HELP US END MASS INCARCERATION The Prison Policy Initiative uses research, advocacy, and organizing to dismantle mass incarceration. We’ve been in this movement for 20 years, thanks to individual donors like you.

Can you help us sustain this work?

Thank you,
Peter Wagner, Executive Director
Donate

Beyond the count: A deep dive into state prison populations

By Leah Wang, Wendy Sawyer, Tiana Herring, and Emily Widra   Tweet this
April 2022
Press release

We know how many people are in state prisons, but what do we really know about who they are or how they ended up there? Over 1 million people are confined in state prisons nationwide, primarily serving sentences of anywhere from a year to life.1 But the walls and restrictions that keep these individuals out of public life also keep them out of the public eye: most of what we know about people in prison comes from the prison system itself. But our analysis of a unique, large-scale survey of incarcerated people provides a richer picture of just who is locked up in state prisons.

From the survey data, we gain a deeper understanding of how mass incarceration has been used to warehouse people with marginalized identities and those struggling with poverty, substance use disorders, and housing insecurity, among other serious problems. Incarcerated people are a diverse cross-section of society whose disadvantages and unmet needs often begin early in life, and persist throughout their often lifelong involvement with the criminal legal system.

This report covers a lot of ground, so we’ve divided it into sections that can be accessed directly here:

  • Demographics: Race, ethnicity, age, gender identity, and sexual orientation 
  • Employment and housing: Incarcerated people were on unstable footing long before prison 
  • Arrested early and often: Age at first arrest, youth confinement, and prior incarceration 
  • Disadvantage dating back to childhood: Family, housing, poverty, and education in youth 
  • Drug use: An extremely common factor leading up to incarceration 
  • Methodology: Details about the data and our analysis  
  • Appendix tables: Explore the data yourself  
 

About the unique data used in this report

This report offers a detailed view of the people in state prisons nationwide, using the most recent self-reported, nationally representative data available, the Bureau of Justice Statistics’ 2016 Survey of Prison Inmates. It may seem like 2016 data about state prison populations would be outdated by now, but because this survey takes a much deeper look into pre-incarceration life experiences — and the data were only released about one year ago — the 2016 data remain essential to understanding incarceration today.

How is this data different from other sources?

There are a few important differences between the Bureau of Justice Statistics’ Survey of Prison Inmates and other government data collections, so some of the statistics in this report will not perfectly match those derived from other data sources…

There are a few important differences between the Bureau of Justice Statistics’ Survey of Prison Inmates and other government data collections, so some of the statistics in this report will not perfectly match those derived from other data sources. The most important differences are who was included, when the data were collected, and how the data were collected.

First, this survey included in its sample people who were physically located in state prisons, often referred to as the “custody” population. This is different from many other data collections and publications that look at the “jurisdictional” population, or only those people who are legally under a given agency’s authority. Our Mass Incarceration: The Whole Pie report, for example, uses data that counts confined people by jurisdiction, not custody. Many people may be surprised that some prisons and jails have people in their custody that aren’t actually under their jurisdiction, but this is relatively common. About 2.3% of the people represented in these survey data were not actually held under state jurisdiction but rather were held for a local jail system, the federal prison system, or another authority.

Second, although the data from this survey were published at the very end of 2020, they were collected in 2016. Other reports, including our Whole Pie report, use more recent population data.

Finally, these data were collected using a survey of a nationally-representative sample of people in prison, as opposed to using administrative (prison system) data that includes everyone in custody. This difference is a tradeoff, of course, because sample sizes limit precision when it comes to population estimates (for example, about 20,000 people in state prisons were surveyed). But self-reported data are often more accurate, especially when it comes to race, ethnicity, gender, sexual orientation, and other aspects of social identity. And of course, surveys allow researchers to ask detailed questions that can’t be answered with the regularly-collected administrative data, such as questions about childhood experiences or the reasons someone chooses to participate in a program. This survey centers incarcerated people in their own data, offering a valuable, if unfortunately rare, source of first-hand information.

