Commentary: Flawed study designs are not salvaged by large samples

W. C Willett
2008 International Journal of Epidemiology  
doi:10.1093/ije/dyn177 pmid:18832386 fatcat:kflh6knunrh6rd7sptfgqkkj2q