Taking on Transformation is a project of the Columbia Justice Lab’s Youth Justice Initiatives. It is a hub for system leaders, and anyone making decisions about youth justice, who are pushing for real transformation of the youth justice system.
Transforming youth justice requires change from within systems, taking steps to align culture, information, and processes with the long term vision.
This module includes ways system leaders can change organizational culture, and create a data-driven culture and use data for action.
Undertaking large-scale system transformation depends on many different factors, including knowing where your jurisdiction currently sits relative to a desired vision; the detail and design of policy changes that move toward that vision; and the beliefs and behaviors of the people who will implement these policies. Collecting, understanding and using the right data can both provide a picture of where the jurisdiction sits with regard to the desired vision and provide a path towards that vision through a data-driven culture and keeping track of relevant metrics.
And lastly, real change depends on recognizing the significant role that organizational culture can play in advancing whether and how vision statements get translated and implemented in policies and practices that positively impact young people and their families on a daily basis. For this reason, taking steps to ensure that agency values and beliefs are internalized at all levels—from the agency director to frontline staff, and among community partners—is integral to achieving the ultimate goals of any transformation. Understanding the context of your system and making these changes from within the system first will be beneficial not only to the young people currently involved in the justice system but to a more seamless transformation process more broadly.
Organizational culture generally refers to the collective set of values, beliefs and behaviors that operate within an institution or agency. The culture of youth justice systems in particular have been shaped by their origins, reflecting the country’s history of racism and paternalism (see more in the chapter on Centering Racial Justice and Equity) and how the values and practices associated with these origins have been reinforced over time. Many other factors impact organizational culture, including current policies and practices, institutional structures, leadership, tenure and experience of staff, channels of formal and informal communication, partnerships, training and metrics. Staff who work with young people every day are a valuable resource and are necessary collaborators to bring new visions to reality in practice. Youth, families and communities are essential partners to defining and realizing a new vision for youth justice that will be youth-centered, family-focused and increasingly community-led. As leaders approach the task of shifting organizational culture to align with new values and goals, it is important to be humble, ask questions and be willing and eager to learn from partners inside and outside of the current system.
As part of this new culture, and in order to both understand the current context of youth justice and see a path to the new vision, data analysis is a powerful tool for transformation. Used thoughtfully, the right data can cast light on current system practice, illuminate the mandate for change, and direct leaders where to focus. Data analysis is essential for measuring progress, keeping leaders accountable to the goals and values of transformation and determining the impact of changes made. Making sense of data and turning it into useful information for action begins with having a clear vision of the results the system should achieve and the many drivers that affect those results. Leaders working to transform a system are more likely to attract support and sustain momentum among partners if they can show how the new direction improves system and individual results. In addition, clarifying how changes in each part of the system are expected to contribute to better results is essential to choosing the data to be tracked and to using data to correct course when necessary and ensure continuous improvement.
Indicators that progress has been made in shifting the culture of a youth justice agency to one that is aligned with transformation.
Indicators that a jurisdiction has made significant progress in developing its data systems to serve transformation.
These steps provide specific guidance for how to change systems from within, and taking steps to align culture, information, and processes. All of these steps are central in youth justice system reform, and can be implemented concurrently.
Engage staff in discussions with leadership about shared goals and values for the youth justice system, and examine existing policies and practices in terms of shared goals and values.
Providing consistent and clear communication around the vision and goals of transformation can build trust between leadership and staff and a healthy organizational culture that propels goals forward.
Invest in staff training and development that emphasizes youth well-being, racial equity and community-based support, and continue training and coaching over the course of transformation.
The data analysis team should include at least a senior manager who can develop the information agenda; a skilled data analyst who can perform statistical studies and analyze data sets, and knows how to present findings in accessible ways; a proficient technical expert with experience in extracting data in various formats from the management information system; and someone with the skills to create engaging visual representations of the data.
