 
measure the extent to which respondents choose to use electronic 
response options including Computer Assisted Telephone Interviewing 
(CATI), Interactive Voice Response (IVR), and Internet. Preliminary 
findings from the RMIE initial mailout component and Operator 
Assistance indicate that Computer Assisted Telephone Interviewing does 
not offer clear advantages relative to the Internet in terms of 
increasing the overall response rate. The IVR mode showed promise but 
requires additional design work while the Internet mode yielded 
relatively high data quality. One major recommendation resulting from 
the RMIE was to investigate the best ways to present the availability 
of response options, and how to word messages included with the mailed 
questionnaire. To take advantage of evolving technology, the Census 
Bureau needs to research various self-response options toward 
developing a strategy that encourages the public to respond to the 
census using either paper or electronic options before Nonresponse 
Follow-up (NRFU) occurs. The method and optimum timing to contact, 
inform and, remind the public should be included.
    The Census Bureau is planning a two-part test in 2003. The first 
part will examine the impact of offering various self-response options 
and the interactions among various options on overall response rates 
and data quality. These options include mail, Internet, Interactive 
Voice Recognition (IVR), and a combination of Internet and IVR. This 
test is also designed to address questions about the relative timing 
and content of various contacts. We hope to answer the following 
questions: (1) What is the effect of offering alternative data 
collection modes on response (i.e. increase, decrease, shift)? and (2) 
what is the effect of new or additional contact strategies on overall 
response?
    The goal of this portion of the test is to identify, for further 
testing in 2004, the best strategy for increasing self-enumerated 
response to the census thus reducing the NRFU workload. Successful 
accomplishment of this goal will greatly improve the data quality of 
Census 2010 while reducing the cost of data collection.
    The second part of the Census 2003 Test will assess the effects of 
dropping the ``Some other race'' response option. This test is designed 
to answer whether item nonresponse to the race question will increase 
if the ``Some other race'' response option with a write-in line is 
deleted, and what effect this will have on the overall quality of race 
reporting. In past decennial censuses, the Census Bureau has received 
an exception from the Office of Management and Budget which allowed it 
to include a ``Some other race'' category. This category is a source of 
noncomparability between the census and surveys and race data produced 
by other agencies. The purpose of this test is to develop and evaluate 
a mailout version of the race question that conforms to OMB standards 
by excluding the ``Some other race'' category. It will also measure the 
effectiveness of revised instructions for the Hispanic origin and race 
questions to convey to respondents the intent of the questions; more 
specifically that different responses are being requested in each of 
these questions. In addition, revisions to the Hispanic origin 
question, including the addition of examples of Hispanic groups to 
obtain more complete reporting of detailed Hispanic subgroups are to be 
tested. Examples for the Other Asian and the Other Pacific Islander 
response categories to the question on race also will be included. It 
is desirable to assess the feasibility of these changes to the 
questions on race and Hispanic origin so that alternatives can be 
developed and tested in a timely way before final question versions are 
adopted. The Census Bureau plans to conduct multiple rounds of 
cognitive testing to identify problems and revise question wordings and 
instructions before finalizing them for this test.
    The goal of the race and Hispanic origin portion of the test is to 
develop question wording and content that will lead to improved self-
reporting of both race and Hispanic origin in the census.

[[Page 39340]]

