Educational Support Reduces Arrest Recidivism for Adolescents

Educational Support Reduces Arrest Recidivism for Adolescents | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Educational Support Reduces Arrest Recidivism for Adolescents Edward Cohen, Roderick Rose This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6994635/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Educational support for both youth and caregivers is one of the priorities that are required to prevent involvement of youth in the justice system. Since August 2018, JusticeEd has matched students with Education Liaisons that provide probation-involved students and their families with individualized educational case management, coaching, and mentoring. We examined the likelihood of re-arrest and time to re-arrest for participants in the JusticeEd (JE) program, in comparison to a sample of youth referred to educational support but who did not receive JusticeEd services. Data were requested from a large California county’s Juvenile Probation Department for all youth arrested and referred to educational services during a 4.5-year timeframe. The data included arrest dates, offenses, adjudication decisions, whether JE services were received, and youth and family risk factors assessed for all youth actively supervised by the Probation Department. Propensity score matching was used to match the comparison group with those receiving JE services. A lower proportion of JusticeEd youth than those in the comparison were re-arrested. For those re-arrested, event history analysis of time to re-arrest showed that the JE youth had a shorter time period to re-arrest during the program period, but after the program period the time to re-arrest was no different than the matched comparison youth. The JE program’s ability to affect re-arrests is most likely due to the program’s intensive approach to supporting families, acting as liaison to the school, and removing barriers to the parent’s and student’s engagement with school. adolescents juvenile justice arrests educational support Hispanic Figures Figure 1 Figure 2 Highlights School absenteeism and dropout are predictors of youth contact with police and arrest. Educational support support for both youth and caregivers is important to prevent adverse school events. This study examined the likelihood of juvenile re-arrest and time to re-arrest for participants in an educational support program at a large California county providing individualized educational case management, coaching, and mentoring to youth and parents, compared to a propensity-scored matched comparison group receiving treatment-as-usual. Bivariate and multivariate analyses showed that youth receiving the educational support program had a lower rate of re-arrest than comparison group youth. The study showed the benefits of educational support for both caregivers and youth in order to re-engage with school and protect against the risks leading to subsequent re-involvement with the justice system. Background Educational support is one of the priorities that are required to prevent involvement of youth in the justice system (Lee & Taxman, 2020 ; Puzzanchera et al., 2022 ). Considering the importance of social support from the family on educational success for Hispanic students (Safavian et al., 2022 ), the majority ethnicity of this study’s participants, help for both the student and family are necessary. There are many barriers to engagement of Hispanic adolescents in schools, however the family is key to successful engagement by, among other supports, helping the adolescent navigate discrimination and cultural differences in day-to-day school life, and providing parental monitoring and supervision (Martinez-Fuentes et al., 2021 ). For Hispanic caregivers, an important issue is their own ability to communicate and become engaged with the school. Schools and support programs are more successful engaging students if they respect parents’ views of the importance of education despite logistical barriers to attend meetings or communicate in English (Carpenter et al., 2016 ). Improving parental engagement with the school will also positively affect students’ attitudes about persevering academically (Mena, 2011 ). Among the predictors of delinquency and recidivism are a history of family violence, learning problems, having a disability, and having behavioral problems (Barrett et al., 2014 ; Simon et al., 2025 ; Zhang et al., 2011 ). Parental mental illness and a history of justice involvement are implicated in their offspring’s contact with police (Athanassiou et al., 2023 ; Calley, 2012 ), as is parental alcohol and drug use (Adamsons & Russell, 2023 ). One serious problem for Hispanic youth is the high rate of school dropout. Although high school dropout rates have been decreasing for all youth over the past decade, Hispanic males currently have higher dropout rates than males or females of all other groups except Native Americans (National Center for Education Statistics, 2024 ). School dropout, suspensions, and absenteeism have all been linked to involvement with the justice system (Robertson & Walker, 2018 ), yet there are emerging practices to prevent these events. Schools that go beyond a rigid disciplinary response and programs that address psychosocial issues for the youth and family have better outcomes in reducing school absences, and thereby improving school engagement (Dembo & Gulledge, 2009 ). Partnerships between schools, parents, and community mentors can reduce chronic absenteeism that often leads to dropping out (Sheldon & Epstein, 2004 ). Educational achievement has a demonstrated effect on reducing the likelihood of re-arrest and the seriousness of subsequent offenses (Blomberg et al., 2011 ), however there are few studies of the impact of educational support on families and youth who come to the attention of the juvenile probation system, even though observers recommend as a best practice linkages between school and probation services to ensure student engagement and educational success (Gary, 2014 ). Abrams et al. ( 2011 ) found that the length of time in a community re-entry program for high-risk juvenile offenders, which included robust support for educational attainment and the strengthening of bonds with caregivers, was protective against repeat convictions. However, for the most part re-entry programs have had mixed results in terms of re-arrests. A study of a re-entry program for gang-involved youth treated in Texas residential facilities, for example, showed no impact on recidivism rates relative to a comparison group within a year after discharge (Spooner et al., 2017 ). Earlier interventions may have more opportunity to affect protective factors, such as a diversion program in Texas targeting youth arrested for the first time (John et al., 2025 ). This program evaluation noted pre- to posttest improvements in relationships with parents and in school adjustment; recidivism, however, was not measured. Diversion programs (which predominantly serve pre-charged youth) whether they are designed to provide warning to the youth without further action, or more invasively require program participation to avoid further court action, have shown an impact on reducing juvenile recidivism rates in a meta-analysis by Wilson & Hoge ( 2013 ). On the other hand, a meta-analysis of diversion programs by Schwalbe et al. ( 2012 ) did not find an overall impact of these programs on recidivism. There were many differences among the studies covered in these meta-analyses, such as having different target populations and varied time periods measuring re-arrest. There were also too few studies with robust group designs. Although attention to school success was a component of many diversion programs, very few specifically targeted the provision of support to both youth and caregivers regarding school engagement. Given the gaps in the literature for early intervention programs that specifically provide educational support, the research questions are 1) For youth arrested, does an educational support program reduce re-arrests? 2) For youth who were re-arrested, what is the impact of an educational support program on how soon youth are re-arrested? Methods Study Site and Sample This study addresses one program designed to re-engage justice-involved youth and their caregivers with school – the JusticeEd (JE) Program of the National Center for Youth Law. The sample were all youth arrested in a large California county during the time period from 1/1/2017-6/30/2022 and assessed as needing some type of education-related services. The data included 965 unduplicated youth, with 7,204 arrests, of which 5,243 were sustained petitions - those that resulted in filing a petition with the court to proceed on charges against the youth, and the charges were found to have merit. Of these youth, 268 were referred to JE, and 155 youth received JE services. The remaining youth in the sample were referred to either a program for youth with learning disabilities and in need of an Individual Education Plan, or to legal representation related to addressing school compliance with school discipline or special education policies. Seven JE youth did not have enough data to complete the propensity score match, and were excluded from analysis yielding a final sample of 148. Of these seven youth, only one was re-arrested during the time periods analyzed. Using the available non-missing data, these youth had no statistically significant differences of any covariates from the larger sample of JE youth. A sample of 148 each in a treatment and a comparison group has a power of .99 to detect an effect size difference of 0.20 between the groups. Descriptive statistics of variables used in the analysis are shown in Table 1 . Table 1. Descriptive Statistics Comparing Groups Variable JE Youth N (%) or mean (sd) N=148 Matched Comparison N=148 Overall non-JE Youth N= (%) or mean (sd) N=836 Average age at first assessment (SD) 16.0 (1.16) 16.53 (1.05) Compared to JE Youth p < .001 15.89 (1.33) Gender Male (1) Female (0) 120 (81%) 28(19%) 127 (86%) 21 (14%) 685 (81.9%) 151 (18.1%) Average number of Arrests pre-JE (SD) 0.81 (1.19) 0.74 (1.62) 0.91 (1.81) Race/Ethnicity Asian Black Caucasian Hispanic Other (reference) 4 (2.7%) 16 (10.8%) 10 (6.8%) 111 (75%) 7 (4.7%) 3 (2%) 10 (6.8%) 6 (4.1%) 127 (85.8%) 2 (1.4%) 56 (6.7%) 57 (6.8%) 106 (12.7%) 578 (69.1%) 39 (4.7%) Absenteeism Problem (1) Not a problem (0) 71 (48%) 77 (52%) 69 (47%) 79 (53%) 357 (42.7%) 479 (57.3%) Behavior Problem Not a problem 98 (66%) 50 (33.8%) 37 (25%) 111 (75%) Compared to JE Youth p = .022 260 (31.1%) 576 (68.9%) Grades Problem Not a problem 98 (66.2%) 50 (33.8%) 92 (62.2%) 56 (39.8%) 406 (48.6%) 430 (51.4%) Compared to JE Youth p < .001 Open to Special Education Open to SE Not open to SE 103 (69.6%) 45 (30.4% 92 (62.2%) 56 (37.8%) 534 (63.9%) 302 (36.1%) Mental health diagnosis Has diagnosis Does not have not a diagnosis 27 (18%) 121 (82%) 19 (12.8%) 129 (87%) 235 (28.1%) 601 (71.9%) Compared to JE Youth p < .007 Average JAIS risk level Low (reference) Medium High 61 (41.2%) 68 (45.9%) 19 (12.8%) 59 (36.9%) 77 (52%) 12 (8.1%) 269 (33.7%) 413 (51.8%) 116 (14.5%) Parent history justice involved Problem Not a problem 67 (45.3%) 81 (54.1%) 68 (45.9%) 80 (54.1%) 423 (50.6%) 413 (49.4%) Parent history drinking/drugs Problem Not a problem 60 (40.5%) 88 (59.5%) 52 (35.1%) 96 (64.9%) 348 (41.6%) 488 (58.4%) Differences between groups not statistically significant unless otherwise stated. [Table 1 . Here] The majority of youth in the full sample were male (82%) and identified as Hispanic (69%). The full list of potential comparison group youth included a) those with learning disabilities and in need of an Individual Education Plan, b) those referred to legal representation, c) those youth referred to JE but did not accept services, and d) those eligible for referral to educational support but determined not to need specialized services at the time of assessment. The IRB at _____University exempted the project from human subjects committee review since the data were de-identified. A university-county memorandum of understanding was developed covering data sharing, as well as a no-cost evaluation agreement between the university and