Working with the data analysis team, use the goals of transformation to analyze the data you will need to understand the current system, what is working to support youth well-being, what changes need to be made, and progress made with these changes and their success in better serving the goals of youth justice.
Learn about how to develop an ongoing staff review team with membership from staff at every level of the agency to facilitate organizational culture change.
Point-in-time analysis involves analysis of a given variable at a certain, fixed point in time, such as the number of admissions to a facility or the number of youth incarcerated in a facility on a given day, the number of youth enrolled in a certain program on a given day, or the average length of stay for youth currently incarcerated. Point-in-time analysis is especially useful for getting a sense of the day-to-day use of and demands on the system and can reveal places where certain resources are under-utilized. For example, this analysis may reveal that specific community programs are not being utilized by the system, which would lead to further analysis as to why, or it might reveal that a certain facility is largely empty, leading to further analyses to explore the possibility of its closure.
Trendline analysis compares point-in-time data over a given period of time—e.g., this month’s data compared with data from previous months or years—to examine the dynamics at work in different parts of the system over time. Usually, this is presented in the form of a simple line graph connecting the dots over given points in time. Examples of data fitting well into trendline analysis include arrests, intake, diversions, adjudications, specific dispositions, including out-of-home placements and average daily population of a facility or program, or average length of stay.
For example, a given trendline analysis might look at the number of diversions to community programs over a given period of time, the average daily population of a facility over a period of time, or the number of youth placed in facilities over a period of time. The results of the analysis may reveal a trend that appears to be aligned with transformation—such as decreasing numbers of youth sent to out-of-home placement, or a trend that calls for further work, such as increasing numbers of out-of-home placements. One trend analysis may call for additional analyses to examine potential underlying causes for a given trend. For example, increases in the daily population of a facility may lead to further analyses to investigate the cause(s) of these increases, such as increased admissions to the system, increased arrest rates, increased court commitments and/or increased filings by the prosecutor. A daily population increase might also reflect increased lengths of stay due to a lack of alternative placements or delays in court processing.
Superimposing multiple trend lines on one graph can also present patterns deserving of deeper investigation. In some cases, such comparisons can reveal problematic trends that lie below the surface of apparently favorable trends. For example, the average daily population trendline for a given jurisdiction’s placement facilities may show steady and fairly substantial reductions over the last decade. Yet the trendline for youth felony arrests may show an even steeper decline, suggesting that a higher proportion of youth with less serious charges are now being incarcerated. Trendline data is one type of comparison: today’s data compared to past data in the same jurisdiction.Other comparisons may include looking at the same data among a variety of similar jurisdictions.For example, a 40% decline in average daily population over a 15-year period may look like laudable progress in a particular jurisdiction, until this decline is compared with declines in other jurisdictions of over 60% during the same time period.
Cohort analysis examines data for a group of youth (a “cohort”) entering the system in the same given time period—for example, all youth who enter during a particular month or a quarter—and tracks their pathways through the system as a collective. This approach can reduce the risk of biased or incomplete analysis from using isolated point-in-time data, and is useful to track groups of youth that are subject to a specific change in policy or practice in order to gauge the effect of that change. For example, assume that 100 youth are detained upon arrest in a given jurisdiction during the month of March. Fifty of them are released back to their families almost immediately, and another 35 remain in detention for less than 30 days, with 15 remaining in detention for 30 days or more. Calculating the average length of stay for the detention population based on a one-day count (a point-in-time analysis) will over-estimate the actual average length of stay of youth in detention, because the count omits many of the youth who had very short stays, and youth with long stays will be over-represented. Cohort analysis can reveal opportunities for significant reduction in bed days, and increased use of alternative programs. For example, youth who leave detention within a day or two should probably be diverted from detention entirely, to a supportive program or otherwise. Case processing times can be improved for youth with very long stays, and considering alternative programs for these youth too can dramatically decrease the total number of bed days overall.