While the Bureau of Justice Statistics, the federal agency that collected the data, has published its own series of reports based on its findings, we designed this report to include many details overlooked in the government’s reports. We surface important details about identity, life experiences, health, and more. Where we can, we compare the data to those collected in decades prior or to measures of the general U.S. population, to get a sense of how prison populations compare and how state policies have changed, or remained stagnant.

Readers may already be asking how the data look by state, or when we’ll have more current data. Unfortunately, these survey data are not available state-by-state, and we estimate it may be another several years before the survey is fielded again (this survey has not been on a regular publication schedule since 2004, with 12 years between the two most recent data collections). But the information presented here is a powerful supplement to other state-specific incarceration data, creating a strong foundation for advocacy and sound policymaking.

   

Demographics: Race, ethnicity, age, gender identity, and sexual orientation

The question of who is locked up in state prisons demands that we look at as many identities as possible, from race and ethnicity, to age and sexual orientation, to sex and gender identity. The Survey of Prison Inmates dataset provides a range of self-reported demographics, data that confirm what many people have said before: that the U.S. criminal justice system targets the least powerful people in society.

We find that people in state prisons disproportionately report marginalized identities related to race, gender, sexual orientation, class, and more.2 These demographic details matter, beyond what they tell us about the identities of criminalized people. Too many examples exist showing that prisons are often neither safe nor equipped to care for the women, people of color, young and older adults, lesbian, gay, bisexual, and transgender (LGBT) people, and others in their care, yet these groups are too often ignored in prison policy and practice.

Race and ethnicity

Nationwide, state prison populations look very different than the general U.S. population:

chart comparing the racial composition of state prisons nationwide to the U.S. population overall in 2016

In state prisons, 34% of people identify as Black and 32% identify as white, groups roughly equal in size. Another 21% identify as Hispanic, 11% identify as two or more races, about 1.4% are American Indian or Alaska Native, and 0.9% were Asian, Native Hawaiian or other Pacific Islander. This is a marked difference from the U.S. population: every race and ethnicity is overrepresented in prisons, except for white and Asian people.3 Comparing women in state prisons to the U.S. female population, the proportion of Hispanic women in state prisons (16%) is roughly the same as in the general population (17%). Still, women of color add up to a slight majority in state prisons (51%, compared to 47% white women) — which is, at least in part, the natural result of racial disparities in the policing of women.

Racial disparity throughout the criminal legal system is not new, of course. But the fact that these disparities persist means that policymakers should always evaluate how proposed criminal justice reforms will — or won’t — advance racial justice. Any policy designed to tackle mass incarceration is incomplete without an acknowledgement of the deep racial disparities that lock up disproportionate numbers of people of color.

Age

Simply by looking at age alone, it’s clear that prison is the wrong place for so many who are there:

chart showing 128,500 people ages 18 to 24 & 159,200 people  55 & older were in state prisons in 2016, & the share of older people in prison has grown from 3 to 13 % since 1986

The number of elderly people in prison is particularly concerning. The proportion of people age 55 and older in prisons has grown over the past few decades, thanks to draconian sentencing laws locking people away for extremely long prison terms. This “graying” of prisons persists despite the low risk of recidivism and high costs of care related to older incarcerated people. (While people who are 55 are not widely considered “elders” outside of prisons, this is not an exaggeration in the prison context: incarceration is known to accelerate biological aging and shorten life expectancy. Advocates, lawmakers, and the Bureau of Justice Statistics itself have used 55 and older as a standard “elder” group.)

Indeed, more than 1 in 8 people in state prisons nationwide (13%) are over the age of 55, and the average age has grown from 35 to 39 years old since 2004. While a smaller percentage of women are 55 or older (7.4%) compared to men (13%), we found only slight variation in older populations along racial and ethnic lines. Reforms targeted toward older people — particularly those that would shorten excessively long sentences or release more people on parole — would make a huge dent in state prison populations (and budgets), without compromising public safety.

chart showing almost two-thirds of people age 18 to 24 in prison are Black or Hispanic

Another 1 in 10 people in state prison (10%) are on the other end of the age range — they’re between 18 and 24 years old. They’re considered “adults” by courts, though a growing body of research shows that this life stage is not full, mature adulthood. And age maps onto other demographic characteristics in ways that demonstrate greater risk for certain young people. For example, over half (63%) of the “emerging adults” in state prison (ages 18-24) are Black or Hispanic — an even higher percentage than in the total state prison population. The ongoing overincarceration of young Black and Hispanic people is not because they are predisposed to commit more crime; rather, policies and social norms have stacked the deck against communities of color for decades, denying the education and opportunity that often prevents crime.