II. Method of Collection

    The methodology for the Census 2003 Test consists of a data 
collection strategy involving fourteen different experimental panels. 
The control panel is a mailing strategy comprised of four pieces--an 
advance letter, an initial questionnaire, a reminder postcard, and a 
replacement questionnaire targeted to non-responding housing units. 
Essentially, this control panel is similar to the Census 2000 mailout 
strategy with the addition of a replacement questionnaire. In addition, 
the timing of each mail piece is different from Census 2000. The 
questionnaire used in nine of the panels will be a Census 2000 short 
form. The remaining six panels will use a Census 2000 short form with 
changes to both the Hispanic origin and race questions, their response 
categories, and instructions to answer both questions. ``Census Day,'' 
the reference date for enumerating respondents, will be February 6, 
2003. The advance letter will be delivered to housing units in the 
sample by the United States Postal Service between January 22 and 24, 
2003. The initial questionnaire will be delivered on January 28 and 30, 
followed by the reminder postcard during February 3-5. On February 10, 
we will determine the universe of non-respondents who will be mailed a 
replacement questionnaire on February 15-18, 2003.
    A national sample of 220,000 addresses will be selected from 
housing units in Census 2000. The sample is restricted to addresses in 
Mailout/Mailback areas that are not in the American Community Survey 
sample during the test period. Based on Census 2000 return rates, 
census blocks will be stratified into high response and low response 
strata. A random sample of 5,000 housing units will be drawn from each 
stratum for each of the eight response strategy test panels, yielding a 
total of 10,000 housing units per panel. For the control panel and each 
of the six race and ethnicity panels, a sample of 10,000 housing units 
from each stratum will be selected, yielding a total of 20,000 housing 
units per panel.
    The eight response strategy test panels consist of various 
treatments providing alternatives and additions to the control panel's 
mailing strategy. The sample households in one panel will have the 
option of responding via the Internet in addition to the option of 
completing a paper questionnaire and returning it by mail. Two other 
panels test a telephone interactive voice recognition (IVR) system as 
an alternative to mailing back the paper questionnaire. The distinction 
among these two panels is the extent to which residents are encouraged 
to choose the IVR option instead of mail. One panel will encourage 
residents to respond by telephone without including a paper 
questionnaire and the second will give them the option of responding by 
telephone or with a questionnaire. Two panels, one without an initial 
questionnaire in the envelope, will give residents both the Internet 
and IVR as response options. Other response strategy treatment panels 
include using a telephone call reminder in lieu of a reminder postcard, 
putting a due date on the questionnaire envelope, and a mailing 
strategy without a replacement questionnaire.
    Responses from paper mail returns, the Internet, and IVR will be 
data captured in order to analyze the demographic characteristics of 
respondents and patterns of item nonresponse. Results of the test will 
help shape the data collection strategy for the next census.
    The six additional test panels are designed to test the effects on 
the overall and item nonresponse of changes to the questions on 
Hispanic origin and race. The purpose of this test is to examine the 
effects of dropping the ``Some other race'' response category from the 
race question, and whether additional instructions can ameliorate the 
resulting increase in race item nonresponse expected, as well as convey 
to respondents the intent of this question. Previously, the 
overwhelming majority of responses in the ``Some other race'' category 
were Hispanic ethnicities. It is vital that respondents understand that 
the intent of the question on race is for them to self-report their 
race using one or more of the race categories shown on the form. In 
addition, revisions to the Hispanic Origin question, including adding 
examples of Hispanic groups to obtain more complete reporting of 
detailed Hispanic subgroups, are being tested.
    Because of the listing of Asian and Pacific Islander ethnicities, 
along with other design effects of the question, some respondents think 
we are asking them to report their ethnicity and not their race. Others 
do not see a difference between race and ethnicity. We also are 
including Other Asian and Other Pacific Islander examples to obtain 
more complete reporting of detailed Other Asian and Other Pacific 
Islander subgroups. The six panels test the effects of the following 
changes compared to the control panel.
    1. A modified Hispanic origin question, including the addition of 
the word ``origin,'' slight revisions to the instruction for the 
question, and removing the slashes (/). (The same modified Hispanic 
question is used in all six panels.)
    2. The inclusion of examples of Hispanic groups and Other Asian and 
Other Pacific Islander groups to obtain more complete reporting of 
detailed other Asian and Other Pacific Islander subgroups.
    3. The deletion of the ``Some other race'' response option and 
write-in area.
    4. The deletion of the ``Some other race'' response option and 
write in area and the addition of examples of Hispanic groups and Other 
Asian and Pacific Islander groups.
    5. The deletion of the ``Some other race'' response and addition of 
an ``informative instruction'' to increase respondents awareness that 
race and Hispanic origin are different.
    6. The deletion of the ``Some other race'' response, addition of 
``informative instruction'' to increase respondents awareness that race 
and Hispanic origin are different and, adding examples of Hispanic 
groups and Other Asian and Other Pacific Islander groups.
    Responses from these paper mail returns also will be data captured 
in order to analyze the demographic characteristics of respondents and 
patterns of item nonresponse.

III. Data

    OMB Number: Not available.
    Form Number(s): DA-1A, 1B, 1C, 1D, 1DD, DA-1(CC-9), 10, 11, 12, 13.
    Type of Review: Regular.
    Affected Public: Individuals or households.
    Estimated Number of Respondents: 220,000.
    Estimate Time Per Response: 10 minutes.
    Estimated Total Annual Burden Hours: 36,666 hours.
    Estimated Total Annual Cost: There is no cost to respondents except 
for their time to respond.
    Respondent Obligation: Mandatory.
    Legal Authority: Title 13 of the United States Code, Sections 141 
and 193.

IV. Request for Comments

    Comments are invited on: (a) Whether the proposed collection of 
information is necessary for the proper performance of the functions of 
the agency, including whether the information shall have practical 
utility; (b) the accuracy of the agency's estimate of the burden 
(including hours and cost) of the proposed collection of information; 
(c) ways to enhance the quality, utility, and clarity of the 
information to be collected; and (d) ways to minimize the burden of the 
collection of information on respondents, including through the use of 
automated collection techniques

[[Page 39341]]

or other forms of information technology.
    Comments submitted in response to this notice will be summarized 
and/or included in the request for OMB approval of this information 
collection; they also will become a matter of public record.

    Dated: June 3, 2002.
Madeleine Clayton,
Departmental Paperwork Clearance Officer, Office of the Chief 
Information Officer.
[FR Doc. 02-14242 Filed 6-6-02; 8:45 am]
BILLING CODE 3510-07-P