Research with Convicted Sex Offenders: Should We Trust the Results?

Background/Aim: Grooming, in the context of child sexual abuse (CSA), refers to a set of behaviors that the perpetrator of sexual crimes uses to initiate or maintain sexual abuse of a child, generally through a series of calculated steps in which an offender achieves their goal of access, compliance, and secrecy (Bennett & O’Donohue, 2014). The sexual offender grooms the child victim, their environment, and institutions both come in contact with (Craven, Brown, & Gilchrist, 2006; McAlinden, 2006; McAlinden, 2012). Whether the grooming behaviors happen in person or online, both serve the same purpose (McAlinden, 2006; Williams et al., 2013). There is no measure that examines these actions, and the current study is a psychometric examination of a new grooming measure that examined offenders’ grooming behaviors before and during sex crimes.

Methods: Once IRB approval was obtained, the researchers worked with a local agency that provides therapeutic group treatment to court mandated clients who had been convicted of sex offenses against adults and children. Once consent was obtained, the participants (sex offenders) anonymously filled out a demographic survey, and the Sex Offenders Grooming Assessment (SOGA), which included eight subscales: Grooming the Environment, Online Grooming, Tricks, Mind-Altering Approaches, Coercion, Bribes, Threats, and Physical Force. Each subscale contained Likert- scaled questions, as well as open-ended qualitative questions.

Sample: The final sample consisted of 50 male convicted sex offenders (n=50), aged 20 through 72 years old, with a mean age of 44. The majority of the sample was white, with one person identifying as Black, and one person identifying as multi-racial.

Results: While the overall Chronbach alpha for the measure was favorable (@=.907), responses to qualitative questions suggested that some of the responses in each subscale were incorrect. For instance, for the subscale that measured ‘Online grooming’: One participant endorsed ‘no’ for online grooming but then wrote in an answer “I groomed my environment by isolating myself and freeing time to get on the internet”. Another participant endorsed ‘no’ for all the questions in online grooming, but when asked how he had access to the victim, answered “Facebook texting” (which is considered an online activity). Obviously, these sex offenders had to have groomed their victims online, but when asked a general question “Did you groom your victims online?”, none of them responded with ‘yes’.

Implications: The measurement of grooming is notoriously difficult, and there is no psychometrically validated measure of grooming in the literature (Bennett & O’Donohue, 2014). In this study, we learned that when undertaking research regarding grooming behaviors, we must always consider the possibilities of social desirability, cognitive dissonance, lying, etc., such that even though the measure appears to be psychometrically valid, this appearance is actually specious when considered in conjunction with qualitative responses of the participants. This measure will now be psychometrically examined with clinicians who work with sex offenders to understand if it can be used as part of an assessment strategy with this population.