How (Not) to Use Risk Ratios in Sociological Research
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
How (Not) to Use Risk Ratios in Sociological Research. / Karlson, Kristian Bernt; Quillian, Lincoln.
In: Socius: Sociological Research for a Dynamic World, Vol. 9, 2023, p. 1-9.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - How (Not) to Use Risk Ratios in Sociological Research
AU - Karlson, Kristian Bernt
AU - Quillian, Lincoln
PY - 2023
Y1 - 2023
N2 - A small but growing literature uses the risk ratio as an association or effect measure. Unlike odds ratios, risk ratios are unaffected by rescaling or non-collapsibility bias and are straightforward to interpret. However, the risk ratio has one unattractive property that researchers need to be aware of: It is not symmetric with respect to the outcome definition. The ratio between two groups’ probability of success does not equal the inverse of the ratio between the two groups’ probability of failure. Choosing the category of a binary outcome to use as the “success” category can significantly affect substantive conclusions, particularly in research comparing risk ratios with highly different baserates of the “success” outcome. We give examples from discrimination and social mobility studies that illustrate this point, and we present rules of thumb for the use of the risk ratio depending on the baserate of the outcome.
AB - A small but growing literature uses the risk ratio as an association or effect measure. Unlike odds ratios, risk ratios are unaffected by rescaling or non-collapsibility bias and are straightforward to interpret. However, the risk ratio has one unattractive property that researchers need to be aware of: It is not symmetric with respect to the outcome definition. The ratio between two groups’ probability of success does not equal the inverse of the ratio between the two groups’ probability of failure. Choosing the category of a binary outcome to use as the “success” category can significantly affect substantive conclusions, particularly in research comparing risk ratios with highly different baserates of the “success” outcome. We give examples from discrimination and social mobility studies that illustrate this point, and we present rules of thumb for the use of the risk ratio depending on the baserate of the outcome.
KW - Faculty of Social Sciences
KW - risk ratio
KW - association measure
KW - effect measure
KW - odd ratio
KW - risk difference
KW - quantitative methods
U2 - 10.1177/23780231231192394
DO - 10.1177/23780231231192394
M3 - Journal article
VL - 9
SP - 1
EP - 9
JO - Socius
JF - Socius
SN - 2378-0231
ER -
ID: 356951370