The same share of people in state prison that identify as lesbian, gay, or bisexual fall into the 18 to 24 age group (10%), but for these young adults, their sexual orientation compounds their risk of abuse, inside and outside of prison. According to a previous Bureau of Justice Statistics report, non-heterosexual people ages 18 to 24 reported higher rates of sexual victimization compared to heterosexual young people while incarcerated. Despite the passage of the federal Prison Rape Elimination Act (PREA), true prevention of sexual violence in prison has been a weak effort; prison is not, and has never been, a safe place, particularly for lesbian, gay, bisexual, and transgender individuals.

Gender identity and sexual orientation

The majority (93%) of all people in state prisons nationwide are male, and the remaining 7% are female according to the Bureau of Justice Statistics’ binary, “administrative” sex variable, which categorizes everyone in a “women’s” prison as female and everyone in a “men’s” prison as male. But for the first time in Bureau of Justice Statistics data collection, survey respondents were also able to report their own gender identity, revealing that about 0.2% of people in state prisons identify as transgender, and a smaller portion identify as neither male, female, nor transgender. While 0.2% may be an underestimate due to the sensitive nature of self-reporting or sampling error, this is the first official data to come out of the Bureau of Justice Statistics to help us understand the experiences and unmet needs of transgender people in prisons, which we discuss in our analysis What the Survey of Prison Inmates tells us about trans people in state prison. (It’s worth noting that in this report, Beyond the Count, when we refer to “men” and “women,” we’re using self-reported gender identity — in other words, men and women who do not identify as transgender).

Among people in state prisons, 4.1% identify as lesbian, gay, bisexual (LGB), or “something else” that is not heterosexual. This is a greater share of the population than the estimated 3.5% of the total U.S. adult population who identify as LGB, and points to alarming rates of incarceration among LGB people. The overcriminalization of LGB and transgender people begins before and persists after a prison sentence, as we know that LGB people are arrested at much higher rates, face longer sentences, and are under correctional control at higher rates compared to straight people.

In line with findings from past research, incarcerated women are much more likely to identify as lesbian or bisexual: Over 22% of women in state prisons report an LGB identity, compared to about 2.5% of men.4 The small number of transgender people in state prisons who participated in the survey also overwhelmingly identify as LGB or something else other than straight.5 White people and people of two or more races are twice as likely to identify as LGB (5.1% and 5.8%, respectively) compared to other racial and ethnic groups, but state prisons confine LGB people of all races and ethnicities.

Recognizing that significant numbers of LGB and transgender people exist in state prisons, it’s imperative to understand how discrimination and stigma lead to their criminalization, and that those who end up behind bars will need extensive policy changes to ensure their safety. Prior research tells us that LGB people are more likely than straight people to come from low-income households (especially LGB people of color), to have a history of sexual victimization, to report poorer mental health (particularly women), and to spend time in segregation or solitary confinement while locked up (again, particularly women). These experiences inside and outside of prison point to a dire need for social supports and an end to the stigmatization of LGB and transgender identities.

   

Employment and housing: Incarcerated people were on unstable footing long before prison

The demographic data alone point to inequities that put some people at greater risk of incarceration than others. But the Survey of Prison Inmates also asked respondents about their socioeconomic backgrounds, and the data tell a more nuanced story about how people end up in state prison — and suggest how policy can stem the tides of mass incarceration by creating equity and opportunity before someone is ever arrested. The survey asked several questions about respondents’ circumstances in the 30 days and 12 months prior to the arrest that led to their current incarceration. The results show that when people most need help, unjust societal forces ensnare them in the criminal legal system instead.