the National Center for Youth Law. Research Design Randomization of youth to educational support services was not feasible at the time of post-arrest assessment since all youth who met the initial criteria based on screening by probation staff were referred to educational support services. Sufficient data were available to conduct a quasi-experimental matched comparison group post-test design. Propensity score matching (PSM) was used to construct a matched comparison group. The PSM procedure involved using logistic regression to generate probability scores for each youth on the nominal dependent variable referral to JE. The assumption is that although they did not meet initial criteria for referral, some youth not referred to JE (or those referred but did not receive services) might nevertheless have benefited from the JE program, and share many risk factors of those who participated in JE. We selected the comparison group cases using nearest neighbor matching. Measures Table 1 lists the variables used for analysis and their characteristics. Re-arrest All arrests in the study were those that resulted in filing a petition for a charge that was sustained by the juvenile court. Charges that were not filed or sustained (e.g. the charges were dropped or there was insufficient evidence) were excluded from analysis. Identifying the index rearrest date for JE youth was based either on the program’s enrollment start or end date – analyses included both scenarios. No enrollment start or end dates were available for the comparison youth. Establishing a comparable time period to identify the index re-arrest date for those youth proved challenging since the JE youth had rolling enrollment dates throughout the range of the data. To create an equitable time period for the comparison group youth, we applied the same treatment entry and exit dates of the JE youth to the corresponding comparison youth matched on their nearest propensity score. The assumption was that if a comparison group youth with similar risk factors were in the JE group, their length of treatment would have been similar. For the survival analysis, we calculated the number of days to the first arrest after the enrollment start dates and, alternatively, the number of days to the first arrest after the enrollment end dates. Survival analysis also requires a nominal variable indicating whether the re-arrest event occurred (coded 1) or whether it did not occur by the end of the observation period (coded 0), again coded separately for arrests after both JE enrollment start and end dates. Youth and Family Characteristics The data included age at first assessment and gender, a dichotomous nominal variable (0 = female, 1 = male, the only gender attributes reported). Race or ethnicity was provided for American Indian, Asian, Black, Caucasian, Hispanic, and “other” categories. For the logistic regression with the re-arrest outcome, dummy variables for each race and ethnicity attribute were coded 1 or 0. “Other” was the excluded reference variable. School problem areas identified in the Juvenile Assessment and Intervention System™ (JAIS) (0 = no problem; 1 = has problem) included absenteeism, behavior problem, poor grades, mental health problem, and whether the student was in special education. The JAIS system was meant to assess needs for planning as well as risk factors. It was originally developed and validated for adults (Baird, 2018 ) and more recently adapted for youth (Evident Change, 2025 ). The JAIS risk level was determined upon assessment, and resulted in an ordinal variable – low, medium, or high risk. If youth were assessed more than once, the most frequent risk score was used. For count ties, the risk level associated with the latest arrest was used. In the PSM matching procedure, one risk level variable with three attributes was used, however in the logistic regression for the outcome analysis dummy variables for “high” and “medium” were included with “low” excluded as the reference. Parental problems assessed by the JAIS (0 = no problem; 1 = has problem) included history of drinking or drug use and history of involvement in the justice system. Variables used for the PSM procedure included: a) Average age of child at first assessment, b) Male gender vs. female, c) Race/ethnicity, d) School absenteeism, e) Behavior problem, f) Poor grades, g) Has open case in special education, h) Youth has a mental health diagnosis, i) JAIS risk score, j) Parent history of justice involvement, and k) Parent history of drinking/drug use. Table 1 shows the descriptive statistics comparing the JE group with the matched comparison and overall non-JE group. The JE and matched comparison groups were statistically different for variables average age at assessment (matched comparison youth averaged half a year older than JE youth) and behavior problem (JE youth had a higher rate). “Off-the-shelf” demographic variables such as age, race/ethnicity, or gender alone do not typically account for imbalances between non-randomly assigned conditions, but the additional characteristics from the JAIS and referral assessment (represented by d-k) are associated with rearrest and referral and we propose that residual imbalances between groups should have a negligible effect on the findings. For a test of the balanced distribution of independent variables we calculated the standard mean deviation (SMD) for each variable between the JE group and matched comparison group (not shown in table). Bai & Clark ( 2019 ) recommend a difference of the SMD below .5, and that there should be at least a 75% overlap for balanced distribution of covariates. For this study, a cutoff of equal to or lower than .2 SMD was chosen as the threshold for balanced covariates. Of all the independent variables, age at first assessment, behavior problem and Hispanic were above the SMD threshold for the matched data. There were no differences between the groups regarding the other risk factors as well as the average number of arrests prior to the start of the JE implementation. The pre-PSM groups, by comparison, had two variables that exceeded the SMD threshold – they were having poor grades and youth mental health problem. Group differences for these variables disappeared in the post-PSM groups. Program Procedures Since August 2018, the JE Program has provided students and their families with individualized educational case management, coaching, and mentoring to support students in achieving their academic goals. Students and families are referred to educational support if they experienced a suspension or expulsion; there are attendance problems; the youth need help with employment, vocational or college readiness skills; or there is a need to help guide the youth and parent towards developing future educational goals. The program uses trained paraprofessionals as Educational Liaisons who a) review and discuss academic progress; b) provide coaching on study skills and social-emotional skill development; c) develop and review education plans and assess progress on student-set education goals; d) engage in post-secondary and career exploration, and d) assist the youth by monitoring completion of school work (National Center for Youth Law, 2022). JE staff work closely with caregivers in order to encourage them to be active, as education champions, in their youths’ education. Typical activities for caregivers with English as a second language, for example, are facilitating meetings with school personnel, providing Spanish translation, and explaining school policies. Analysis Procedures With the matched treatment and comparison groups, we conducted two separate bivariate chi square test of association analyses to determine differences between the groups of the number of youth re-arrested, using both the enrollment start dates and enrollment end dates to identify the index re-arrest. We also conducted a separate “doubly robust” logistic regression of re-arrest after the enrollment end dates on covariates, including propensity scores as recommended by Bai & Clark ( 2019 ), in order to further minimize selection bias in examining the effect of JE participation. Separate Kaplan-Meier survival analyses with the log rank test of equality of survivor curves were also conducted to explore time to first re-arrest from the enrollment start dates as well as the enrollment end dates. The alpha for all analyses was .05. Results Tables 2 and 3 show the results of the Chi Square analysis of the proportions of youth re-arrested after the program enrollment start and end dates, respectively. Table 2 Re-arrests After Program Start Date by Group Assignment Group JE Group Matched Comparison Group n (%) n (%) Re-arrested 48 (32.4%) 71 (48%) Not re-arrested 100 (67.6%) 77 (52%) Total 148 (100%) 148 (100%) p = .004 Table 3 Re-arrests After Program End Date by Group Assignment Group JE Group Matched Comparison Group n (%) n (%) Re-arrested 13 (8.8%) 35 (23.6%) Not re-arrested 135 (91.2%) 113 (76.4%) Total 148 (100%) 148 (100%) p < .001 [Table 2 here] [Table 3 here] A significantly lower proportion of JE youth were arrested relative to the comparison group youth after both enrollment start, X 2 (1, N = 296) = 7.43, p = .004 and end dates, X 2 (1, N = 296) = 12.04, p < .001, although the re-arrest rate decreased for both groups by the enrollment end dates. Table 4 shows the results of the logistic regression of being re-arrested on JE participation controlling for covariates, using the enrollment end dates. Table 4 Logistic regression of re-arrest on covariates from enrollment end date Variable B SE B Wald p OR 95% CI for OR Lower Upper JE group (= 1) -1.86 0.48 14.92 0.000 0.16 0.06 0.40 Age at first assessment -2.97 1.64 3.27 0.071 0.05 0.00 1.28 Male (= 1) 0.33 0.58 0.32 0.571 1.39 0.45 4.33 Ethnicity ("Other" is reference) Hispanic (= 1) -3.23 1.77 3.35 0.067 0.04 0.00 1.26 Caucasian (= 1) 1.80 1.78 1.03 0.311 6.07 0.19 199.18 Black (= 1) -3.63 2.44 2.20 0.138 0.03 0.00 3.20 Asian (= 1) 0.83 1.64 0.26 0.612 2.30 0.09 57.61 Absenteeism (= 1) 1.72 0.61 7.81 0.005 5.57 1.67 18.57 Behavior problems (= 1) 0.01 0.47 0.00 0.989 1.01 0.40 2.52 Poor grades (= 1) -7.04 3.88 3.28 0.070 < 0.00 0.00 1.78 In special education (= 1) -0.08 0.40 0.04 0.833 0.92 0.42 2.01 Mental health (= 1) 2.50 2.22 1.27 0.260 12.15 0.16 940.69 Parent history justice system (= 1) 1.03 0.93 1.23 0.267 2.81 0.45 17.49 Parent history drinking, drugs (= 1) 0.36 0.44 0.67 0.413 1.43 0.61 3.36 JAIS risk level ("Low risk" is reference) Medium_risk (= 1) 1.59 0.93 2.92 0.088 4.89 0.79 30.19 High_risk (= 1) 3.28 1.56 4.44 0.035 26.57 1.26 561.99 Propensity score 36.00 23.16 2.42 0.120 4.33E + 15 0.00 2.23E + 35 Constant 38.70 20.91 3.43 0.064 6.44E + 16 OR = odds ratio; CI = confidence interval [Table 4 here] The Hosmer and Lemeshow test of model fit shows that the overall model is predictive of re-arrest, X 2 (8) = 10.8, p = .221. After controlling for all covariates, the youth in the JE group were less likely to be re-arrested, (OR = 0.16 p < .001). Of the other covariates, school absenteeism (OR = 5.57, p = .005) and having a JAIS assessment as high risk (compared to low risk) (OR = 26.56, p = .035) showed an impact on increasing the odds of re-arrest. Since observational studies are vulnerable from threats to internal validity, a sensitivity analysis was performed in order to explore the possibility that unmeasured confounders might affect the association between JE participation and re-arrest. We calculated the E-value, an estimate of sensitivity, as presented by VanderWeele & Ding ( 2017 ). The E-value is a single estimate of the minimum strength of an omnibus confounder, in the form of a risk ratio, that would explain the effect in a way that makes a causal claim no longer credible, over and above the actual confounders already in the model. A large E-value implies that considerable unmeasured confounding would be required to fully explain away the association between JE and re-arrest, whereas a small E-value indicates that even a weak unmeasured confounder could explain the association (i.e. the lowest value of 1.0 indicates no confounders would be necessary to nullify the association). The E-value for the existing model was 27.92. Although there is no specific validated cutoff since