The Department of Commerce will submit to the Office of Management and Budget (OMB) for clearance the following proposal for collection of information under the provisions of the Paperwork Reduction Act (44 U.S.C. chapter 35).
Agency: U.S. Census Bureau.
Title: 2015 National Content Test
OMB Control Number: 0607-XXXX.
Questionnaire Cover Letters
DE-17(L1)Start Printed Page 29610
Internet Instrument Spec
Reinterview Instrument Spec (Coverage)
Reinterview Instrument Spec (Race)
Wording for Emails and Text Messages
Type of Request: New Collection.
Number of Respondents: 1.3 million households.
Average Hours per Response: 0.2.
Burden Hours: 216,667.
Estimated Burden Hours for 2015 National Content Test
| ||Total number of respondents||Estimated response time (minutes)||Estimated burden hours|
Needs and Uses: During the years preceding the 2020 Census, the Census Bureau will pursue its commitment to reducing the cost of conducting the next decennial census while maintaining the highest data quality possible. A primary decennial census cost driver is the collection of data from members of the public for which the Census Bureau received no reply via initially offered response options. We refer to these cases as nonrespondents. Increasing the number of people who take advantage of self response options (such as completing a paper questionnaire and mailing it back to the Census Bureau, or responding via telephone or Internet alternatives) can contribute to a less costly census.
The 2015 National Content Test (NCT) is part of the research and development cycle leading up to the 2020 Census.
The first objective of this test is to evaluate and compare different versions of questions about such things as race and Hispanic origin, relationship, and within-household coverage. The 2015 NCT is the primary mid-decade opportunity to compare different versions of questions prior to making final decisions for the 2020 Census. The test will include a reinterview to further assess the accuracy and reliability of the question alternatives about race and origin and within-household coverage.
For the decennial census, the Census Bureau adheres to the U.S. Office of Management and Budget's (OMB) October 30, 1997 “Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity” (see www.whitehouse.gov/omb/fedreg_1997standards) for classifying racial and ethnic responses. There are five broad categories for data on race: “White,” “Black or African American,” “American Indian or Alaska Native,” “Asian,” and “Native Hawaiian or Other Pacific Islander.” There are two broad categories for data on ethnicity: “Hispanic or Latino” and “Not Hispanic or Latino.” The OMB standards advise that respondents shall be offered the option of selecting one or more racial designations. The OMB standards also advise that race and ethnicity are two distinct concepts; therefore, Hispanics or Latinos may be of any race.
Additionally, the 1997 OMB standards permit the collection of more detailed information on population groups, provided that any additional groups can be aggregated into the standard broad set of categories. Currently, the Census Bureau collects additional detailed information on Hispanic or Latino groups, American Indian and Alaska Native tribes, Asian groups, and Native Hawaiian and Other Pacific Islander groups.
For example, responses to the race question such as Navajo Nation, Nome Eskimo Community, and Mayan are collected and tabulated separately in Census Bureau censuses and surveys, but also are aggregated and tabulated into the total American Indian or Alaska Native population. Similarly, responses to the race question such as Chinese, Asian, Indian, and Vietnamese are collected and tabulated separately, but also aggregated and tabulated into the total Asian population, while responses such as Native Hawaiian, Chamorro, or Fijian are collected and tabulated separately, but also tabulated, and aggregated into the total Native Hawaiian or Other Pacific Islander population. Responses to the ethnicity question such as Mexican, Puerto Rican, and Cuban are collected and tabulated separately, but also are tabulated and aggregated in Census Bureau censuses and surveys, but also tabulated and aggregated into the total Hispanic or Latino population.
The 2015 NCT will test ways to collect and tabulate detailed information for the detailed groups, not just to the broad groups identified above. Detailed data for specific White population groups, such as German, Irish, and Polish, and Black population groups, such as African American, Jamaican, and Nigerian, will be collected and tabulated, and may be aggregated into the total “White” or “Black or African American” populations respectively.
