|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
LETTER |
Sandro Galea is with the School of Public Health, University of Michigan, Ann Arbor. Jennifer Ahern is with the University of California, Berkeley.
Correspondence: Request for reprints should be sent to Sandro Galea, MD, DrPH, Department of Epidemiology, School of Public Health, University of Michigan, 1214 S. University, Ann Arbor, MI 48104-2548 (e-mail: sgalea@umich.edu).
|
|||
Moore et al. raise a concern about collinearity in our study, a concern that we shared when conducting this analysis. There are 2 distinct issues that merit discussion in relation to collinearity. First, when independent variables are strongly correlated, standard errors may be inflated and one of the independent variables may wash out the effect of the other. To examine whether these were problems in our analysis we calculated the bootstrap confidence intervals for the final models and found only minimal inflation of the confidence intervals and no change in inference for any of the models. Both mean education and
This article has been cited by other articles:
![]() |
B. Kamoie, R. M. Pestronk, P. Baldridge, D. Fidler, L. Devlin, G. A. Mensah, and M. Doney Assessing laws and legal authorities for public health emergency legal preparedness. J. Law Med. Ethics, March 1, 2008; 36(1 Suppl): 23 - 27. [PDF] |
||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |