Associations were examined through multivariate regression models, controlling for community sociodemographics, clustered on county, with robust standard errors.
Through a survey using stated choice scenarios discrete choice models controlling for context and experience effects are specified.
As the Times put it, "in various models controlling for different factors and using different definitions of tense situations, Mr. Fryer found that blacks were either less likely to be shot or there was no difference between blacks and whites".
But in various models controlling for different factors and using different definitions of tense situations, Mr. Fryer found that blacks were either less likely to be shot or there was no difference between blacks and whites.
This relationship lost significance in models controlling for reported hypertension.
Findings were unchanged in multivariable models controlling for demographic factors.
Multivariate logistic and normal regression models, controlling for baseline characteristics, examined relationships between race and outcomes.
Hierarchical logistic regression models controlling for patient and hospital-level factors were used.
We also estimated several models controlling for potential selection bias in the migrant coefficient.
In multivariate binomial regression models controlling for covariates, the association was attenuated but still present.
These results remained significant in multivariate logistic regression models controlling for relevant variables including PTSD, MDD, and TBI.
