Erratum to: ann. behav. med.

DOI 10.1007/s12160-012-9366-5

In “Predicting human papillomavirus vaccine uptake in young adult women: Comparing the Health Belief Model and Theory of Planned Behavior” (DOI 10.1007/s12160-012-9366-5) we inadvertently failed to label and interpret the R-squared statistics reported in the article as pseudo R-squareds. Because our primary outcome variable (human papillomavirus [HPV] vaccine uptake) was dichotomous, it is not possible to compute “true” R-squared statistics, as would be generated with ordinary least squares (OLS) regression. Pseudo R-squareds are conceptually analogous to OLS R-squareds in that they can provide information about comparative model fit. Pseudo R-squareds are particularly useful when comparing models that predict the same binary outcome variable from the same dataset, as was the case in the present study. In these instances, it is possible to determine which model better predicts the outcome variable by comparing pseudo R-squareds; models with higher pseudo R-squared values suggest better model fit [1]. Nevertheless, although pseudo R-squared values can be used to assess comparative model fit, they should not be interpreted as indicators of variance accounted for in the outcome variable.