The 2015 NCT also includes testing of a separate “Middle Eastern or North African” (MENA) category and the collection of data on detailed MENA groups, such as Lebanese, Egyptian, and Iranian. Currently, following the 1997 OMB standards, Middle Eastern and North African responses are classified under the White racial category, per OMB's definition of “White.”Start Printed Page 29611
The second objective of the NCT is to test different contact strategies for optimizing self-response. The Census Bureau has committed to using the Internet as a primary response option in the 2020 Census. The 2015 NCT includes nine different approaches to encouraging households to respond and, specifically, to respond using the less costly and more efficient Internet response option. These approaches include altering the timing of the first reminder, use of email as a reminder, altering the timing for sending the mail questionnaire, use of a third reminder, and sending a reminder letter in place of a paper questionnaire to non-respondents.
One benefit of the Internet response mode is that it allows for more functionality and greater flexibility in designing questions compared to paper, which is constrained by space availability. The 2015 NCT will utilize web-based technology, such as the Internet, smart phones, and tablets to improve question designs, and to optimize reporting of detailed racial and ethnic groups (e.g., Samoan, Iranian, Blackfeet Tribe, Filipino, Jamaican, Puerto Rican, Irish, etc.).
Web-based designs also provide much more utility and flexibility for using detailed checkboxes and write-in spaces to elicit and collect data for detailed groups than traditional paper questionnaires, and will help collect data for both the broader OMB categories, as well as more detailed responses across all groups.
Components of the Test
Race and Origin Content
The 2015 NCT builds on extensive research previously conducted by the Census Bureau as part of the 2010 Census. One major study was the 2010 Census Race and Hispanic Origin Alternative Questionnaire Experiment (AQE) (for details, see www.census.gov/2010census/news/press-kits/aqe/aqe.html). The 2010 AQE examined alternative strategies for improving the collection of data on a race and Hispanic origin, with four goals in mind:
1. Increasing reporting in the standard race and ethnic categories as defined by the U.S. Office of Management and Budget;
2. Decreasing item non-response for these questions;
3. Increasing the accuracy and reliability of the results for this question; and
4. Eliciting detailed responses for all racial and ethnic communities (e.g., Chinese, Mexican, Jamaican, etc.).
Some of the findings from this research include:
- Combining race and ethnicity into one question did not change the proportion of people who reported as Hispanics, Blacks, Asians, American Indians and Alaska Natives, or Native Hawaiians and Other Pacific Islanders.
- The combined question yielded higher item response rates, compared with separate question approaches.
- The combined question increased reporting of detailed responses for most groups, but decreased reporting for others.
The successful strategies from the AQE research have been employed in the design of the Census Bureau's 2020 Census research. Four key dimensions of the questions on race and Hispanic origin are being tested in the 2015 NCT. These include question format, response categories, wording of the instructions, and question terminology.
The 2015 NCT will evaluate the use of two alternative question approaches for collecting detailed data on race and ethnicity. One approach uses two separate questions: The first about Hispanic origin, and the second about race. The other approach combines the two items into one question about race and origin. The 2015 NCT research will test both approaches with new data collection methods, including Internet, telephone, and in-person response. Each approach is described below, with its associated data collection modes.
1. Separate Race and Origin Questions (Paper and Internet)
This is a modified version of the race and Hispanic origin approach used in the 2010 Census. Updates since the 2010 Census include added write-in spaces and examples for the White response category and the Black or African American response category, removal of the term “Negro,” and the addition of an instruction to allow for multiple responses in the Hispanic origin question.
2. Combined Question With Checkboxes and Write-Ins Visible at Same Time (Paper)
This is a modified version of the combined question approaches found to be successful in the 2010 AQE research. Checkboxes are provided for the U.S. Office of Management and Budget (OMB) broad categories (per the 1997 Standards for the Classification of Federal Data on Race and Ethnicity), with a corresponding write-in space for detailed response to each checkbox category. In this version, all checkboxes and write-in spaces are visible at all times. Each response category contains six example groups, which represent the diversity of the geographic definitions of the OMB category. For instance, the Asian category examples of Chinese, Filipino, Asian, Indian, Vietnamese, Korean, and Japanese represent the six largest detailed Asian groups in the United States, reflecting OMB's definition of Asian (“A person having origins in any of the original peoples of the Far East, Southeast Asia, and the Indian subcontinent.”). Respondents do not have to select an OMB checkbox, but may enter a detailed response in the write-in space without checking a category.