Disaggregation analysis breaks down a given data point by subgroups of youth with unique characteristics, such as race, ethnicity, gender, age, sexual orientation, and geography, to understand differences in practice that may exist based on those characteristics. For example, a jurisdiction might break down data on admissions to detention by a young person’s instant offense type and their race, to determine if Black youth are more likely to be detained than white youth while presenting similar instant offenses.
Statistical summaries when used alone can mask significant disparities in practice related to different subgroups of youth. For example, overall length of stay in detention or placement and daily institutional population maybe falling in the aggregate. But when these variables are disaggregated by race, the data may reveal that the length of stay and average daily population for Black youth have actually risen during the same time period, exacerbating racial disparities in the use of detention or placement. A similar finding might apply to young women, LBGTQ youth, youth from a particular neighborhood or youth charged with certain types of offenses. All data analysis should include disaggregation to assess for systemic disparities and well as system bias.
Learn about how to how to structure and utilize
participatory assessments to advance system transformation.
Point-in-time analysis involves analysis of a given variable at a certain, fixed point in time, such as the number of admissions to a facility or the number of youth incarcerated in a facility on a given day, the number of youth enrolled in a certain program on a given day, or the average length of stay for youth currently incarcerated. Point-in-time analysis is especially useful for getting a sense of the day-to-day use of and demands on the system and can reveal places where certain resources are under-utilized. For example, this analysis may reveal that specific community programs are not being utilized by the system, which would lead to further analysis as to why, or it might reveal that a certain facility is largely empty, leading to further analyses to explore the possibility of its closure.
Trendline analysis compares point-in-time data over a given period of time—e.g., this month’s data compared with data from previous months or years—to examine the dynamics at work in different parts of the system over time. Usually, this is presented in the form of a simple line graph connecting the dots over given points in time. Examples of data fitting well into trendline analysis include arrests, intake, diversions, adjudications, specific dispositions, including out-of-home placements and average daily population of a facility or program, or average length of stay.
For example, a given trendline analysis might look at the number of diversions to community programs over a given period of time, the average daily population of a facility over a period of time, or the number of youth placed in facilities over a period of time. The results of the analysis may reveal a trend that appears to be aligned with transformation—such as decreasing numbers of youth sent to out-of-home placement, or a trend that calls for further work, such as increasing numbers of out-of-home placements. One trend analysis may call for additional analyses to examine potential underlying causes for a given trend. For example, increases in the daily population of a facility may lead to further analyses to investigate the cause(s) of these increases, such as increased admissions to the system, increased arrest rates, increased court commitments and/or increased filings by the prosecutor. A daily population increase might also reflect increased lengths of stay due to a lack of alternative placements or delays in court processing.
Superimposing multiple trend lines on one graph can also present patterns deserving of deeper investigation. In some cases, such comparisons can reveal problematic trends that lie below the surface of apparently favorable trends. For example, the average daily population trendline for a given jurisdiction’s placement facilities may show steady and fairly substantial reductions over the last decade. Yet the trendline for youth felony arrests may show an even steeper decline, suggesting that a higher proportion of youth with less serious charges are now being incarcerated. Trendline data is one type of comparison: today’s data compared to past data in the same jurisdiction.Other comparisons may include looking at the same data among a variety of similar jurisdictions.For example, a 40% decline in average daily population over a 15-year period may look like laudable progress in a particular jurisdiction, until this decline is compared with declines in other jurisdictions of over 60% during the same time period.
