Back

Alternate Methods of Asking Race in Telephone Surveys

Preparing for the Y2K Changes

 

John M Kennedy

and

Nancy G Bannister

 

Indiana University

Center for Survey Research

1022 East Third Street, Bloomington, IN

Overview

This report presents data on four methods of asking respondents to classify their race in a telephone interview. In 1997, OMB revised Directive 15 that specified how racial and ethnic data are to be collected. The most significant change is that respondents will be allowed to select multiple races.

We tested alternative wordings and response patterns for administering the race categories question in telephone surveys conducted in fall 1998 and spring 1999.

The following tables show the race distributions for Indiana based on:

  • Four different methods of collecting race data using multiple categories
  • One method that allowed only one racial category
  • The 1990 Census race distributions
  • The 1997 population estimates of the racial and ethnic composition of Indiana
  • 1990 Indiana Poll

    What is your race? Are you:
  • White
  • Black, African-American
  • American Indian
  • Asian or Pacific Islander
  • Or, some other race
  • Interviewers were instructed to read each choice and accept only one answer.
  • If the respondent insisted on mentioning more than one, or responded with an ethnic or nationality response, the interviewers took a note and requested that the respondent select the listed race category that most closely described his/her race.
  •  

    Race distribution for the 1990 Indiana Poll and the 1990 Census for Indiana

     

    1990 Indiana Poll

    1990 Census Indiana

    White 91.2 90.6
    Black, African American 5.9 7.8
    American Indian or Alaskan Native 0.8 0.2
    Asian or Pacific Islander 0.7 0.7
    Something else 1.4 0.7

     

    The Indiana Poll

    The Indiana Poll is a statewide telephone survey of adult Indiana residents conducted by the Indiana University Center for Survey Research in Bloomington, Indiana. At each residential telephone number, a respondent was randomly selected to be interviewed from all the household members age 18 and older.

    1990 Indiana Poll

  • Interviews were conducted with 1010 adults.
  • Two-stage Mitofsky-Waksberg RDD
  • Conducted in April and May, 1990
  • 1998 Indiana PollA

  • Interviews were conducted with 507 adults
  • List-assisted RDD sample from the Genesys sampling system
  • Conducted in October, 1998
  • 1998 Indiana PollB

  • Interviews were conducted with 505 adults
  • List-assisted RDD sample from the Genesys sampling system
  • Conducted in November, 1998
  • 1999 Indiana Short Poll

  • Interviews were conducted with 561 adults
  • List-assisted RDD sample from the Genesys sampling system
  • There was oversampling of exchanges with higher incidence of African Americans
  • Conducted in March, 1999
  •  

    1998 Indiana PollA

    What is your race or ethnic background?
  • American Indian or Alaska Native?
  • Asian or Pacific Islander?
  • Black or African American?
  • Hispanic or Latino?
  • White or Caucasian?
  • Something else?
  • Interviewers were instructed not to read the choices unless necessary.
  • They could accept more than one response if the respondent volunteered more than one, but respondents were not told they could answer with more than one response.
  • Interviewers coded each category as no (not mentioned) or yes (mentioned).
  • "Hispanic or Latino" was included in this response set rather than as a separate variable.
  •  

    Race distribution for the 1998 Indiana PollA and 1997 Census estimates of race for Indiana

     

    1998 Indiana PollA

    1997 Census Indiana Estimates

    White or Caucasian 88.1 90.6
    Black or African American 5.1 8.2
    American Indian or Alaska Native 1.2 0.2
    Asian or Pacific Islander 1.0 0.9
    Something else 0.6  
    Hispanic or Latino 1.0  
    American Indian/Alaskan Native & White/Caucasian 0.6  
    White/Caucasian & Something else 0.4  
    Three or more races 0.6  
    Don't know 0.6  
    Refused 0.8  

     

    1998 Indiana PollB

    Please tell me your race, using one or more of the following categories. Please answer yes or no to each.
  • American Indian or Alaska Native?
  • Asian or Pacific Islander?
  • Black or African American?
  • White or Caucasian?
  • Something else?
  • Interviewers were instructed to read each category in the order given.
  • The respondents were instructed to respond with yes or no for each category as it was read.
  • Respondents were explicitly invited to report more than one category.
  • No Hispanic/Latino question.
  •  

    Race distribution for the 1998 Indiana PollB and 1997 Census estimates of race for Indiana

     

    1998 Indiana PollB

    1997 Census Indiana Estimates

    White or Caucasian 82.7 90.6
    Black or African American 4.2 8.2
    American Indian or Alaskan Native 0.6 0.2
    Asian or Pacific Islander 0.4 0.9
    Something else 0.4
    American Indian/Alaskan Native & Black/African American 0.4
    American Indian/Alaskan Native & White/Caucasian 6.2
    Asian/Pacific Islander & White/Caucasian 0.4
    Black/African American & Something else 0.2
    White/Caucasian & Something else 1.8
    Three or more races 2.0
    Don’t know 0.2
    Refused 0.6

