The evaluation of risk factors in dental research frequently uses observati
ons at multiple sites in the same patient. For this reason, statistical met
hods that accommodate correlated data are generally used to assess the sign
ificance of the risk factors (e.g., generalized estimating equations, gener
alized linear mixed models). In applications of these methods, it is typica
lly assumed (implicitly, if not explicitly) that between-subject and within
-subject comparisons will produce the same estimated effect of the risk fac
tor. When between- and within-subject comparisons conflict, the statistical
methods can give biased estimates or results that are difficult to interpr
et. For illustration, we present two examples from periodontal disease stud
ies in which different statistical methods give different estimates and sig
nificance levels for a risk factor. Statistical analyses in dental research
should assess whether different sources of information give similar conclu
sions about risk factors or treatments.