What is the reason for conducting a human abuse potential (HAP) study? Are we interested is finding signals that indicate abuse potential, or are we interested in “proving” there is no signal? Regulated HAP studies are often geared towards excluding false negatives (i.e., concluding no abuse liability, when it exists), usually at the expense of false positives results (concluding abuse liability when there isn’t any). The pharmacokinetic/pharmacodynamic relationships associated with drugs of abuse can present additional challenges in interpretation, as can the use of subjective measures, in the form of high variability and “distributional” violations. Are we using appropriate statistical approaches for interpreting these data? Do our interpretations map to the real world? Are we abusing statistics to get the answers we want as opposed to the answers we need? This talk will discuss the pro and cons associated with a select set of recommended study designs and analyses.