Review 5 of "Forced Interactions with Sheriff Deputies Over Time and their Influence on Stigma and Self Identities among Individuals Convicted of Sex Crimes"

Vote: Reject

Overall Explanation: When I received this manuscript for review, I was excited about the potential for contribution to knowledge. The interactions of individuals convicted of sexual offenses (ICSOs) and the criminal justice system is a thoroughly understudied topic, and one where there is a significant opportunity for contribution to theory concerning recidivism, desistance, and alienation. I believe that this research has the potential for this contribution, but I am not convinced that the manuscript in its current form delivers on that potential. The title and abstract for the paper got me excited about this but the analysis does not quite deliver as strongly as it could. The interactions with law enforcement only enter the analysis at the tail end, and represent a relatively small component of the results. This read strangely to me, because as I read through the methods it seemed like the authors had an incredibly rich prospective data collection over 7 years. But ultimately, change over time really does not materialize in the analysis at all. It is presented as cross sections at various time points (pre-/post-conviction), but there is no systematic attempt to trace change over time for any given individual. At the same time, the authors are never quite clear as to whether they have more than one interview with each subject. If there is indeed some manner of prospective data collection as hinted on page 8, there is much more the authors can do here, such as mapping out changes in narrative at key points of follow-up. 

In the end, there is some fine analysis concerning “sex offender” identity management and stigma, and I am sure that there is a worthy contribution here, but the manuscript in its current form requires considerable refinement to deliver on its intended contribution on forced interactions over time. 

[Please put additional info below, as/if you see fit:]

Inconsistencies in Person-first Language: The juxtaposition of the manuscript opening with a discussion of stigmatizing labels, but using the term “sex offender” (even in quotes) throughout is off-putting. I also notice that the authors eventually stop using quotes around “sex offender” later in the manuscript, and occasionally mix in other terms like “registered citizen.” I would request some uniformity in person first language throughout the manuscript.

Organization of Front End: As I read through, the connections from section to section or even paragraph to paragraph are unclear. The authors could make the connections from section to section more explicit as it relates to their research objectives.

Narrative Research on Stigma: The literature discussed in the stigma internalization and management section is clearly seminal, but there has been considerable progress in this area in recent years. I believe that the contribution of the current work can be accentuated by ensuring that the backdrop is based off the contemporary narrative desistance literature. For instance, Rebecca Stone’s work on desistance and identity repair seems especially relevant, as they draw on some of this more recent work, and since publication has been cited by the most recent work in narrative desistance research.

Specificity on Data Collection: On Page 8 the authors discuss how the sample was reduced to the (impressive) 63 subjects. But language in this paragraph seems to suggest that data were collected prospectively over 2012-2019. This seems to get glossed over (see comment below on time lags), because there is little information given on the specifics of the data collection. If the primary data collection came from interviews, how many contacts were there per subject (average, min, max)? This seems like an incredible source of strength and richness for the design, but almost seems hidden away by the text. What I think the manuscript would benefit from is a more explicit and systematic account of the longitudinal nature of the design, and how that was used to generate findings concerning change.

Time Lags?: The mention of the 3 and 5 year time lags in the opening section ended up being somewhat confusing for me. As I read through the methods I was expecting to see some concrete discussion of how the interviewers attempted to isolate these time periods during retrospection and techniques for minimizing recall bias (e.g., calendars, mentioning significant events [Before Trump presidency, etc.]). Ultimately the authors note that they attempted to use triangulation with loved ones as a check, and then (realistically) highlight that if the reader is concerned about recall bias here then it’s time to throw out life history interviewing as a method anyways. As much as I appreciate the “Will you just get off our backs? Sheesh!” attitude from the authors – as this is how I usually want to write my papers anyways – I find this line of argument unsatisfying. By identifying *specific* time points for the sake of research design, this enhances the possibilities for recall bias unless the authors took design-based precautions. Then again, when I get to the analysis, I really just see a pre-conviction period and a post-conviction. The idea of change over time during post-conviction has seemingly been forgotten. I believe the authors can do better than this. Be more straightforward with the design intentions, and ensure that the design maps to the analysis presented.

Opportunities to Link Results to Literature: As the results are presented, there are opportunities for solidifying the contribution of this research that I believe the authors could take better advantage of. For instance, in the pre-conviction identity section, the findings concerning the rejection of sex offender identity and the presentations of circumstances concerning the offenses are consistent with presentational strategies reported in qualitative research on “sex offender” treatment groups (e.g., Lacombe, 2008) and Heiith Copes’ research on “violent offenders” (Hochstetler, Copes, & Williams, 2010) and “crack heads” (Copes, Hochstetler, & Williams, 2008). These sorts of linkages would strengthen the foundation for the results by identifying the literatures being furthered by these findings.