the interpretation of the E-value is contingent on existing measured confounders, this value would be considered very high, indicating that the existing model is robust and adequate to explain the association between JE and re-arrest. Figures 1 and 2 show the Kaplan-Meier analyses of time to re-arrest by enrollment start and end dates, respectively. [Figure 1 here] [Figure 2 here] In Fig. 1 , the JE youth were re-arrested sooner than comparison youth after their program start dates, X 2 (1, N = 119) = 11.05, p < .001. The difference between the two groups, however, disappears for time to re-arrest after the enrollment end date, as shown in Fig. 2, X 2 (1, N = 48) = 1.16, p = .281. We were also curious about whether length of time in the program for JE youth was associated with time to first re-arrest. An additional Cox proportional hazards regression was performed of time to re-arrest after program end with all independent variables plus JE program length (not shown in tables). There was no association between length of program and time to re-arrest, neither for the entire sample nor for the JE group. Discussion The JusticeEd Program educational support program has had an impact on re-arrests – a smaller proportion than those of a matched comparison group of youth considered for, or referred to, educational services. This is most likely due to the program’s intensive approach to supporting families, acting as liaison to the school, and removing barriers to the parent’s and student’s engagement with school. JE youth had a faster time to re-arrest than matched comparison youth using time to rearrests from their enrollment start dates. This may be due to the disproportionate number of youth with behavioral problems in the JE group. The time-to-re-arrest difference between the two groups, however, disappears for re-arrests occurring after the enrollment end dates. Further analysis is recommended of factors related to time to re-arrest considering that both groups experienced re-arrest, and considering the differences in survival curves depending on whether re-arrest was measured after program start vs. program end. There are important limitations of the study. We note that the comparison group consists of youth with a variety of educational needs. Some of the services students actually received beyond special education support, which were not noted in the data, could result in treatment imitation. In some studies this might reduce the differences in re-arrest between the treatment and comparison groups. This was not the case in our analysis, although we note that the comparison group youth were likely to have benefited from the various support services available to them from the probation system. Subsequent studies should include a more thorough description of the types of support received from comparison group youth. Although there were some covariate differences between the two groups, they were within the threshold as defined by PSM authors. Data limited to a specific time period are also subject to incomplete information about arrests that may occur after the end of the date range for the data set. In addition, missing from the data are youth who completed the JE Program but their probation files were sealed and data not shared with the researchers. These youths’ characteristics could differ from youth whose records were not sealed, which could affect differences in their outcomes. One difference, for example, might be duration to first arrest. Eliminating seven JE youth due to missing propensity scores reduced the sample size, however, including them would likely have resulted in an even lower re-arrest rate for that group. We also note that the specific geographical location of the study site, with its predominately Hispanic population in the probation system, might limit the generalizability of findings to other study sites and populations. About the procedure used to identify the index re-arrest date - applying the individual JE youths’ enrollment start and end dates to their matched comparison youths relies on an assumption about the similarity of potential treatment duration for the comparison group, which is a theoretical, not a data-based, assumption. A potential research design that could address this would be a waitlist control group if this were feasible, where control group youths’ actual waitlist time period dates could be used to identify their index re-arrest. While PSM, under certain conditions, is considered an adequate substitute for a randomized treatment control design, any limitations in the reliability of the risk data derived from self-report could lead to inaccurate matching. With the current data, the PSM procedures performed well on some risk factors but not others – namely age at first assessment and having behavior problems. Differences between groups of average age and behavior problems, for example, might be important factors in influencing arrest outcomes. Although JE youth were younger than their matched comparisons, they had substantially more behavioral problems, suggesting that this confounder may make rearrest in the treated group more likely. This further strengthens confidence in the findings. As with any observational study, there is always the possibility that unmeasured confounders are affecting the outcome of re-arrest, such as family support, neighborhood factors, illicit drug involvement, etc. However, the sensitivity analysis showed that our model with the limited data available to us was robust enough to show a positive effect of JE on youth re-arrest. Despite these limitations, this study shows promising evidence of the impact of educational support for both youth and caregivers, in order to re-engage with school and protect against the risks leading to subsequent involvement with the justice system. A next step for research could be analyzing an expanded comparison group of all youth arrested, not limited to those referred to educational support. Such a comparison group might include youth not referred to educational services but who might also have similar risk factors which could be matched to youth receiving specialized services. In addition, process evaluation studies could provide insight into which components of educational support work better for specific youth and families. The available data mainly included risk factors, however there are familial and community resiliencies that most likely affect the outcomes of youth. For Hispanic families, for example, these include an emphasis on rule setting, family values, successfully navigating acculturative tasks, and valuing the importance of education (Halgunseth et al., 2006 ). These should be measured in subsequent studies. Longitudinal research could monitor continued school engagement, graduation rates, and the likelihood of subsequent school suspensions or dropout. Such studies might also include the prospective recruitment and observation of the experiences of transition-age youth whose court records would be sealed after juvenile case closure. Declarations Author Contribution E.C. headed the study and wrote the draft manuscript. R.R. provided statistical analyses and contributed text to substantial sections of the manuscript. Both authors reviewed the final draft for submission. Acknowledgement The authors acknowledge the collaboration with staff of the National Center for Youth Law, and thank the County of Santa Clara Department of Probation for supplying data and their helpful comments. References Abrams, L. S., Terry, D., & Franke, T. M. (2011). Community-based juvenile reentry services: The effects of service dosage on juvenile and adult recidivism. Journal of Offender Rehabilitation , 50 (8), 492–510. https://doi.org/10.1080/10509674.2011.596919 Adamsons, K., & Russell, B. (2023). Longitudinal transmission of risk behaviors between mothers, fathers, and adolescents. Journal of Family Psychology , 37 (7), 1016–1025. https://doi.org/10.1037/fam0001086 Athanassiou, U., Green, M. J., Tzoumakis, S., Whitten, T., Laurens, K. R., Harris, F., Carr, V. J., & Dean, K. (2023). Relationships between parental mental illness and/or offending and offspring contact with the police in childhood: Findings from a longitudinal record‐linkage study. Criminal Behaviour and Mental Health , 33 (1), 72–84. https://doi.org/10.1002/cbm.2274 Bai, H., & Clark, M. H. (2019). Propensity score methods and applications . Sage. Baird, C. (2018). The CAIS/JAIS approach to assessment. In J. P. Singh, D. G. Kroner, J. S. Wormith, S. L. Desmarais, & Z. Hamilton (Eds.), Handbook of recidivism risk/needs assessment tools (pp. 33–47). John Wiley & Sons. Barrett, D. E., Katsiyannis, A., Zhang, D., & Zhang, D. (2014). Delinquency and recidivism: A multicohort, matched-control study of the role of early adverse experiences, mental health problems, and disabilities. Journal of Emotional and Behavioral Disorders , 22 (1), 3–15. https://doi.org/10.1177/1063426612470514 Blomberg, T. G., Bales, W. D., Mann, K., Piquero, A. R., & Berk, R. A. (2011). Incarceration, education and transition from delinquency. Journal of Criminal Justice , 39 (4), 355–365. https://doi.org/10.1016/j.jcrimjus.2011.04.003 Calley, N. G. (2012). Juvenile offender recidivism: An examination of risk factors. Journal of Child Sexual Abuse , 21 (3), 257–272. https://doi.org/10.1080/10538712.2012.668266 Carpenter, B. W., Young, M. D., Bowers, A., & Sanders, K. (2016). Family involvement at the secondary level: Learning from Texas borderland schools. NASSP Bulletin , 100 (1), 47–70. https://doi.org/10.1177/0192636516648208 Dembo, R., & Gulledge, L. M. (2009). Truancy intervention programs: Challenges and innovations to implementation. Criminal Justice Policy Review , 20 (4), 437–456. https://doi.org/10.1177/0887403408327923 Evident Change. (2025). Juvenile Assessment and Intervention System TM (JAIS) . https://evidentchange.org/assessment/juvenile-assessment-and-intervention-system-jais/ Gary, J. (2014). Education connection: A prison to school pipeline: Preparing incarcerated youth for reentry into society and public school. Children’s Legal Rights Journal , 34 , 241–244. Halgunseth, L. C., Ispa, J. M., & Rudy, D. (2006). Parental control in Latino families: An integrated review of the literature. Child Development , 77 (5), 1282–1297. John, A., Mosley, S., Wheeler, S., Okech, J., & Lewright, C. (2025). Second Opportunity for Success: Evaluation of a juvenile diversion program. Child & Youth Services , 46 (1), 129–152. https://doi.org/10.1080/0145935X.2023.2243437 Lee, J. A. S., & Taxman, F. S. (2020). Using latent class analysis to identify the complex needs of youth on probation. Children and Youth Services Review , 115 (May), 105087. https://doi.org/10.1016/j.childyouth.2020.105087 Martinez-Fuentes, S., Jager, J., & Umaña-Taylor, A. J. (2021). The mediation process between Latino youths’ family ethnic socialization, ethnic-racial identity, and academic engagement: Moderation by ethnic-racial discrimination? Cultural Diversity & Ethnic Minority Psychology , 27 (2), 296–306. https://doi.org/10.1037/cdp0000349 Mena, J. A. (2011). Latino parent home-based practices that bolster student academic persistence. Hispanic Journal of Behavioral Sciences , 33 (4), 490–506. https://doi.org/10.1177/0739986311422897 National Center for Education Statistics. (2024). Status Dropout Rates . https://nces.ed.gov/programs/coe/indicator/coj/status-dropout-rates National Center for Youth Law. (2022, May 19). JusticeEd: Year Two Progress Update . https://youthlaw.org/resources/justiceed-year-two-progress-update Puzzanchera, C., Hockenberry, S., & Sickmund, M. (2022). Youth and the juvenile justice system: 2022 National Report . National Center for Juvenile Justice. Robertson, A. A., & Walker, C. S. (2018). Predictors of justice system involvement: Maltreatment and education. Child Abuse & Neglect , 76 , 408–415. https://doi.org/10.1016/j.chiabu.2017.12.002 Safavian, N., Lee, G., Dicke, A.