Cohort analysis examines data for a group of youth (a “cohort”) entering the system in the same given time period—for example, all youth who enter during a particular month or a quarter—and tracks their pathways through the system as a collective. This approach can reduce the risk of biased or incomplete analysis from using isolated point-in-time data, and is useful to track groups of youth that are subject to a specific change in policy or practice in order to gauge the effect of that change. For example, assume that 100 youth are detained upon arrest in a given jurisdiction during the month of March. Fifty of them are released back to their families almost immediately, and another 35 remain in detention for less than 30 days, with 15 remaining in detention for 30 days or more. Calculating the average length of stay for the detention population based on a one-day count (a point-in-time analysis) will over-estimate the actual average length of stay of youth in detention, because the count omits many of the youth who had very short stays, and youth with long stays will be over-represented. Cohort analysis can reveal opportunities for significant reduction in bed days, and increased use of alternative programs. For example, youth who leave detention within a day or two should probably be diverted from detention entirely, to a supportive program or otherwise. Case processing times can be improved for youth with very long stays, and considering alternative programs for these youth too can dramatically decrease the total number of bed days overall.
Disaggregation analysis breaks down a given data point by subgroups of youth with unique characteristics, such as race, ethnicity, gender, age, sexual orientation, and geography, to understand differences in practice that may exist based on those characteristics. For example, a jurisdiction might break down data on admissions to detention by a young person’s instant offense type and their race, to determine if Black youth are more likely to be detained than white youth while presenting similar instant offenses.
Statistical summaries when used alone can mask significant disparities in practice related to different subgroups of youth. For example, overall length of stay in detention or placement and daily institutional population maybe falling in the aggregate. But when these variables are disaggregated by race, the data may reveal that the length of stay and average daily population for Black youth have actually risen during the same time period, exacerbating racial disparities in the use of detention or placement. A similar finding might apply to young women, LBGTQ youth, youth from a particular neighborhood or youth charged with certain types of offenses. All data analysis should include disaggregation to assess for systemic disparities and well as system bias.
Take a look at some tips and notable examples of places working to change systems from within harnessing data to drive youth justice transformation.
Skilled, experienced facilitators can organize and run conversations between staff and leadership engaged in youth justice transformation, hold space for dialogue around potential conflicts and help the group arrive at a list of shared values and goals that resonate with staff and leadership and can be used to guide transformation and the culture change process. Useful questions a facilitator might ask to begin and guide conversations include:
Reviewing a list of shared values and goals common to youth justice system transformation can be helpful to assess how much alignment exists between staff and leadership. Lists of shared values from other jurisdictions that have engaged successfully in system transformation efforts can be helpful to draw upon as an example for both staff and leadership. Examples of values can also be found in our chapter on Developing a Shared Vision for Transformation. Below are some potential shared values and goals to work through in conversations.
Youth justice leaders may want to work with facilitators to consider what strategies and forums might be best suited to launch and continue dialogues with staff throughout the agency.Conversations might begin amongst the executive team and supervisors, ensuring that leadership is prepared to both implement and model behavior with their teams in a way that contributes to a healthy, asset-based culture, discussed further below. Leaders might then organize a series of conversations with staff in a variety of forums, ranging from town halls where all staff are convened and are able to ask questions openly, to working group meetings where staff collaborate with smaller numbers of peers representing various positions, to opportunities for more informal conversations such as office hours for staff to share questions or concerns with leadership in a more confidential arena.
Pierce County, Washington has been highlighted numerous times in this Guide, and in other resources about transforming juvenile probation and youth justice work. They have pioneered anOpportunity-Based Probation model with a positive youth justice framework, and not only changed their practice, but started from a place of centering racial justice and equity, collaborating with youth and their families and communities and working with staff at all levels to change the organizational culture.Leadership in Pierce County has noted that staff buy-in is crucial to the success of young people and of the program.
Pierce County has been thoughtful and deliberate about its collaboration with community partners from the very inception of its transformative programming. This process included staff on the front lines being involved in conversations with youth and family about what types of services they wanted and needed, and then finding community-based programs already in existence which met those needs, were aligned with theCounty’s vision and were willing to work with young people.
Culture is well reflected in routine interactions and language used between staff and youth. For example, currently in many jurisdictions it is not uncommon to hear staff refer to young people as “offenders.” This language may be so ingrained in agency culture that there is little to no reflection about how these terms shape perspectives of or interactions with young people.