3. Combined Question With Major Checkboxes, Detailed Checkboxes, and Write-Ins (Paper)
This is a modified version of the combined question approaches found to be successful in the 2010 AQE. Checkboxes are provided for the OMB categories, along with a series of detailed checkboxes under each major category, and a corresponding write-in space and examples to elicit and collect all other detailed responses within the major category. In this version, all checkboxes and write-in spaces are visible at all times. Again, the detailed response categories represent the diversity of the geographic definitions of the OMB category.
For instance, under the Asian category (and major checkbox), a series of detailed checkboxes is presented for Chinese, Filipino, Asian Indian, Vietnamese, Korean, and Japanese, which represent the six largest detailed Asian groups in the United States. Then, instructions to enter additional detailed groups (with the examples of “Pakistani, Thai, Hmong, etc.”) precede a dedicated write-in area to collect other detailed responses. Again, these detailed groups reflect OMB's definition of Asian (“A person having origins in any of the original peoples of the Far East, Southeast Asia, and the Indian subcontinent.”). Respondents do not have to select an OMB checkbox, but may enter a detailed response in the write-in space without checking a category.
4. Combined Question With Major Checkboxes and Write-Ins on Separate Screens (Internet)
In this version, the detailed origin groups are solicited on subsequent screens after the OMB response categories have been selected. On the first screen, the OMB checkbox categories are shown along with their six representative example groups. Once Start Printed Page 29612the OMB categories have been selected, one at a time, subsequent screens solicit further detail for each category that was chosen (e.g., Asian), using a write-in space, with examples, to collect the detailed groups (e.g., Korean and Japanese). The intent is to separate mouse click tasks (checkbox categories) and typing tasks (write-ins) in an attempt to elicit responses that are more detailed. This approach was used as one of three race and origin Internet panels in the 2014 Census Test.
5. Combined Question Branching With Detailed Checkbox Screens (Internet)
This version is an alternative method of soliciting detailed origin groups using separate screens, detailed checkboxes, and write-in spaces. On the first screen, the OMB checkbox categories are shown along with their six representative example groups. Once the OMB categories have been selected, one at a time, subsequent screens solicit further detail for each category, this time using a series of additional checkboxes for the six largest detailed groups (e.g., Chinese, Filipino, Asian, Indian, Vietnamese, Korean, and Japanese) with a write-in space also provided to collect additional groups.
Race Response Categories
The 2015 NCT will also evaluate the use of a “Middle Eastern or North African” (“MENA”) response category. There will be two treatments for testing this dimension:
1. Use of a MENA category: This treatment tests the addition of a MENA checkbox category to the race question. The MENA category is placed within the current category lineup, based on estimates of population size, between the category for American Indians and Alaska Natives and the category for Native Hawaiians and Other Pacific Islanders. With the addition of this new category, the “White” example groups are revised. The Middle Eastern and North African examples of Lebanese and Egyptian are replaced with the European examples of Polish and French. The MENA checkbox category will have the examples of Lebanese, Iranian, Egyptian, Syrian, Moroccan, and Algerian. All other checkbox categories and write-in spaces remain the same.
2. No separate MENA category: This treatment tests approaches without a separate MENA checkbox category, and represents the current OMB definition of White (“A person having origins in any of the original peoples of Europe, the Middle East, or North Africa.”). Here the category will provide examples of Middle Eastern and North African origins (e.g., Lebanese; Egyptian) along with examples of European origins (e.g., German; Irish) as part of the “White” racial category.