-L., Karabenick, S. A., & Eccles, J. S. (2022). Disentangling perceived educational support sources and types in adolescence and Latinas’ educational attainment in adulthood. Hispanic Journal of Behavioral Sciences , 44 (2), 123–148. https://doi.org/10.1177/07399863231153292 Schwalbe, C. S., Gearing, R. E., MacKenzie, M. J., Brewer, K. B., & Ibrahim, R. (2012). A meta-analysis of experimental studies of diversion programs for juvenile offenders. Clinical Psychology Review , 32 (1), 26–33. https://doi.org/10.1016/j.cpr.2011.10.002 Sheldon, S. B., & Epstein, J. L. (2004). Getting students to school: Using family and community involvement to reduce chronic absenteeism. School Community Journal , 14 (2), 39–56. Simon, E., Smith, T. J., & Dillahunt-Aspillaga, C. (2025). Interactions of juveniles with intellectual and developmental disabilities in the criminal justice system: A systematized review. Journal of Vocational Rehabilitation , 10522263241310062. https://doi.org/10.1177/10522263241310062 Spooner, K., Pyrooz, D., Webb, V., & Fox, K. (2017). Recidivism among juveniles in a multi-component gang reentry program: Findings from a program evaluation in Harris County, Texas. Journal of Experimental Criminology , 13 (2), 275–285. https://doi.org/10.1007/s11292-017-9288-0 VanderWeele, T. J., & Ding, P. (2017). Sensitivity analysis in observational research: Introducing the E-Value. Annals of Internal Medicine , 167 (4), 268–274. https://doi.org/10.7326/M16-2607 Wilson, H. A., & Hoge, R. D. (2013). The effect of youth diversion programs on recidivism. Criminal Justice and Behavior , 40 (5), 497–518. https://doi.org/10.1177/0093854812451089 Zhang, D., Barrett, D. E., Katsiyannis, A., & Yoon, M. (2011). Juvenile offenders with and without disabilities: Risks and patterns of recidivism. Learning and Individual Differences , 21 (1), 12–18. https://doi.org/10.1016/j.lindif.2010.09.006 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6994635","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":487059751,"identity":"642cc0e9-1c04-49d8-a65a-9539e62407f2","order_by":0,"name":"Edward Cohen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7ElEQVRIiWNgGAWjYFCCBCBmA2IJIP4AZkHYxGlhnEGyFmYeqBheLfzsOYafK8oY5Pmlm489tvnDZ29wgPngbR48WiR73hhLnjnHYDhzzrF049w2tsQNB9iSrfFpMbiRu0GysY0hweBGjpl0bgNbgsEBHjNpfFrsb+Ru/gnRkv9N2uIPG9Bh/N/wajGQyN0Gs4VNmoGNjXHDAR42vFokzrz/ZtlwTsJw5ow0M8leoF9mHmYztpyDRwt/e1ryzYYyG3l+ieRnEj/+HLPnO9788MYbPFpglsEYx4CxQ1g5CqghUf0oGAWjYBSMBAAAERNHDmOlVroAAAAASUVORK5CYII=","orcid":"","institution":"San Jose State University","correspondingAuthor":true,"prefix":"","firstName":"Edward","middleName":"","lastName":"Cohen","suffix":""},{"id":487059752,"identity":"00cf1118-9aaf-4b10-8d15-41eb9af389d0","order_by":1,"name":"Roderick Rose","email":"","orcid":"","institution":"University of Maryland","correspondingAuthor":false,"prefix":"","firstName":"Roderick","middleName":"","lastName":"Rose","suffix":""}],"badges":[],"createdAt":"2025-06-27 22:53:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6994635/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6994635/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87359855,"identity":"b735dbab-96bd-45aa-bf4f-8083c3c43521","added_by":"auto","created_at":"2025-07-23 05:41:33","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":21611,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eTime to Arrest from Enrollment Start Date, by Group\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6994635/v1/196df0f5a4ec25a023cc9854.png"},{"id":87362816,"identity":"b6117378-cdd7-4d88-b803-14af91e09b31","added_by":"auto","created_at":"2025-07-23 05:57:33","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":165097,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eTime to Arrest from Enrollment End Date, by Group\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6994635/v1/5416fe8de947764fd75aa300.jpeg"},{"id":87362817,"identity":"52798943-c5ee-4dce-bdc0-c2e5eb371d32","added_by":"auto","created_at":"2025-07-23 05:57:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":842577,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6994635/v1/897f33b0-7539-490b-a7d7-64d3494680c9.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Educational Support Reduces Arrest Recidivism for Adolescents","fulltext":[{"header":"Highlights","content":"\u003cul\u003e\n \u003cli\u003eSchool absenteeism and dropout are predictors of youth contact with police and arrest.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eEducational support support for both youth and caregivers is important to prevent adverse school events.\u003c/li\u003e\n \u003cli\u003eThis study examined the likelihood of juvenile re-arrest and time to re-arrest for participants in an educational support program at a large California county providing individualized educational case management, coaching, and mentoring to youth and parents, compared to a propensity-scored matched comparison group receiving treatment-as-usual.\u003c/li\u003e\n \u003cli\u003eBivariate and multivariate analyses showed that youth receiving the educational support program had a lower rate of re-arrest than comparison group youth.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eThe study showed the benefits of educational support for both caregivers and youth in order to re-engage with school and protect against the risks leading to subsequent re-involvement with the justice system.\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"Background","content":"\u003cp\u003eEducational support is one of the priorities that are required to prevent involvement of youth in the justice system (Lee \u0026amp; Taxman, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Puzzanchera et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Considering the importance of social support from the family on educational success for Hispanic students (Safavian et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), the majority ethnicity of this study\u0026rsquo;s participants, help for both the student and family are necessary. There are many barriers to engagement of Hispanic adolescents in schools, however the family is key to successful engagement by, among other supports, helping the adolescent navigate discrimination and cultural differences in day-to-day school life, and providing parental monitoring and supervision (Martinez-Fuentes et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). For Hispanic caregivers, an important issue is their own ability to communicate and become engaged with the school. Schools and support programs are more successful engaging students if they respect parents\u0026rsquo; views of the importance of education despite logistical barriers to attend meetings or communicate in English (Carpenter et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Improving parental engagement with the school will also positively affect students\u0026rsquo; attitudes about persevering academically (Mena, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAmong the predictors of delinquency and recidivism are a history of family violence, learning problems, having a disability, and having behavioral problems (Barrett et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Simon et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Parental mental illness and a history of justice involvement are implicated in their offspring\u0026rsquo;s contact with police (Athanassiou et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Calley, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), as is parental alcohol and drug use (Adamsons \u0026amp; Russell, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). One serious problem for Hispanic youth is the high rate of school dropout. Although high school dropout rates have been decreasing for all youth over the past decade, Hispanic males currently have higher dropout rates than males or females of all other groups except Native Americans (National Center for Education Statistics, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). School dropout, suspensions, and absenteeism have all been linked to involvement with the justice system (Robertson \u0026amp; Walker, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), yet there are emerging practices to prevent these events. Schools that go beyond a rigid disciplinary response and programs that address psychosocial issues for the youth and family have better outcomes in reducing school absences, and thereby improving school engagement (Dembo \u0026amp; Gulledge, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Partnerships between schools, parents, and community mentors can reduce chronic absenteeism that often leads to dropping out (Sheldon \u0026amp; Epstein, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2004\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eEducational achievement has a demonstrated effect on reducing the likelihood of re-arrest and the seriousness of subsequent offenses (Blomberg et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), however there are few studies of the impact of educational support on families and youth who come to the attention of the juvenile probation system, even though observers recommend as a best practice linkages between school and probation services to ensure student engagement and educational success (Gary, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Abrams et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) found that the length of time in a community re-entry program for high-risk juvenile offenders, which included robust support for educational attainment and the strengthening of bonds with caregivers, was protective against repeat convictions. However, for the most part re-entry programs have had mixed results in terms of re-arrests. A study of a re-entry program for gang-involved youth treated in Texas residential facilities, for example, showed no impact on recidivism rates relative to a comparison group within a year after discharge (Spooner et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Earlier interventions may have more opportunity to affect protective factors, such as a diversion program in Texas targeting youth arrested for the first time (John et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This program evaluation noted pre- to posttest improvements in relationships with parents and in school adjustment; recidivism, however, was not measured. Diversion programs (which predominantly serve pre-charged youth) whether they are designed to provide warning to the youth without further action, or more invasively require program participation to avoid further court action, have shown an impact on reducing juvenile recidivism rates in a meta-analysis by Wilson \u0026amp; Hoge (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). On the other hand, a meta-analysis of diversion programs by Schwalbe et al. (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) did not find an overall impact of these programs on recidivism. There were many differences among the studies covered in these meta-analyses, such as having different target populations and varied time periods measuring re-arrest. There were also too few studies with robust group designs. Although attention to school success was a component of many diversion programs, very few specifically targeted the provision of support to both youth and caregivers regarding school engagement.\u003c/p\u003e\u003cp\u003eGiven the gaps in the literature for early intervention programs that specifically provide educational support, the research questions are 1) For youth arrested, does an educational support program reduce re-arrests? 2) For youth who were re-arrested, what is the impact of an educational support program on how soon youth are re-arrested?