First, we find that many people were struggling to make ends meet before being arrested and sent to prison:

  • Four in 10 (39%) people in state prison did not have a job (i.e., were “jobless”) in the month before their arrest. Black people had the highest jobless rates6 of any race or ethnicity (at 46%), and women were slightly more likely not to have a job (53%) than to have one (47%) in the month before arrest.
  • Of those who weren’t working before their arrest, 4 in 10 (38%) were looking for work, a requirement for being counted as “unemployed.” That’s roughly equivalent to a 15% unemployment rate, an astonishingly high rate that we’ve only seen in the national population twice in recorded history — during the Great Depression and during the early pandemic in 2020.7
  • Among those who did work before their arrest, 1 in 5 (20%) were actually working two or more jobs, compared to just 1 in 20 (5%) of the U.S. population who held more than one job in 2016.
  • Whether they worked or not, 1 in 5 (20%) people in state prisons relied on income from other, non-job sources like public assistance, Social Security, a retirement fund, unemployment benefits, or disability benefits in the 30 days before their arrest and incarceration.8
chart comparing unemployment and homelessness among people in state prisons and the total U.S. population in 2016

In a previous analysis of the 2004 Survey of Prison Inmates data, we found that before they were incarcerated, working-age people in prison were making less than $20,000 per year on average, which was over 40% less than similar non-incarcerated people. Incarcerated women of all races (but especially Black and Hispanic women) and incarcerated Black men were earning much less than this average. Unfortunately, because of data collection issues from the relevant questions, we did not update our analysis of the pre-incarceration incomes of people in prison.9 However, our analysis does show that when the incomes of people in prison dried up upon arrest or incarceration, it had a devastating impact: Of those who lived with children before their arrest, the vast majority (83%) provided at least half of their household’s financial support.10

Policymakers often frame housing insecurity and homelessness as a reentry issue, yet for many people, unstable housing was a fact of life before prison, and this precarity may have contributed to their later incarceration. The survey data show that 4.9% of people in prison were living in a homeless shelter or on the streets in the 30 days before arrest, another 3.4% were living in transitional housing or a residential treatment facility, and another 3.7% were in prison or jail in the 30 days before their arrest. Finally, 6.5% of people in state prisons were living in a hotel, motel, or rooming house; altogether, more than one-fifth (22%) were experiencing housing instability or homelessness shortly before they were incarcerated.11 These figures may sound low, but they’re wildly out of proportion with how the general U.S. population experiences housing insecurity. It’s clear that investing in the basic building blocks of communities and individuals’ lives — such as a “housing first” approach, and ensuring employment opportunities for those who already have conviction records — will help to prevent the contact with law enforcement that too often, for the most disempowered, leads to prison.

   

Arrested early and often: Age at first arrest, youth confinement, and prior incarceration

Our analysis shows that many people in prison have a personal history of household instability, poverty, and overall lack of opportunity in their home communities, unjustly setting them up for a long-term relationship with the criminal legal system. It’s no secret that law enforcement agencies have historically targeted poorer communities and criminalized many survival behaviors, initiating cycles of arrest, incarceration, and supervision. The Survey of Prison Inmates data demonstrate that our cultural and policy responses to social inequities — not “criminals” — are to blame for the size of our incarcerated population. Solutions then, too, are located in policies that affect communities, not just individuals who end up in prison.

chart showing the percent of people in prison in 2016 who were first arrested at age 18 or younger, by gender and race A shocking 38% of people in state prisons in 2016 were first arrested before their 16th birthdays. View an alternate version of this chart for people first arrested at 15 or younger here.

