Corrections to Quinsey et al., (2006) Violent Offenders: Appraising and Managing Risk, Second Edition




Note: Regarding the scoring of the SORAG and the item pertaining to the Cormier-Lang score for criminal charges for violent offenses prior to the index offense, the first edition of the book assigned a SORAG score of -2 for a Cormier-Lang score of zero (0). In the second edition, this was altered to -1 because either was consistent with the data for the development sample and the latter was more conservative.


Unfortunately, the first printing of our book has a couple of errors.

Pg. 165, Chapter 8 in Exhibit 8.2, PRORATING SCORES:

The first example should read (corrections in boldface):

     For example, suppose an offender is scored on the VRAG. Items 10 (Meets DSM-III criteria for any personality disorder), and 11 (Meets DSM-III criteria for schizophrenia) are scored as -2 and -3 as substitutions obtained by using DSM-IV diagnoses. Items 9 and 12 are unavailable. The offender obtains a score of -2 on the first 8 items of the VRAG scored according to the scoring criteria outlined in Appendix A. The total possible negative score for these items is -15. Thus, the offender obtained 2/15 or .133 of the available negative points for the 8 items that could be scored according to the scoring criteria outlined in Appendix A. The available negative points for the missing items (9 and 12) is 6. Multiplying .133 times 6 gives .798. Thus, the offender's total score including prorating is -2 (score on the 8 items scored by the scoring criteria in Appendix A), plus -.798 (the addition for prorating), plus -5 (for the substituted items) = -7.798. This number can then be rounded to the nearest whole number, in this case, -8.

Pg. 300: Under ETHICS AND BIAS point number 4:

The answer to the question should say, “No. We excluded…recidivism. Note, however, that there are two items (7 and 8) on the SORAG that do include noncontact offenses that are known to be sexual. Note also….”


  Back to Risk Assessment Page