\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStudy Site and Sample\u003c/h2\u003e\u003cp\u003eThis study addresses one program designed to re-engage justice-involved youth and their caregivers with school \u0026ndash; the JusticeEd (JE) Program of the National Center for Youth Law. The sample were all youth arrested in a large California county during the time period from 1/1/2017-6/30/2022 and assessed as needing some type of education-related services. The data included 965 unduplicated youth, with 7,204 arrests, of which 5,243 were sustained petitions - those that resulted in filing a petition with the court to proceed on charges against the youth, and the charges were found to have merit. Of these youth, 268 were referred to JE, and 155 youth received JE services. The remaining youth in the sample were referred to either a program for youth with learning disabilities and in need of an Individual Education Plan, or to legal representation related to addressing school compliance with school discipline or special education policies. Seven JE youth did not have enough data to complete the propensity score match, and were excluded from analysis yielding a final sample of 148. Of these seven youth, only one was re-arrested during the time periods analyzed. Using the available non-missing data, these youth had no statistically significant differences of any covariates from the larger sample of JE youth. A sample of 148 each in a treatment and a comparison group has a power of .99 to detect an effect size difference of 0.20 between the groups.\u003c/p\u003e\u003cp\u003eDescriptive statistics of variables used in the analysis are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cem\u003eTable 1. Descriptive Statistics Comparing Groups\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"625\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003eJE Youth\u003c/p\u003e\n \u003cp\u003eN (%) or mean (sd)\u003c/p\u003e\n \u003cp\u003eN=148\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003eMatched Comparison\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eN=148\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003eOverall non-JE Youth\u003c/p\u003e\n \u003cp\u003eN= (%) or mean (sd)\u003c/p\u003e\n \u003cp\u003eN=836\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eAverage age at first assessment (SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e16.0 (1.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e16.53 (1.05)\u003c/p\u003e\n \u003cp\u003eCompared to JE Youth \u003cem\u003ep\u0026nbsp;\u003c/em\u003e\u0026lt; .001\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e15.89 (1.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003cp\u003eMale (1)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eFemale (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e120 (81%)\u003c/p\u003e\n \u003cp\u003e28(19%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e127 (86%)\u003c/p\u003e\n \u003cp\u003e21 (14%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e685 (81.9%)\u003c/p\u003e\n \u003cp\u003e151 (18.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eAverage number of Arrests pre-JE (SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e0.81 (1.19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e0.74 (1.62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e0.91 (1.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eRace/Ethnicity\u003c/p\u003e\n \u003cp\u003eAsian\u003c/p\u003e\n \u003cp\u003eBlack\u003c/p\u003e\n \u003cp\u003eCaucasian\u003c/p\u003e\n \u003cp\u003eHispanic\u003c/p\u003e\n \u003cp\u003eOther (reference)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e4 (2.7%)\u003c/p\u003e\n \u003cp\u003e16 (10.8%)\u003c/p\u003e\n \u003cp\u003e10 (6.8%)\u003c/p\u003e\n \u003cp\u003e111 (75%)\u003c/p\u003e\n \u003cp\u003e7 (4.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e3 (2%)\u003c/p\u003e\n \u003cp\u003e10 (6.8%)\u003c/p\u003e\n \u003cp\u003e6 (4.1%)\u003c/p\u003e\n \u003cp\u003e127 (85.8%)\u003c/p\u003e\n \u003cp\u003e2 (1.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e56 (6.7%)\u003c/p\u003e\n \u003cp\u003e57 (6.8%)\u003c/p\u003e\n \u003cp\u003e106 (12.7%)\u003c/p\u003e\n \u003cp\u003e578 (69.1%)\u003c/p\u003e\n \u003cp\u003e39 (4.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eAbsenteeism\u003c/p\u003e\n \u003cp\u003eProblem (1)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eNot a problem (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e71 (48%)\u003c/p\u003e\n \u003cp\u003e77 (52%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e69 (47%)\u003c/p\u003e\n \u003cp\u003e79 (53%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e357 (42.7%)\u003c/p\u003e\n \u003cp\u003e479 (57.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eBehavior\u003c/p\u003e\n \u003cp\u003eProblem\u003c/p\u003e\n \u003cp\u003eNot a problem\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e98 (66%)\u003c/p\u003e\n \u003cp\u003e50 (33.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e37 (25%)\u003c/p\u003e\n \u003cp\u003e111 (75%)\u003c/p\u003e\n \u003cp\u003eCompared to JE Youth \u003cem\u003ep\u003c/em\u003e = .022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e260 (31.1%)\u003c/p\u003e\n \u003cp\u003e576 (68.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eGrades\u003c/p\u003e\n \u003cp\u003eProblem\u003c/p\u003e\n \u003cp\u003eNot a problem\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e98 (66.2%)\u003c/p\u003e\n \u003cp\u003e50 (33.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e92 (62.2%)\u003c/p\u003e\n \u003cp\u003e56 (39.8%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e406 (48.6%)\u003c/p\u003e\n \u003cp\u003e430 (51.4%)\u003c/p\u003e\n \u003cp\u003eCompared to JE Youth \u003cem\u003ep\u003c/em\u003e \u0026lt; .001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eOpen to Special Education\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eOpen to SE\u003c/p\u003e\n \u003cp\u003eNot open to SE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e103 (69.6%)\u003c/p\u003e\n \u003cp\u003e45 (30.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e92 (62.2%)\u003c/p\u003e\n \u003cp\u003e56 (37.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e534 (63.9%)\u003c/p\u003e\n \u003cp\u003e302 (36.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eMental health diagnosis\u003c/p\u003e\n \u003cp\u003eHas diagnosis\u003c/p\u003e\n \u003cp\u003eDoes not have not a diagnosis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e27 (18%)\u003c/p\u003e\n \u003cp\u003e121 (82%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e19 (12.8%)\u003c/p\u003e\n \u003cp\u003e129 (87%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e235 (28.1%)\u003c/p\u003e\n \u003cp\u003e601 (71.9%)\u003c/p\u003e\n \u003cp\u003eCompared to JE Youth \u003cem\u003ep\u003c/em\u003e \u0026lt; .007\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eAverage JAIS risk level\u003c/p\u003e\n \u003cp\u003eLow (reference)\u003c/p\u003e\n \u003cp\u003eMedium\u003c/p\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e61 (41.2%)\u003c/p\u003e\n \u003cp\u003e68 (45.9%)\u003c/p\u003e\n \u003cp\u003e19 (12.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e59 (36.9%)\u003c/p\u003e\n \u003cp\u003e77 (52%)\u003c/p\u003e\n \u003cp\u003e12 (8.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e269 (33.7%)\u003c/p\u003e\n \u003cp\u003e413 (51.8%)\u003c/p\u003e\n \u003cp\u003e116 (14.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eParent history justice involved\u003c/p\u003e\n \u003cp\u003eProblem\u003c/p\u003e\n \u003cp\u003eNot a problem\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e67 (45.3%)\u003c/p\u003e\n \u003cp\u003e81 (54.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e68 (45.9%)\u003c/p\u003e\n \u003cp\u003e80 (54.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e423 (50.6%)\u003c/p\u003e\n \u003cp\u003e413 (49.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 158px;\"\u003e\n \u003cp\u003eParent history drinking/drugs\u003c/p\u003e\n \u003cp\u003eProblem\u003c/p\u003e\n \u003cp\u003eNot a problem\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e60 (40.5%)\u003c/p\u003e\n \u003cp\u003e88 (59.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 162px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e52 (35.1%)\u003c/p\u003e\n \u003cp\u003e96 (64.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 154px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e348 (41.6%)\u003c/p\u003e\n \u003cp\u003e488 (58.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eDifferences between groups not statistically significant unless otherwise stated.\u0026nbsp;\u003c/p\u003e\u003cp\u003e[Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Here]\u003c/p\u003e\u003cp\u003eThe majority of youth in the full sample were male (82%) and identified as Hispanic (69%). The full list of potential comparison group youth included a) those with learning disabilities and in need of an Individual Education Plan, b) those referred to legal representation, c) those youth referred to JE but did not accept services, and d) those eligible for referral to educational support but determined not to need specialized services at the time of assessment. The IRB at _____University exempted the project from human subjects committee review since the data were de-identified. A university-county memorandum of understanding was developed covering data sharing, as well as a no-cost evaluation agreement between the university and the National Center for Youth Law.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eResearch Design\u003c/h3\u003e\n\u003cp\u003eRandomization of youth to educational support services was not feasible at the time of post-arrest assessment since all youth who met the initial criteria based on screening by probation staff were referred to educational support services. Sufficient data were available to conduct a quasi-experimental matched comparison group post-test design. Propensity score matching (PSM) was used to construct a matched comparison group. The PSM procedure involved using logistic regression to generate probability scores for each youth on the nominal dependent variable referral to JE. The assumption is that although they did not meet initial criteria for referral, some youth not referred to JE (or those referred but did not receive services) might nevertheless have benefited from the JE program, and share many risk factors of those who participated in JE. We selected the comparison group cases using nearest neighbor matching.\u003c/p\u003e\n\u003ch3\u003eMeasures\u003c/h3\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e lists the variables used for analysis and their characteristics.\u003c/p\u003e\n\u003ch3\u003eRe-arrest\u003c/h3\u003e\n\u003cp\u003eAll arrests in the study were those that resulted in filing a petition for a charge that was sustained by the juvenile court. Charges that were not filed or sustained (e.g. the charges were dropped or there was insufficient evidence) were excluded from analysis. Identifying the index rearrest date for JE youth was based either on the program\u0026rsquo;s enrollment start or end date \u0026ndash; analyses included both scenarios. No enrollment start or end dates were available for the comparison youth. Establishing a comparable time period to identify the index re-arrest date for those youth proved challenging since the JE youth had rolling enrollment dates throughout the range of the data. To create an equitable time period for the comparison group youth, we applied the same treatment entry and exit dates of the JE youth to the corresponding comparison youth matched on their nearest propensity score. The assumption was that if a comparison group youth with similar risk factors were in the JE group, their length of treatment would have been similar. For the survival analysis, we calculated the number of days to the first arrest after the enrollment start dates and, alternatively, the number of days to the first arrest after the enrollment end dates. Survival analysis also requires a nominal variable indicating whether the re-arrest event occurred (coded 1) or whether it did not occur by the end of the observation period (coded 0), again coded separately for arrests after both JE enrollment start and end dates.