Characterizing score reliability for the South Oaks Gambling Screen

Accurate measurement of problem gambling and subsequent interpretation of results require that assessment devices yield strong psychometric properties. In terms of consistent measurement, reliability estimates must be calculated with each administration, as changes in sample characteristics may alter the scale's ability to generate reliable scores. The purpose in the present study was to explore reliability reporting practices and variability in reliability estimates on a frequently used problem gambling detection measure, the South Oaks Gambling Screen. All empirical research employing the South Oaks Gambling Screen was retained for analysis. Of the 121 studies used in the present study, only 13 (10.7%) reported reliability for the data at hand. The remaining studies either induced reliability (42%) or did not cite reliability at all (46%). The median internal consistency estimate across 15 samples was.86, ranging from.69 to.97. Predictor variables did not demonstrate significant relationships with score reliability. Results from this study indicate wide variability in reliability coefficients across studies. Therefore, it is suggested that further psychometric analysis be conducted on the SOGS in order to yield more consistent and accurate measurement of problem gambling.