We find that people in prison tend to have too much experience with law enforcement, too early in life:

  • More than 2 in 3 people in state prison (68%) are first arrested before the age of 19; shockingly, more than 1 in 3 people in state prison (38%) are first arrested before they turn 16. This is an extraordinary number of youth being arrested, a traumatizing experience that also increases the risk of re-arrest.
  • Women in state prisons experienced their first arrests later in life than men: 16% of women in state prison (but fewer than 7% of men) had a first arrest between age 26 and 35. Another 1 in 10 (10%) women were first arrested over the age of 35, compared to less than 6% of men. These findings align with the concept of “gendered pathways” to prison, which explains that women tend to become involved in unlawful behavior in different ways than men.12
  • Most Black people in prison were first arrested as youth, with over 3 in 4 (76%) reporting a first arrest before age 19. Black people and those with “Other” racial identities in state prison were also the least likely to experience their first arrest after age 22, showing the magnitude of young people of color being funneled into jails and prisons instead of being diverted to more supportive interventions.
  • Whatever their age when first arrested, the average person in state prison has been arrested 9.3 times. This level of contact with law enforcement does little to help people struggling with substance use, poverty and homelessness, or mental health issues, and may actually exacerbate these problems.

A full one-third (34%) of people in state prisons served time in a juvenile facility. This amounts to over 425,000 people in state prisons who were previously locked up as children or adolescents, sometimes for behaviors that are not crimes at all. The survey data don’t tell us how long people spent in youth detention, but confinement is inherently traumatizing and the transitions in and out of confinement inevitably destabilizing.

The interaction of homelessness, youth confinement, and discrimination is a good example of why connecting the dots between these disparate data points is imperative. According to a Center for American Progress report, homelessness is the greatest predictor of involvement with the juvenile justice system. And the survey data we analyzed show that about 1 in 6 (16%) people in state prisons who was first arrested at age 18 or younger also reported being homeless before turning 18. But this correlation is even more dramatic among lesbian, gay, and bisexual people in prison: about a quarter (24%) of LGB people in prison who were first arrested at 18 or younger experienced homelessness before age 18. This is largely explained by the fact that homeless youth disproportionately identify as lesbian, gay, bisexual (LGB), or transgender (40% do). Once again, where marginalized statuses overlap is exactly where we find people at high risk of entering the juvenile justice or adult criminal legal systems.

Unsurprisingly, almost all people in state prisons (94%) spend time in a local jail before being sentenced and transferred to a state prison. Whether due to unaffordable bail, overburdened courts and public defenders, or “holds” for probation, parole, or other authorities, individuals spend anywhere from days to years in jail before going to prison.13 Given the frequency of prior arrests, many people have been to jail before: nearly 3 out of every 5 people in state prison have already served a jail sentence of some sort. Among racial and ethnic groups, we found that American Indian or Alaska Native people in prison may cycle in and out of jails most frequently, with 8.1% reporting more than 10 jail stints before their current incarceration. It may seem like there would be a natural limit to cycling in and out of jail, but many people report dozens upon dozens of jail stints, an undeniably disruptive, costly, and ineffectual cycle.14

Prisons vs. jails: What's the difference?

They've got a lot in common, but they're far from the same thing…

Prisons are facilities under state or federal control where people who have been convicted (usually of felonies) go to serve their sentences. Jails are city- or county-run facilities where a majority of people locked up are there awaiting trial (in other words, still legally innocent), many because they can’t afford to post bail. To make things a little more complicated, some people do serve their sentences in local jails, either because their sentences are short or because the jail is renting space to the state prison system.

We find that community supervision also plays a central role in the personal timelines of people in state prisons. More than half of those in prison had been placed on probation once or twice before their current incarceration. Previous research tells us that violations of probation and parole contribute significantly to state prison populations, accounting for a whopping 45% of admissions every year. This is supported by our analysis of the survey data, which finds nearly 40% of people in prison were actively on probation or parole at the time of their arrest.