\u003c/p\u003e\n\u003ch3\u003eYouth and Family Characteristics\u003c/h3\u003e\n\u003cp\u003eThe data included age at first assessment and gender, a dichotomous nominal variable (0\u0026thinsp;=\u0026thinsp;female, 1\u0026thinsp;=\u0026thinsp;male, the only gender attributes reported). Race or ethnicity was provided for American Indian, Asian, Black, Caucasian, Hispanic, and \u0026ldquo;other\u0026rdquo; categories. For the logistic regression with the re-arrest outcome, dummy variables for each race and ethnicity attribute were coded 1 or 0. \u0026ldquo;Other\u0026rdquo; was the excluded reference variable. School problem areas identified in the Juvenile Assessment and Intervention System\u0026trade; (JAIS) (0\u0026thinsp;=\u0026thinsp;no problem; 1\u0026thinsp;=\u0026thinsp;has problem) included absenteeism, behavior problem, poor grades, mental health problem, and whether the student was in special education. The JAIS system was meant to assess needs for planning as well as risk factors. It was originally developed and validated for adults (Baird, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) and more recently adapted for youth (Evident Change, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The JAIS risk level was determined upon assessment, and resulted in an ordinal variable \u0026ndash; low, medium, or high risk. If youth were assessed more than once, the most frequent risk score was used. For count ties, the risk level associated with the latest arrest was used. In the PSM matching procedure, one risk level variable with three attributes was used, however in the logistic regression for the outcome analysis dummy variables for \u0026ldquo;high\u0026rdquo; and \u0026ldquo;medium\u0026rdquo; were included with \u0026ldquo;low\u0026rdquo; excluded as the reference. Parental problems assessed by the JAIS (0\u0026thinsp;=\u0026thinsp;no problem; 1\u0026thinsp;=\u0026thinsp;has problem) included history of drinking or drug use and history of involvement in the justice system.\u003c/p\u003e\u003cp\u003eVariables used for the PSM procedure included: a) Average age of child at first assessment, b) Male gender vs. female, c) Race/ethnicity, d) School absenteeism, e) Behavior problem, f) Poor grades, g) Has open case in special education, h) Youth has a mental health diagnosis, i) JAIS risk score, j) Parent history of justice involvement, and k) Parent history of drinking/drug use.\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the descriptive statistics comparing the JE group with the matched comparison and overall non-JE group. The JE and matched comparison groups were statistically different for variables average age at assessment (matched comparison youth averaged half a year older than JE youth) and behavior problem (JE youth had a higher rate). \u0026ldquo;Off-the-shelf\u0026rdquo; demographic variables such as age, race/ethnicity, or gender alone do not typically account for imbalances between non-randomly assigned conditions, but the additional characteristics from the JAIS and referral assessment (represented by d-k) are associated with rearrest and referral and we propose that residual imbalances between groups should have a negligible effect on the findings. For a test of the balanced distribution of independent variables we calculated the standard mean deviation (SMD) for each variable between the JE group and matched comparison group (not shown in table). Bai \u0026amp; Clark (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) recommend a difference of the SMD below .5, and that there should be at least a 75% overlap for balanced distribution of covariates. For this study, a cutoff of equal to or lower than .2 SMD was chosen as the threshold for balanced covariates. Of all the independent variables, age at first assessment, behavior problem and Hispanic were above the SMD threshold for the matched data. There were no differences between the groups regarding the other risk factors as well as the average number of arrests prior to the start of the JE implementation. The pre-PSM groups, by comparison, had two variables that exceeded the SMD threshold \u0026ndash; they were having poor grades and youth mental health problem. Group differences for these variables disappeared in the post-PSM groups.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eProgram Procedures\u003c/h2\u003e\u003cp\u003eSince August 2018, the JE Program has provided students and their families with individualized educational case management, coaching, and mentoring to support students in achieving their academic goals. Students and families are referred to educational support if they experienced a suspension or expulsion; there are attendance problems; the youth need help with employment, vocational or college readiness skills; or there is a need to help guide the youth and parent towards developing future educational goals. The program uses trained paraprofessionals as Educational Liaisons who a) review and discuss academic progress; b) provide coaching on study skills and social-emotional skill development; c) develop and review education plans and assess progress on student-set education goals; d) engage in post-secondary and career exploration, and d) assist the youth by monitoring completion of school work (National Center for Youth Law, 2022). JE staff work closely with caregivers in order to encourage them to be active, as education champions, in their youths\u0026rsquo; education. Typical activities for caregivers with English as a second language, for example, are facilitating meetings with school personnel, providing Spanish translation, and explaining school policies.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eAnalysis Procedures\u003c/h3\u003e\n\u003cp\u003eWith the matched treatment and comparison groups, we conducted two separate bivariate chi square test of association analyses to determine differences between the groups of the number of youth re-arrested, using both the enrollment start dates and enrollment end dates to identify the index re-arrest. We also conducted a separate \u0026ldquo;doubly robust\u0026rdquo; logistic regression of re-arrest after the enrollment end dates on covariates, including propensity scores as recommended by Bai \u0026amp; Clark (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), in order to further minimize selection bias in examining the effect of JE participation. Separate Kaplan-Meier survival analyses with the log rank test of equality of survivor curves were also conducted to explore time to first re-arrest from the enrollment start dates as well as the enrollment end dates. The alpha for all analyses was .05.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eTables\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e show the results of the Chi Square analysis of the proportions of youth re-arrested after the program enrollment start and end dates, respectively.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eRe-arrests After Program Start Date by Group Assignment\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGroup\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eJE Group\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMatched Comparison Group\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRe-arrested\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e48 (32.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e71 (48%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNot re-arrested\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e100 (67.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e77 (52%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e148 (100%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e148 (100%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.004\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eRe-arrests After Program End Date by Group Assignment\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGroup\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eJE Group\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMatched Comparison Group\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRe-arrested\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13 (8.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e35 (23.6%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNot re-arrested\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e135 (91.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e113 (76.4%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e148 (100%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e148 (100%)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;.001\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e[Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e here]\u003c/p\u003e\u003cp\u003e[Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e here]\u003c/p\u003e\u003cp\u003eA significantly lower proportion of JE youth were arrested relative to the comparison group youth after both enrollment start, \u003cem\u003eX\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e(1, N\u0026thinsp;=\u0026thinsp;296)\u0026thinsp;=\u0026thinsp;7.43, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.004 and end dates, \u003cem\u003eX\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e(1, N\u0026thinsp;=\u0026thinsp;296)\u0026thinsp;=\u0026thinsp;12.04, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001, although the re-arrest rate decreased for both groups by the enrollment end dates.\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the results of the logistic regression of being re-arrested on JE participation controlling for covariates, using the enrollment end dates.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eLogistic regression of re-arrest on covariates from enrollment end date\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"8\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eB\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eSE B\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eWald\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eOR\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u003cp\u003e95% CI for OR\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eLower\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eUpper\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eJE group (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-1.86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e14.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.40\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge at first assessment\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-2.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.071\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e1.28\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.571\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e4.33\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEthnicity (\"Other\" is reference)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHispanic (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-3.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.067\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e1.26\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCaucasian (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.311\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e6.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e199.18\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBlack (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-3.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2.20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.138\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e3.20\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAsian (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.612\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e57.61\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAbsenteeism (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.005\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e5.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e18.57\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBehavior problems (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.989\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e2.52\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePoor grades (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-7.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.070\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e1.78\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIn special education (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.833\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e2.01\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMental health (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.260\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e12.