A core part of the culture change process will be to retrain staff in using person-first language that humanizes young people who are involved with the youth justice system, as well as their families and communities, builds a strengths-based perspective of youth and repositions staff as agents of support and change for young people. Using simple, human terms such as “young people with strengths and challenges,” for example, immediately provides a more human and complex image than distancing and dehumanizing terms such as “delinquents” or “offenders,” and leads staff to consider ways in which they can support young people and invest in their strengths.Of course, language changes alone do not create change; they go hand in hand with all other practical aspects of transformation described in this chapter and in this Guide. Specific training such as anti-racism training may be useful in addition to open, honest and collaborative conversations between staff, leadership, young people, families and representatives of impacted communities to develop language that is supportive of a new paradigm and youth experience of justice.
Typically, systems collect quantitative data—data that can be measured and reported in raw numbers. Less often, systems collect qualitative data—narrative reports of system experiences and functioning. Both types of data are essential to understanding how the system operates and its impact on youth, families and communities. While each youth justice system is different, quantitative information is usually more readily available. Nevertheless, both sources of data will likely require further development as part of a longer-term data development agenda.
Leaders should ask the data analysis team to review the following key performance measures relevant to the system’s dynamics and case flow. Where relevant, the data should include the number and percent for each category and trends over time.
Many systems will need to invest in developing capacity to collect, analyze and interpret data bearing on the results of the system’s interventions. While transformation efforts should not be delayed while capacity is built, leaders should ensure that tracking and reporting on system results is a high priority for the data development agenda.
As they build out the data development agenda, leaders should establish regular, systematic processes for gathering information and feedback from young people who experience the system, their families and members of their communities, as well as youth justice staff, leaders and partners from other youth-serving agencies.These efforts should include formal surveys and group forums as well as interviews and other individualized conversations. This qualitative information is essential to gaining a much clearer picture of the young people the system is serving, their experiences in the system, current system failures and opportunities for transformation that will improve success.
A combination of quantitative and qualitative data can provide system leaders and partners with critical information about gaps in existing services when assessed alongside the needs of young people in the system and their families. This gap analysis is especially important when trying to understand system failures, such as a significant uptick in secure custody or other out-of-home placement, or an increase in the number of youth failing to complete probation or other programs, and how to respond with system changes as part of the transformation process.
This panel discussed why data is so crucial to efforts to transform youth justice systems and the way we think about justice for young people, how data has been used in past transformation efforts, and some key ways that other leaders can use data (and how to do it).
This discussion covered both the why and the how of having the most impacted youth and families leading transformation.
.jpeg)
National Youth Partnership Strategist,
Youth First Initiative
New Mexico Youth Justice Coalition
.jpeg)
Executive Director,
New Jersey Parents Caucus
.jpeg)
Administrator,
Pierce County Juvenile Court
Kathy Wright, Executive Director of New Jersey Parents Caucus, talks about the need to learn about the impact of the system directly from young people who experienced it and to trust them to lead if we are to create change.
Xiuhtecutli (Xiuy) Soto of the New Mexico Youth Justice Coalition speaks about how transforming youth justice begins with having patience with, providing support for, and relating to young people like him.
TJ Bohl, Administrator at Pierce County Juvenile Court, on some of the cultural obstacles inside the system to collaborating with communities, and the need for system leaders to overcome defensiveness to building a path forward together.
Explore the importance of bringing together a diverse group of stakeholders to develop a shared vision for youth justice.
Explore why data is so crucial to efforts to transform youth justice systems and key ways that leaders can use data.
Explore the importance of cultivating true partnership with young people and families at the individual case level and in system transformation.
Explore why transforming the roles of communities and sharing authority and responsibility with them is central to youth justice transformation.
.svg){kind=small}
.png){kind=small}