   

Disadvantage dating back to childhood: housing, poverty, and education

Going back further in time, many people in prison faced serious disadvantages early on that put them at heightened risk of involvement in the criminal legal system. Our analysis of the Survey of Prison Inmates data indicates strong connections between adverse childhood conditions related to housing, poverty, education, family history of incarceration and other disruptions, and eventual incarceration in a state prison.

chart summarizing the family, housing, economic, and educational challenges faced by people in prison in their youth and discussed in this section

In a previous section, we discussed the prevalence of homelessness among people in prison right before their arrest and incarceration. But many incarcerated people experienced homelessness much earlier: About 1 in 10 people in state prison was homeless at some point before turning 18.15 An even greater share experienced some combination of poverty16 and/or family disruption. Almost 1 in 5 people in state prison (19%) report living in public housing as youth, and 2 in 5 (42%) report that their family received public assistance (welfare) at some point before they turned 18. Further, 1 in 5 (18%) report spending time in foster care, and 1 in 3 (33%) report that a parent was incarcerated at some point in their lives.17 This is an overwhelming figure that points to devastating cycles of intergenerational disadvantage and incarceration. These disadvantages hit Black people in prison hardest: As youth, Black people were three times as likely to have lived in public housing as white people in prison were (31% vs. 9%), more than half (52%) report that their family received public assistance during their childhood, and 37% have a history of parental incarceration.

According to our analysis, the younger someone is when they’re arrested for the first time, the more likely they are to have also been homeless, in public housing, in foster care, or for their family to have received public assistance before they turned 18.18 Similarly, people who were confined in a juvenile facility at least once were more likely to report these experiences.

Before going to prison, less than half (38%) of people in state prisons complete high school, or education beyond high school — while almost 90% of adults in the U.S. are at least high school graduates. The average person in state prison, who is 39 years old, has a 10th grade education upon admission to prison. These shockingly low rates of educational attainment may be linked to how frequently people in prison report confinement as youth; by definition, incarceration interrupts regular schooling and destabilizes a young person’s life. School-based discipline, and the widely recognized “school-to-prison pipeline,” has been linked to experience in the juvenile justice system, especially for Black students.

chart showing educational attainment before admission to prison, by race, and that over two-thirds of Black & Hispanic people in prison have less than a high school diploma

Compared to the average, more women (47%) but fewer Black and Hispanic people (21% and 20%, respectively) entered prison with a high school education; just 4% of people in prison enter with a college degree. Despite the known return on investment in prison education, large swaths of the incarcerated population have been denied meaningful access to college due to the ban on federal Pell grants in 1994 (and ensuing bans on similar, state-based grants).19 At a time when the prison population was exploding, college became out of reach for most incarcerated people, whether they had a chance at release or not.

The low educational attainment of people in state prisons makes investments in education, and in stopping the school-to-prison pipeline, an urgent and indisputable priority. While education in prison is a valuable offering and promising area of reform, it should coexist with universal access to high-quality public education and other youth programming that prevents criminal legal system involvement.

   

Drug use: An extremely common factor leading up to state prison

The Survey of Prison Inmates data and previous studies show clear connections between disadvantages early in life, substance use disorders, and unlawful behaviors that land someone in juvenile detention, jail, and prison. Our analysis of the survey data allows us to look closely at drug use as it directly relates to someone’s conviction, and as it maps on to their demographic characteristics, creating an even more vivid image of hardship and unmet need before prison. Specifically, we find that:

  • A majority (65%) of people in state prisons report using some illicit substance in the 30 days prior to their arrest,20 whereas in the general U.S. population, only about 11% of people age 18 and older have used illicit drugs in the past month. Moreover, 39% of people in state prisons report using drugs at the time of their offense, a statistic that underscores the reality of substance use and misuse in the circumstances leading up to arrest and eventual imprisonment.
  • Half (49%) of people in state prisons met the criteria for a substance use disorder (SUD), including about 59% of women and 48% of men.21 And almost half (46%) of people in state prisons (presumably, nearly all of those with a SUD diagnosis) report undergoing SUD treatment at some point before their most recent admission to prison. American Indian or Alaska Native people in prison have the highest rates of SUD (62%) and reported SUD treatment (62%), but white and multiracial people also had SUD and treatment rates over 50%, whereas other racial and ethnic groups did not.
  • Women are more likely to have a history of a substance use disorder (SUD) (59% versus 48%) and/or SUD treatment (55% versus 46%) compared to men, though they are not always more likely to have a SUD; there are gendered differences in misuse that depend on the substance. Remarkably, the same percentage of imprisoned women reported SUD treatment back in 1991, suggesting that we haven’t figured out a better, more humane way than incarceration to respond to problems related to substance use in the intervening 25 years.
chart showing that people in state prisons report much higher rates of drug use before their arrest - including 39% who used drugs during the offense itself