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e940.69\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eParent history justice system (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.267\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e17.49\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eParent history drinking, drugs (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.413\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e3.36\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eJAIS risk level (\"Low risk\" is reference)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMedium_risk (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.088\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e30.19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHigh_risk (=\u0026thinsp;1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e3.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e4.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.035\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e26.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e561.99\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePropensity score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e36.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e23.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.120\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.33E\u0026thinsp;+\u0026thinsp;15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e2.23E\u0026thinsp;+\u0026thinsp;35\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConstant\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e38.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e20.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.064\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e6.44E\u0026thinsp;+\u0026thinsp;16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"8\"\u003eOR\u0026thinsp;=\u0026thinsp;odds ratio; CI\u0026thinsp;=\u0026thinsp;confidence interval\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e[Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e here]\u003c/p\u003e\u003cp\u003eThe Hosmer and Lemeshow test of model fit shows that the overall model is predictive of re-arrest, \u003cem\u003eX\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e (8)\u0026thinsp;=\u0026thinsp;10.8, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.221. After controlling for all covariates, the youth in the JE group were less likely to be re-arrested, (OR\u0026thinsp;=\u0026thinsp;0.16 \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). Of the other covariates, school absenteeism (OR\u0026thinsp;=\u0026thinsp;5.57, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.005) and having a JAIS assessment as high risk (compared to low risk) (OR\u0026thinsp;=\u0026thinsp;26.56, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.035) showed an impact on increasing the odds of re-arrest.\u003c/p\u003e\u003cp\u003eSince observational studies are vulnerable from threats to internal validity, a sensitivity analysis was performed in order to explore the possibility that unmeasured confounders might affect the association between JE participation and re-arrest. We calculated the E-value, an estimate of sensitivity, as presented by VanderWeele \u0026amp; Ding (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The E-value is a single estimate of the minimum strength of an omnibus confounder, in the form of a risk ratio, that would explain the effect in a way that makes a causal claim no longer credible, over and above the actual confounders already in the model. A large E-value implies that considerable unmeasured confounding would be required to fully explain away the association between JE and re-arrest, whereas a small E-value indicates that even a weak unmeasured confounder could explain the association (i.e. the lowest value of 1.0 indicates no confounders would be necessary to nullify the association). The E-value for the existing model was 27.92. Although there is no specific validated cutoff since the interpretation of the E-value is contingent on existing measured confounders, this value would be considered very high, indicating that the existing model is robust and adequate to explain the association between JE and re-arrest.\u003c/p\u003e\u003cp\u003eFigures \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e show the Kaplan-Meier analyses of time to re-arrest by enrollment start and end dates, respectively.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e[Figure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e here]\u003c/p\u003e\u003cp\u003e[Figure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e here]\u003c/p\u003e\u003cp\u003eIn Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, the JE youth were re-arrested sooner than comparison youth after their program start dates, \u003cem\u003eX\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e (1, N\u0026thinsp;=\u0026thinsp;119)\u0026thinsp;\u003cem\u003e=\u003c/em\u003e\u0026thinsp;11.05, \u003cem\u003ep\u0026thinsp;\u0026lt;\u003c/em\u003e\u0026thinsp;.001. The difference between the two groups, however, disappears for time to re-arrest after the enrollment end date, as shown in Fig.\u0026nbsp;2, \u003cem\u003eX\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e (1, N\u0026thinsp;=\u0026thinsp;48)\u0026thinsp;\u003cem\u003e=\u003c/em\u003e\u0026thinsp;1.16, \u003cem\u003ep\u0026thinsp;=\u003c/em\u003e\u0026thinsp;.281. We were also curious about whether length of time in the program for JE youth was associated with time to first re-arrest. An additional Cox proportional hazards regression was performed of time to re-arrest after program end with all independent variables plus JE program length (not shown in tables). There was no association between length of program and time to re-arrest, neither for the entire sample nor for the JE group.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe JusticeEd Program educational support program has had an impact on re-arrests \u0026ndash; a smaller proportion than those of a matched comparison group of youth considered for, or referred to, educational services. This is most likely due to the program\u0026rsquo;s intensive approach to supporting families, acting as liaison to the school, and removing barriers to the parent\u0026rsquo;s and student\u0026rsquo;s engagement with school.\u003c/p\u003e\u003cp\u003eJE youth had a faster time to re-arrest than matched comparison youth using time to rearrests from their enrollment start dates. This may be due to the disproportionate number of youth with behavioral problems in the JE group. The time-to-re-arrest difference between the two groups, however, disappears for re-arrests occurring after the enrollment end dates. Further analysis is recommended of factors related to time to re-arrest considering that both groups experienced re-arrest, and considering the differences in survival curves depending on whether re-arrest was measured after program start vs. program end.\u003c/p\u003e\u003cp\u003eThere are important limitations of the study. We note that the comparison group consists of youth with a variety of educational needs. Some of the services students actually received beyond special education support, which were not noted in the data, could result in treatment imitation. In some studies this might reduce the differences in re-arrest between the treatment and comparison groups. This was not the case in our analysis, although we note that the comparison group youth were likely to have benefited from the various support services available to them from the probation system. Subsequent studies should include a more thorough description of the types of support received from comparison group youth. Although there were some covariate differences between the two groups, they were within the threshold as defined by PSM authors.\u003c/p\u003e\u003cp\u003eData limited to a specific time period are also subject to incomplete information about arrests that may occur after the end of the date range for the data set. In addition, missing from the data are youth who completed the JE Program but their probation files were sealed and data not shared with the researchers. These youths\u0026rsquo; characteristics could differ from youth whose records were not sealed, which could affect differences in their outcomes. One difference, for example, might be duration to first arrest. Eliminating seven JE youth due to missing propensity scores reduced the sample size, however, including them would likely have resulted in an even lower re-arrest rate for that group. We also note that the specific geographical location of the study site, with its predominately Hispanic population in the probation system, might limit the generalizability of findings to other study sites and populations.\u003c/p\u003e\u003cp\u003eAbout the procedure used to identify the index re-arrest date - applying the individual JE youths\u0026rsquo; enrollment start and end dates to their matched comparison youths relies on an assumption about the similarity of potential treatment duration for the comparison group, which is a theoretical, not a data-based, assumption. A potential research design that could address this would be a waitlist control group if this were feasible, where control group youths\u0026rsquo; actual waitlist time period dates could be used to identify their index re-arrest.\u003c/p\u003e\u003cp\u003eWhile PSM, under certain conditions, is considered an adequate substitute for a randomized treatment control design, any limitations in the reliability of the risk data derived from self-report could lead to inaccurate matching. With the current data, the PSM procedures performed well on some risk factors but not others \u0026ndash; namely age at first assessment and having behavior problems. Differences between groups of average age and behavior problems, for example, might be important factors in influencing arrest outcomes. Although JE youth were younger than their matched comparisons, they had substantially more behavioral problems, suggesting that this confounder may make rearrest in the treated group more likely. This further strengthens confidence in the findings. As with any observational study, there is always the possibility that unmeasured confounders are affecting the outcome of re-arrest, such as family support, neighborhood factors, illicit drug involvement, etc. However, the sensitivity analysis showed that our model with the limited data available to us was robust enough to show a positive effect of JE on youth re-arrest.\u003c/p\u003e\u003cp\u003eDespite these limitations, this study shows promising evidence of the impact of educational support for both youth and caregivers, in order to re-engage with school and protect against the risks leading to subsequent involvement with the justice system. A next step for research could be analyzing an expanded comparison group of all youth arrested, not limited to those referred to educational support. Such a comparison group might include youth not referred to educational services but who might also have similar risk factors which could be matched to youth receiving specialized services. In addition, process evaluation studies could provide insight into which components of educational support work better for specific youth and families. The available data mainly included risk factors, however there are familial and community resiliencies that most likely affect the outcomes of youth. For Hispanic families, for example, these include an emphasis on rule setting, family values, successfully navigating acculturative tasks, and valuing the importance of education (Halgunseth et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). These should be measured in subsequent studies. Longitudinal research could monitor continued school engagement, graduation rates, and the likelihood of subsequent school suspensions or dropout. Such studies might also include the prospective recruitment and observation of the experiences of transition-age youth whose court records would be sealed after juvenile case closure.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eE.C. headed the study and wrote the draft manuscript. R.R. provided statistical analyses and contributed text to substantial sections of the manuscript. Both authors reviewed the final draft for submission.