Measuring addictive behaviors accurately is very difficult. Among the many complications are recruiting subjects, relying on their memory and self-report and quantifying the difference between pathological actions and actions that are merely unusual or uncommon. (Jerome Wakefield, Ph.D., discussed this topic in detail at the NCRG conference 2010.) One particularly difficult aspect of studying addiction is the diagnostic screening instrument (when scored, the instrument shows whether or not a person should be diagnosed with a particular disorder). Diagnostic screens of all kinds pose the problem of advancement versus consistency: It is always possible to make a newer and better instrument, but a new instrument means that studies conducted with the older one cannot be directly compared to studies conducted with the new instrument. This tension dictates that new instruments must be shown as better along several lines to make them worth implementing. Two researchers at the forefront of this discussion for gambling disorders are Erica Fortune and Adam Goodie, Ph.D., at the University of Georgia. Fortune and Goodie recently published a study, partly funded by a grant from the NCRG to Dr. Goodie, that compares the performance of two diagnostic screens in a population of college students (Fortune & Goodie, 2010).