Wording of the Instructions
The 2015 NCT will evaluate the use of different approaches for wording the instructions used to collect data on race and ethnicity. The 2010 AQE research found that respondents frequently overlook the instruction to “Mark [X] one or more boxes” and have difficulty understanding the instructions. From the 2010 AQE qualitative research we learned that some respondents stop reading the instruction after noticing the visual cue [X] and proceed directly to do just that—mark a box—overlooking the remainder of the instruction. The new instruction being tested in the 2015 NCT (“Mark all boxes that apply”) is an attempt to improve the clarity of the question and make it more apparent that more than one group may be selected. The following options will be tested in the 2015 NCT.
1. “Mark [X] one or more”: One version (old instructions) will advise respondents to, “Mark [X] one or more boxes AND print [origins/ethnicities/details].”
2. “Mark all that apply”: An alternative version (new instructions), will advise respondents to, “Mark all boxes that apply AND print [origins/ethnicities/details] in the spaces below. Note, you may report more than one group.”
Instructions for American Indian and Alaska Native (AIAN) Write-In Area
The 2015 NCT will also examine different instructions to optimize detailed reporting within the AIAN write-in area. From the 2010 AQE research and recent 2014 qualitative research that the Census Bureau conducted with American Indians, Alaska Natives, and Central and South American Indian respondents, we know the instruction to “Print enrolled or principal tribe” causes confusion for many AIAN respondents and means Start Printed Page 29613different things to different people. The research found that AIAN respondents were confused by the use of different terms and concepts (e.g., “enrolled”, “affiliated,” “villages,” “race,” “origin,” “tribe,” etc.) and there was disagreement among focus group participants as to what “affiliated tribe” or “enrolled” or “villages” meant.
The overwhelming sentiment from 2014 AIAN focus group participants was that they want to be treated equally with other race/ethnic groups, and this was accomplished by not using different terminology (i.e., enrolled, affiliated, villages, etc.). Asking “What is your race or origin?” in conjunction with “Print, for example, . . .” (along with AIAN example groups) allowed the respondents to understand what the question asked them to report (their race or origin) and did not limit their write-in response by confounding the instructions with terms that mean different things to different people (e.g., tribes, villages, etc.). Therefore, the instruction to, “Print, for example, . . .” presented a viable alternative for further exploration in 2015 NCT research.
Based on the findings and recommendations from this research, the 2015 NCT will test variations of the instructions for the AIAN write-in area. We plan to test the instruction, “Print enrolled or principal tribe, for example . . .” on control versions, and the instruction, “Print, for example . . .” on experimental versions, to see how they perform.
The 2015 NCT will evaluate the use of different conceptual terms (e.g., origin, ethnicity, or no terms) in the wording of questions for collecting data on race and ethnicity. Recent qualitative focus groups and qualitative research (e.g., 2010 AQE research; 2013 Census Test research; cognitive pre-testing for the 2016 American Community Survey (ACS) Content Test) found that the terms “race,” “ethnicity,” and “origin” are confusing or misleading to many respondents, and mean different things to different people. The 2010 AQE research tested the removal of the term “race” from the question, and showed no evidence that removal of the term had any effect on either unit or item response rates. Recent cognitive research for the 2016 ACS Content Test tested an open-ended instruction (“Which categories describe you?”) and found that respondents did not have issues with understanding what the question was asking. The following options will be tested in the 2015 NCT.
1. “Origin” term: The current version of the race and Hispanic origin questions, and the combined question, use the terms “race” and/or “origin” to describe the concepts and groups in the question stem and/or instructions. For instance, in the combined race and Hispanic origin approach, the question stem is “What is Person 1's race or origin?”
2. “Ethnicity” term: One alternative option being explored tests the use of both the terms “ethnicity” along with “race” in the question stem and/or instructions (e.g., “What is Person 1's race or ethnicity?”).
3. NO terms: A second alternative option being explored tests the removal of the terms “race,” “origin,” and “ethnicity” from the question stem and instructions. Instead, a general approach asks, “Which categories describe Person 1?”
