Differential Development of Perceived Costs and Rewards of Offending the Relevance for Predicting Offending in Early Adulthood: A Group-Based Trajectory Modeling Approach

1. This cap was set at 15% of the total sample. The Pathways to Desistance research team set a predetermined number on the total number of participants to enroll. When the capped limit was nearly reached on this total, the research team stopped approaching male drug offenders regarding interest in the study. More details on this procedure are explained by Schubert et al. (Citation2004).

2. Individual cost items included: Hanging out with your friends; Choosing how to spend your time; Deciding what to eat; Being able to leave someplace when you want; Sleeping late; Going out with your boyfriend/girlfriend; Watching TV when you want; Dressing the way you want; Listening to the kind of music you like; Seeing whatever movies you want; Being in a warm, comfortable place; Being in a place that looks good; Choosing what to wear; Talking on the phone whenever you want; Reading whatever you want to; Being by myself if you want; Drinking or getting high when you want; Buying things. Preliminary analyses indicated a large amount of missingness on the pooled indicator comprising every wave of data for this construct (47.39% missing). It is unclear as to exactly why this was, though Stata/MP 16 indicated that 42.04% of the sample did not “skip into this section.” Correlation analyses were conducted to determine if key variables were associated with this particular kind of missingness. Those who did not “skip into this section” were more likely to be female, White, have weaker social support, and older. They were less likely to report exposure to violence, had fewer deviant peers, reported less negative affect, and were less likely to be Black. All of these data were imputed using full information maximum likelihood in trajectory analysis, thus, they were assigned to a trajectory group even despite having missing data in the pooled sample. This made it possible to include them in regression modeling in the distal outcome analysis phase of analyses.

3. Individual reward items included: Fighting; Robbery with a gun; Stabbing someone; Breaking into a store or home; Stealing clothes from the store; Vandalism; Auto theft.

4. Offenses included in the variety score included: Destroyed/damaged property; Set fire to house/building/car/vacant lot; Entered building to steal; Shoplifted; Bought/received/sold stolen property; Used checks/credit cards illegally; Stolen car/motorcycle; Sold marijuana; Sold other illegal drugs; Carjacked someone; Drove drunk or high; Been paid by someone for sex; Forced someone to have sex; Killed someone; Shot someone (where bullet hit); Shot at someone (pulled trigger); Too something by force using weapon; Took something by force no weapon; Beaten up somebody badly needed doctor; Been in fight; Beaten up someone as part of gang; Carried a gun; Broke into car to steal something; Gone joy-riding.

5. The following ordinal ranking system existed for educational attainment: 1 = Some grad or prof school/prof or grad school; 2 = College graduate; 3 = Business or trade school/Some college/grad of 2-yr college; 4 = High school diploma; 5 = Some high school; 6 = Grade school or less. Occupational prestige ordinal categories were: 1 = Farm Laborers/Menial Service Workers; 2 = Unskilled Workers; 3 = Machine Operators and Semiskilled Workers; 4 = Smaller Business Owners, Skilled Manual Workers, Craftsmen, and Tenant Farmers; 5 = Clerical and Sales Workers, Small Farm and Business Owners; 6 = Technicians, Semiprofessionals, Small Business Owners; 7 = Smaller Business Owners, Farm Owners, Managers, Minor Professionals; 8 = Administrators, Lesser Professionals, Proprietors of Medium-Sized Businesses; 9 = Higher Executives, Proprietors of Large Businesses, and Major Professionals.

6. The following types of direct victimization experiences were included in this measure: have you been attacked with a weapon, like a knife, box cutter, or bat?; have you been shot at?; have you been shot?; have you been raped, had someone attempt to rape you or been sexually attacked in some other way?; have you been beaten up, mugged, or seriously threatened by another person?; have you been chased where you thought you might be seriously hurt? These same six items were assessed in terms of whether a participant witnessed these events occur, as well as asking participants if they saw someone die in the previous observation period.

7. The specific antisocial acts that participants were asked about were: How many of your friends have suggested that you should go out drinking with them?; How many of your friends have suggested or claimed that you have to get drunk to have a good time?; How many of your friends have suggested or claimed that you have to be high on drugs to have a good time?; How many of your friends have suggested that you should sell drugs?; How many of your friends have suggested that you should steal something?; How many of your friends have suggested that you should hit or beat someone up?; How many of your friends have suggested that you should carry a weapon?

8. The following domains of social support were assessed: Are there any adults who you admire and would want to be like?; If you needed some information or advice about something, is there someone you could talk to?; If you were having trouble at home, is there someone you could talk to?; If you got an award or did something well, is there someone you would tell?; Is there an adult with whom you can talk about important decisions in your life?; Is there an adult you can depend on for help if you really need it?; Is there an adult you feel comfortable talking about problems with?; Is there a special adult person in your life who cares about your feelings?

9. Age 23 measures of each time-variant control variable and offending seriousness were used in phase two of analyses, as this was the terminal age for the trajectory groups.

10. These groups were chosen because they should both theoretically reflect the groups that were at lowest risk for offending, thus, providing a consistent understanding of how gradients of offending risk should theoretically be reflected across each of the other groups in each model.

11. Sensitivity analyses were conducted to test for robustness of results. A binary indicator of offending at age 23 was used as the outcome variable in these supplemental analyses (0 = No offending in prior observation period; 1 = Offended during prior observation period). Results were robust with this alternative outcome, with all significant and nonsignificant effects related to trajectory group membership remaining significant or nonsignificant.

12. The PHR group was chosen as the reference group because of the similar intercepts of these groups at age 16 and the great divergence in development thereafter, with the PHR group demonstrating persistently high perceived rewards of offending.