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors acknowledge the collaboration with staff of the National Center for Youth Law, and thank the County of Santa Clara Department of Probation for supplying data and their helpful comments.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbrams, L. S., Terry, D., \u0026amp; Franke, T. M. (2011). Community-based juvenile reentry services: The effects of service dosage on juvenile and adult recidivism. \u003cem\u003eJournal of Offender Rehabilitation\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(8), 492\u0026ndash;510. https://doi.org/10.1080/10509674.2011.596919\u003c/li\u003e\n\u003cli\u003eAdamsons, K., \u0026amp; Russell, B. (2023). Longitudinal transmission of risk behaviors between mothers, fathers, and adolescents. \u003cem\u003eJournal of Family Psychology\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(7), 1016\u0026ndash;1025. https://doi.org/10.1037/fam0001086\u003c/li\u003e\n\u003cli\u003eAthanassiou, U., Green, M. J., Tzoumakis, S., Whitten, T., Laurens, K. R., Harris, F., Carr, V. J., \u0026amp; Dean, K. (2023). Relationships between parental mental illness and/or offending and offspring contact with the police in childhood: Findings from a longitudinal record‐linkage study. \u003cem\u003eCriminal Behaviour and Mental Health\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(1), 72\u0026ndash;84. https://doi.org/10.1002/cbm.2274\u003c/li\u003e\n\u003cli\u003eBai, H., \u0026amp; Clark, M. H. (2019). \u003cem\u003ePropensity score methods and applications\u003c/em\u003e. Sage.\u003c/li\u003e\n\u003cli\u003eBaird, C. (2018). The CAIS/JAIS approach to assessment. In J. P. Singh, D. G. Kroner, J. S. Wormith, S. L. Desmarais, \u0026amp; Z. Hamilton (Eds.), \u003cem\u003eHandbook of recidivism risk/needs assessment tools\u003c/em\u003e (pp. 33\u0026ndash;47). John Wiley \u0026amp; Sons.\u003c/li\u003e\n\u003cli\u003eBarrett, D. E., Katsiyannis, A., Zhang, D., \u0026amp; Zhang, D. (2014). Delinquency and recidivism: A multicohort, matched-control study of the role of early adverse experiences, mental health problems, and disabilities. \u003cem\u003eJournal of Emotional and Behavioral Disorders\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(1), 3\u0026ndash;15. https://doi.org/10.1177/1063426612470514\u003c/li\u003e\n\u003cli\u003eBlomberg, T. G., Bales, W. D., Mann, K., Piquero, A. R., \u0026amp; Berk, R. A. (2011). Incarceration, education and transition from delinquency. \u003cem\u003eJournal of Criminal Justice\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(4), 355\u0026ndash;365. https://doi.org/10.1016/j.jcrimjus.2011.04.003\u003c/li\u003e\n\u003cli\u003eCalley, N. G. (2012). Juvenile offender recidivism: An examination of risk factors. \u003cem\u003eJournal of Child Sexual Abuse\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(3), 257\u0026ndash;272. https://doi.org/10.1080/10538712.2012.668266\u003c/li\u003e\n\u003cli\u003eCarpenter, B. W., Young, M. D., Bowers, A., \u0026amp; Sanders, K. (2016). Family involvement at the secondary level: Learning from Texas borderland schools. \u003cem\u003eNASSP Bulletin\u003c/em\u003e, \u003cem\u003e100\u003c/em\u003e(1), 47\u0026ndash;70. https://doi.org/10.1177/0192636516648208\u003c/li\u003e\n\u003cli\u003eDembo, R., \u0026amp; Gulledge, L. M. (2009). Truancy intervention programs: Challenges and innovations to implementation. \u003cem\u003eCriminal Justice Policy Review\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(4), 437\u0026ndash;456. https://doi.org/10.1177/0887403408327923\u003c/li\u003e\n\u003cli\u003eEvident Change. (2025). \u003cem\u003eJuvenile Assessment and Intervention System\u003csup\u003eTM\u003c/sup\u003e (JAIS)\u003c/em\u003e. https://evidentchange.org/assessment/juvenile-assessment-and-intervention-system-jais/\u003c/li\u003e\n\u003cli\u003eGary, J. (2014). Education connection: A prison to school pipeline: Preparing incarcerated youth for reentry into society and public school. \u003cem\u003eChildren\u0026rsquo;s Legal Rights Journal\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e, 241\u0026ndash;244.\u003c/li\u003e\n\u003cli\u003eHalgunseth, L. C., Ispa, J. M., \u0026amp; Rudy, D. (2006). Parental control in Latino families: An integrated review of the literature. \u003cem\u003eChild Development\u003c/em\u003e, \u003cem\u003e77\u003c/em\u003e(5), 1282\u0026ndash;1297.\u003c/li\u003e\n\u003cli\u003eJohn, A., Mosley, S., Wheeler, S., Okech, J., \u0026amp; Lewright, C. (2025). Second Opportunity for Success: Evaluation of a juvenile diversion program. \u003cem\u003eChild \u0026amp; Youth Services\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(1), 129\u0026ndash;152. https://doi.org/10.1080/0145935X.2023.2243437\u003c/li\u003e\n\u003cli\u003eLee, J. A. S., \u0026amp; Taxman, F. S. (2020). Using latent class analysis to identify the complex needs of youth on probation. \u003cem\u003eChildren and Youth Services Review\u003c/em\u003e, \u003cem\u003e115\u003c/em\u003e(May), 105087. https://doi.org/10.1016/j.childyouth.2020.105087\u003c/li\u003e\n\u003cli\u003eMartinez-Fuentes, S., Jager, J., \u0026amp; Uma\u0026ntilde;a-Taylor, A. J. (2021). The mediation process between Latino youths\u0026rsquo; family ethnic socialization, ethnic-racial identity, and academic engagement: Moderation by ethnic-racial discrimination? \u003cem\u003eCultural Diversity \u0026amp; Ethnic Minority Psychology\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(2), 296\u0026ndash;306. https://doi.org/10.1037/cdp0000349\u003c/li\u003e\n\u003cli\u003eMena, J. A. (2011). Latino parent home-based practices that bolster student academic persistence. \u003cem\u003eHispanic Journal of Behavioral Sciences\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(4), 490\u0026ndash;506. https://doi.org/10.1177/0739986311422897\u003c/li\u003e\n\u003cli\u003eNational Center for Education Statistics. (2024). \u003cem\u003eStatus Dropout Rates\u003c/em\u003e. https://nces.ed.gov/programs/coe/indicator/coj/status-dropout-rates\u003c/li\u003e\n\u003cli\u003eNational Center for Youth Law. (2022, May 19). \u003cem\u003eJusticeEd: Year Two Progress Update\u003c/em\u003e. https://youthlaw.org/resources/justiceed-year-two-progress-update\u003c/li\u003e\n\u003cli\u003ePuzzanchera, C., Hockenberry, S., \u0026amp; Sickmund, M. (2022). \u003cem\u003eYouth and the juvenile justice system: 2022 National Report\u003c/em\u003e. National Center for Juvenile Justice.\u003c/li\u003e\n\u003cli\u003eRobertson, A. A., \u0026amp; Walker, C. S. (2018). Predictors of justice system involvement: Maltreatment and education. \u003cem\u003eChild Abuse \u0026amp; Neglect\u003c/em\u003e, \u003cem\u003e76\u003c/em\u003e, 408\u0026ndash;415. https://doi.org/10.1016/j.chiabu.2017.12.002\u003c/li\u003e\n\u003cli\u003eSafavian, N., Lee, G., Dicke, A.-L., Karabenick, S. A., \u0026amp; Eccles, J. S. (2022). Disentangling perceived educational support sources and types in adolescence and Latinas\u0026rsquo; educational attainment in adulthood. \u003cem\u003eHispanic Journal of Behavioral Sciences\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(2), 123\u0026ndash;148. https://doi.org/10.1177/07399863231153292\u003c/li\u003e\n\u003cli\u003eSchwalbe, C. S., Gearing, R. E., MacKenzie, M. J., Brewer, K. B., \u0026amp; Ibrahim, R. (2012). A meta-analysis of experimental studies of diversion programs for juvenile offenders. \u003cem\u003eClinical Psychology Review\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e(1), 26\u0026ndash;33. https://doi.org/10.1016/j.cpr.2011.10.002\u003c/li\u003e\n\u003cli\u003eSheldon, S. B., \u0026amp; Epstein, J. L. (2004). Getting students to school: Using family and community involvement to reduce chronic absenteeism. \u003cem\u003eSchool Community Journal\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(2), 39\u0026ndash;56.\u003c/li\u003e\n\u003cli\u003eSimon, E., Smith, T. J., \u0026amp; Dillahunt-Aspillaga, C. (2025). Interactions of juveniles with intellectual and developmental disabilities in the criminal justice system: A systematized review. \u003cem\u003eJournal of Vocational Rehabilitation\u003c/em\u003e, 10522263241310062. https://doi.org/10.1177/10522263241310062\u003c/li\u003e\n\u003cli\u003eSpooner, K., Pyrooz, D., Webb, V., \u0026amp; Fox, K. (2017). Recidivism among juveniles in a multi-component gang reentry program: Findings from a program evaluation in Harris County, Texas. \u003cem\u003eJournal of Experimental Criminology\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(2), 275\u0026ndash;285. https://doi.org/10.1007/s11292-017-9288-0\u003c/li\u003e\n\u003cli\u003eVanderWeele, T. J., \u0026amp; Ding, P. (2017). Sensitivity analysis in observational research: Introducing the E-Value. \u003cem\u003eAnnals of Internal Medicine\u003c/em\u003e, \u003cem\u003e167\u003c/em\u003e(4), 268\u0026ndash;274. https://doi.org/10.7326/M16-2607\u003c/li\u003e\n\u003cli\u003eWilson, H. A., \u0026amp; Hoge, R. D. (2013). The effect of youth diversion programs on recidivism. \u003cem\u003eCriminal Justice and Behavior\u003c/em\u003e, \u003cem\u003e40\u003c/em\u003e(5), 497\u0026ndash;518. https://doi.org/10.1177/0093854812451089\u003c/li\u003e\n\u003cli\u003eZhang, D., Barrett, D. E., Katsiyannis, A., \u0026amp; Yoon, M. (2011). Juvenile offenders with and without disabilities: Risks and patterns of recidivism. \u003cem\u003eLearning and Individual Differences\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(1), 12\u0026ndash;18. https://doi.org/10.1016/j.lindif.2010.09.006\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"adolescents, juvenile justice, arrests, educational support, Hispanic","lastPublishedDoi":"10.21203/rs.3.rs-6994635/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6994635/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEducational support for both youth and caregivers is one of the priorities that are required to prevent involvement of youth in the justice system. Since August 2018, JusticeEd has matched students with Education Liaisons that provide probation-involved students and their families with individualized educational case management, coaching, and mentoring. We examined the likelihood of re-arrest and time to re-arrest for participants in the JusticeEd (JE) program, in comparison to a sample of youth referred to educational support but who did not receive JusticeEd services. Data were requested from a large California county\u0026rsquo;s Juvenile Probation Department for all youth arrested and referred to educational services during a 4.5-year timeframe. The data included arrest dates, offenses, adjudication decisions, whether JE services were received, and youth and family risk factors assessed for all youth actively supervised by the Probation Department. Propensity score matching was used to match the comparison group with those receiving JE services. A lower proportion of JusticeEd youth than those in the comparison were re-arrested. For those re-arrested, event history analysis of time to re-arrest showed that the JE youth had a shorter time period to re-arrest during the program period, but after the program period the time to re-arrest was no different than the matched comparison youth. The JE program\u0026rsquo;s ability to affect re-arrests is most likely due to the program\u0026rsquo;s intensive approach to supporting families, acting as liaison to the school, and removing barriers to the parent\u0026rsquo;s and student\u0026rsquo;s engagement with school.\u003c/p\u003e","manuscriptTitle":"Educational Support Reduces Arrest Recidivism for Adolescents","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-23 05:41:28","doi":"10.21203/rs.3.rs-6994635/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"554bb214-0f75-4891-a997-d79f4f817425","owner":[],"postedDate":"July 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-07-23T05:41:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-23 05:41:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6994635","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6994635","identity":"rs-6994635","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}