A treatment program dealing with violent offenders adapted a shame transformation process; treatment is based on experiencing victims’ feelings through differentiating then activating shame and guilt. This study provided mixed results concerning the known-group validation. Cronbach’s alpha, the measure of internal consistency, supports the initial clinically observed assessment that three groups of offenders exist, while goodness-to-fit indices with the exception of the chi-square probability value are less supportive. Since these values were below the conventional 0.9 threshold, the original hypothesized initial model was rejected. However, a later, condensed model that omitted low performing variables, performed better, although still not optimally. The construct validation of the SGBI reported here attempts to systematically integrate theory, research, and experience derived from an incipient shame management approach to treating offenders in a clinical setting. Such an approach can be utilized with all offender populations when seeking to elicit a guilt-response, notably when calling for accountability, taking responsibility for negative behavior, and also recognizing its impact on victims. By recognizing the impact of negative behaviors on themselves, an offender ultimately gains empathy for others. This approach draws on a global move toward more emotionally intelligent justice and the successes of restorative justice in reducing recidivism, particularly among violent offenders. Both call for addressing and reintegrating the shame inherent in traditional, rational, or retributive criminal justice. Tables and references