Two versions of the relationship question will be tested. Both versions are the same as those used in a split-sample in the 2014 Census Test, with no changes. These relationship categories were previously tested in other Census Bureau surveys including the American Housing Survey, American Community Survey, and the Survey of Income and Program Participation (currently used in production). Although research to date has been informative, leading to the development of the revised relationship question, additional quantitative testing is needed. Because the incidence of some household relationships—such as same-sex couples—is relatively low in the general population, the revised question needs to be tested with large, nationally representative samples prior to a final decision to include them in the 2020 Census questionnaire.
The first version uses the 2010 Census relationship question response options, but in a new order, starting with “husband or wife” and then the “unmarried partner” category. This version also re-introduces the foster child category, which was removed from the 2010 Census form due to space issues.
The second version includes the same basic response options as the 2010 Census version, but modifies/expands the “husband or wife” and “unmarried partner” categories to distinguish between same-sex and opposite-sex relationships.
Coverage Content (Internet Only)
The 2012 NCT experimented with several methods to improve within-household coverage for Internet respondents. One benefit of the online response mode is that it allows for more functionality and greater flexibility in designing questions compared to paper, which is constrained by space availability. The 2012 NCT included a coverage follow-up reinterview to evaluate the different Internet design options, but some results were inconclusive. In the 2015 NCT, two designs will be tested to compare different approaches for helping respondents provide a more accurate roster of household residents.
The first approach is the “Rules-Based” approach, and will allow us to see whether the presence of a question asking the number of people in the household, along with the residence rule instructions, helps respondents create an accurate roster. This is similar to the approach used across all modes in Census 2000 and the 2010 Census, where the respondent was expected to understand various applications of our residence rules and apply them to their household. The roster creation is followed by a household-level question that probes to determine if any additional people not listed originally should be included for consideration as residents of the household (several types of people and living situations are shown in a list).
The “Question-Based” approach allows us to ask guided questions to help improve resident responses. Respondents are not shown the residence rule instructions and are only asked to create an initial roster of people they consider to be living or staying at their address on Census Day. This is followed by several short household-level questions about types of people and living situations that might apply to people in the household that were not listed originally.
The materials mailed to the respondents will inform them that the survey is mandatory in accordance with Title 13, United States Code, Sections 141 and 193. This information also will be available via a hyperlink from within the Internet Instrument.
The results of the 2015 NCT will help guide the design of additional 2020 Census testing later this decade. The 2015 NCT will be the only opportunity to test content with a nationally representative sample prior to the 2020 Census. Testing in 2015 is necessary to establish recommendations for contact strategies, response options, and content options that can be further refined and tested in later tests. At this point in the decade, the Census Bureau needs to acquire evidence showing whether the strategies being tested can reduce the cost per housing unit during a decennial census, while providing high quality and accuracy of the census data. The nationally-representative sample is Start Printed Page 29614designed to ensure that the unbiased estimates from this test accurately reflect the nation as a whole, across a variety of demographic characteristics.
Along with other results, the response rates to paper and Internet collection will be used to help inform 2020 Decennial program planning and cost estimation metrics values. In addition, several demographic questions and coverage probes are included in this test to achieve improved coverage by future decennial censuses and surveys.
Information quality is an integral part of the pre-dissemination review of the information disseminated by the Census Bureau (fully described in the Census Bureau's Information Quality Guidelines). Information quality is also integral to the information collections conducted by the Census Bureau and is incorporated into the clearance process required by the Paperwork Reduction Act.
Affected Public: Individuals or Households.
Frequency: One Time.
Respondent's Obligation: Mandatory.
Legal Authority: Title 13 U.S.C. 141 and 193.
This information collection request may be viewed at www.reginfo.gov. Follow the instructions to view Department of Commerce collections currently under review by OMB.
Written comments and recommendations for the proposed information collection should be sent within 30 days of publication of this notice to OIRA_Submission@omb.eop.gov or fax to (202) 395-5806.
Dated: May 14, 2015.
Management Analyst, Office of the Chief Information Officer.
[FR Doc. 2015-12140 Filed 5-21-15; 8:45 am]
BILLING CODE 3510-07-P