Inter-observer agreement on subjects' race and race-informative characteristics

Heather J.H. Edgar, Shamsi Daneshvari, Edward Harris, Philip J. Kroth

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Health and socioeconomic disparities tend to be experienced along racial and ethnic lines, but investigators are not sure how individuals are assigned to groups, or how consistent this process is. To address these issues, 1,919 orthodontic patient records were examined by at least two observers who estimated each individual's race and the characteristics that influenced each estimate. Agreement regarding race is high for African and European Americans, but not as high for Asian, Hispanic, and Native Americans. The indicator observers most often agreed upon as important in estimating group membership is name, especially for Asian and Hispanic Americans. The observers, who were almost all European American, most often agreed that skin color is an important indicator of race only when they also agreed the subject was European American. This suggests that in a diverse community, light skin color is associated with a particular group, while a range of darker shades can be associated with members of any other group. This research supports comparable studies showing that race estimations in medical records are likely reliable for African and European Americans, but are less so for other groups. Further, these results show that skin color is not consistently the primary indicator of an individual's race, but that other characteristics such as facial features add significant information.

Original languageEnglish (US)
Article numbere23986
JournalPLoS ONE
Volume6
Issue number8
DOIs
StatePublished - Aug 29 2011

Fingerprint

Skin
Skin Pigmentation
Color
Asian Americans
African Americans
Hispanic Americans
color
Health
research support
North American Indians
American Indians
Orthodontics
Names
Medical Records
socioeconomics
shade
Research Personnel
Light
European Americans
Research

All Science Journal Classification (ASJC) codes

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Inter-observer agreement on subjects' race and race-informative characteristics. / Edgar, Heather J.H.; Daneshvari, Shamsi; Harris, Edward; Kroth, Philip J.

In: PLoS ONE, Vol. 6, No. 8, e23986, 29.08.2011.

Research output: Contribution to journalArticle

Edgar, Heather J.H. ; Daneshvari, Shamsi ; Harris, Edward ; Kroth, Philip J. / Inter-observer agreement on subjects' race and race-informative characteristics. In: PLoS ONE. 2011 ; Vol. 6, No. 8.
@article{c0847d71e9c244719d125d24d0ff8538,
title = "Inter-observer agreement on subjects' race and race-informative characteristics",
abstract = "Health and socioeconomic disparities tend to be experienced along racial and ethnic lines, but investigators are not sure how individuals are assigned to groups, or how consistent this process is. To address these issues, 1,919 orthodontic patient records were examined by at least two observers who estimated each individual's race and the characteristics that influenced each estimate. Agreement regarding race is high for African and European Americans, but not as high for Asian, Hispanic, and Native Americans. The indicator observers most often agreed upon as important in estimating group membership is name, especially for Asian and Hispanic Americans. The observers, who were almost all European American, most often agreed that skin color is an important indicator of race only when they also agreed the subject was European American. This suggests that in a diverse community, light skin color is associated with a particular group, while a range of darker shades can be associated with members of any other group. This research supports comparable studies showing that race estimations in medical records are likely reliable for African and European Americans, but are less so for other groups. Further, these results show that skin color is not consistently the primary indicator of an individual's race, but that other characteristics such as facial features add significant information.",
author = "Edgar, {Heather J.H.} and Shamsi Daneshvari and Edward Harris and Kroth, {Philip J.}",
year = "2011",
month = "8",
day = "29",
doi = "10.1371/journal.pone.0023986",
language = "English (US)",
volume = "6",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "8",

}

TY - JOUR

T1 - Inter-observer agreement on subjects' race and race-informative characteristics

AU - Edgar, Heather J.H.

AU - Daneshvari, Shamsi

AU - Harris, Edward

AU - Kroth, Philip J.

PY - 2011/8/29

Y1 - 2011/8/29

N2 - Health and socioeconomic disparities tend to be experienced along racial and ethnic lines, but investigators are not sure how individuals are assigned to groups, or how consistent this process is. To address these issues, 1,919 orthodontic patient records were examined by at least two observers who estimated each individual's race and the characteristics that influenced each estimate. Agreement regarding race is high for African and European Americans, but not as high for Asian, Hispanic, and Native Americans. The indicator observers most often agreed upon as important in estimating group membership is name, especially for Asian and Hispanic Americans. The observers, who were almost all European American, most often agreed that skin color is an important indicator of race only when they also agreed the subject was European American. This suggests that in a diverse community, light skin color is associated with a particular group, while a range of darker shades can be associated with members of any other group. This research supports comparable studies showing that race estimations in medical records are likely reliable for African and European Americans, but are less so for other groups. Further, these results show that skin color is not consistently the primary indicator of an individual's race, but that other characteristics such as facial features add significant information.

AB - Health and socioeconomic disparities tend to be experienced along racial and ethnic lines, but investigators are not sure how individuals are assigned to groups, or how consistent this process is. To address these issues, 1,919 orthodontic patient records were examined by at least two observers who estimated each individual's race and the characteristics that influenced each estimate. Agreement regarding race is high for African and European Americans, but not as high for Asian, Hispanic, and Native Americans. The indicator observers most often agreed upon as important in estimating group membership is name, especially for Asian and Hispanic Americans. The observers, who were almost all European American, most often agreed that skin color is an important indicator of race only when they also agreed the subject was European American. This suggests that in a diverse community, light skin color is associated with a particular group, while a range of darker shades can be associated with members of any other group. This research supports comparable studies showing that race estimations in medical records are likely reliable for African and European Americans, but are less so for other groups. Further, these results show that skin color is not consistently the primary indicator of an individual's race, but that other characteristics such as facial features add significant information.

UR - http://www.scopus.com/inward/record.url?scp=80052279742&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052279742&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0023986

DO - 10.1371/journal.pone.0023986

M3 - Article

VL - 6

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 8

M1 - e23986

ER -