     

    1998 Indiana PollB

    Versions 1 and 2

  • After 368 interviews (of 505), the order of the race categories was changed because there was concern that more people than usual were reporting as American Indian.
  • It was possibly because "American Indian" was the first category read and it started with "American."
  • However, when the order was changed to read "White or Caucasian" first, there was no change in the frequency of respondents who reported as American Indian.
  • Indiana PollB Versions 1 and 2 and the 1997 Census estimates of race for Indiana

     

    1998 Indiana PollB1

    1998 Indiana PollB2

    1997 Census Indiana Estimates

    White or Caucasian

    83.7

    79.4

    90.6

    Black or African American

    3.5

    5.9

    8.2

    American Indian or Alaskan Native

    0.5

    0.7

    0.2

    Asian or Pacific Islander

    0.5

    0.0

    0.9

    Something else

    0.3

    0.7

     
    American Indian/Alaskan Native & black/African American

    0.5

    0.0

     
    American Indian/Alaskan Native & white/Caucasian

    6.2

    5.9

     
    Asian/Pacific Islander & white/Caucasian

    0.5

    0.0

     
    Black/African American & Something else

    0.0

    0.7

     
    White/Caucasian & Something else

    1.4

    2.9

     
    Three or more races

    1.6

    2.9

     
    Don't know

    0.3

    0.0

     
    Refusal

    0.5

    0.7

     

     

     

    Race distribution for the 1998 Indiana PollA, the 1998 Indiana PollB, and 1997 Census estimates of race for Indiana

     

    1998 Indiana PollA

    1998 Indiana PollB  1998

    1997 Census Indiana Estimates

    White or Caucasian 88.1 82.7 90.6
    Black or African American 5.1 4.2 8.2
    American Indian or Alaskan Native 1.2 0.6 0.2
    Asian or Pacific Islander 1.0 0.4 0.9
    Something else 0.6 0.4  
    American Indian/Alaskan Native & black/African American 0.0 0.4  
    American Indian/Alaskan Native & white/Caucasian 0.6 6.2  
    Asian/Pacific Islander & white/Caucasian 0.0 0.4  
    Black/African American & Something else 0.0 0.2  
    White/Caucasian & Something else 0.4 1.8  
    Three or more races 0.6 2.0  
    Don’t know 0.6 0.2  
    Refused 0.8 0.6  

     

     

    1999 Indiana Short Poll

    Now, I'm going to read a list of racial categories. Please select one or more categories that describe your racial background.

    As I read each category, please answer with a yes or no.

  • Are you Alaska Native?
  • American Indian?
  • Asian?
  • Black or African American?
  • Native Hawaiian?
  • Pacific Islander?
  • White or Caucasian?
  • Are you another race I have not mentioned?
  • The respondents were instructed to respond with yes or no for each category as it was read.
  • Respondents were explicitly invited to report more than one category.
  • Some categories were split to reflect the Census pretest race categories.
  • The data are not comparable to other polls because there was oversampling of exchanges with higher incidence of African-Americans.
  •  

     

     

    Distribution of race from the 1999 Indiana Short Poll

     

    1999 Indiana Short Poll

    White or Caucasian 71.5
    Black or African American 14.1
    American Indian 0.2
    Alaska Native 0.0
    Asian 0.4
    Native Hawaiian 0.0
    Pacific Islander 0.0
    Another race 2.6
    American Indian & Black/African American 1.1
    American Indian & white/Caucasian 4.2
    Asian & white/Caucasian 0.2
    Black/African American & white/Caucasian 0.4
    Black/African American & another race 0.5
    White/Caucasian and another race 2.4
    Three or more races 2.6

    Conclusion

  • In general, the proportions of minorities in our telephone survey samples are relatively close to census counts and estimates when we either allow only one response or we do not read the race categories.
  • Whenever "American Indian" is read as a category, we get much higher proportions than we would expect given census estimates. Various attempts to minimize this problem have not been successful. We suspect the respondents are categorizing themselves as "American."
  • We anticipate that mode effects will cause differences in distributions of race collected in telephone surveys after the Census 2000.
  • Information Sources

    Race population estimates for 1997 (Indiana) were taken from: Population Estimates Program, Population Division, US Bureau of the Census, Washington, DC 20233, Internet Release date: September 4, 1998. www.census.gov/population/estimates

    Census 2000 race question development information was from: "Results of the 1996 Race and Ethnic Targeted Test," U.S. Census Bureau, the Official Statistics. ™ www.census.gov/population/www/documentation/twps0018/section-1.html

    Information on revisions to OMB Directive 15 was from: "Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity."

    www.whitehouse.gov/WH/EOP/OMB/html/fedreg/Ombdir15.html

    Back to Top