The survey data don’t distinguish between court-ordered and voluntary treatment, but substance use disorder (SUD) treatment — as handed down by the criminal legal system — may not be working as intended. This is especially true when treatment is involuntary or coerced, when people are re-incarcerated for positive drug tests, or when they’re disqualified for diversion programs based on their history of substance use. Instead, as the Drug Policy Alliance suggests, SUD treatment should always be voluntary, involve clinicians rather than law enforcement, incorporate positive incentives, and use gold-standard medication and other services.

The high rate of substance use disorders among people in prison points to a serious nationwide misunderstanding: Our society still responds to substance use (and related crimes) as an individual failure requiring punishment, rather than as a public health problem, and it’s not working. Many people needing care are instead arrested and jailed over and over until finally, one event lands them in state prison, a practice with high moral and fiscal costs. Instead, states should decriminalize or reclassify low-level offenses that, for many people with substance abuse disorders, are associated with survival and health.

   

Conclusions

Most correctional data collected by the government reduces people in prison to beds occupied, or the crimes they’ve committed. This is dehumanizing, and makes it easier to dismiss the needs of incarcerated people, their dignity, and the reality that their outcomes in life impact everyone. On the other hand, the data derived from the Survey of Prison Inmates — reported by incarcerated people themselves — look beyond the headcount and help to connect these numbers to the society that fashioned them, revealing how many come from, or have been relegated to, the margins of society. The data illustrate how people with intersecting identities and disadvantages are targeted by policies that criminalize the behaviors they engage in as they try to survive and cope under these conditions. Given that so many of the stories of people in prison are shaped by social and economic policies that have eroded safety nets and neglected entire communities, the most impactful policy changes to reduce incarceration and enhance safety will also begin outside of prison walls, addressing communities rather than individuals.

   

Methodology

The Bureau of Justice Statistics’ 2016 Survey of Prison Inmates was conducted through in-person interviews of people in state and federal prisons, and provides self-reported data on a wide range of characteristics and subjects. The target survey population included people ages 18 and older who were currently in the custody of a state or federal prison, whether or not they were sentenced to a period of incarceration. A total of 28,848 people participated in the survey (20,064 from 306 state prisons and 4,784 from 58 federal prisons). For more information, see the Bureau of Justice Statistics’ Methodology: Survey of Prison Inmates, 2016.

The state and federal survey data were made available, in separate files, to researchers in late December 2020. We elected to focus on the state prison data only, since it’s primarily state policy that drives mass incarceration. Additionally, the state and federal prison populations differ in substantial ways that make it worthwhile to look at them separately. For more information about how state, federal, and local systems of confinement differ, see our report Mass Incarceration: The Whole Pie.

Using the programming language R and the survey’s weighted data, we generated crosstabs of all variables of interest, recoding where necessary to derive meaning (for example, we created age categories rather than comparing people of specific ages). As a survey rather than an administrative dataset, the Survey of Prison Inmates provides data self-reported from incarcerated people in addition to some administrative data. We opted to use self-reported data whenever possible. Specifically:

  • We always analyzed data using self-reported gender identity, rather than the binary sex variable derived from administrative records, to more accurately account for people who identify as transgender or something else. For the most part however, in this report, we did not generalize or discuss the survey results for transgender people, due to the very small sample size: only 29 people surveyed identified as transgender. Because we believe it’s critical to highlight the experiences of transgender people in the criminal legal system, we present many of these results in our briefing What the Survey of Prison Inmates tells us about trans people in state prison.
  • We analyzed race data using one combined race/ethnicity variable from the survey that categorizes anyone identifying as Hispanic separately from other racial identities, although we recognize that many people identify as both Hispanic and another race, such as white, Black, or American Indian or Alaska Native. This methodology is consistent with how the Bureau of Justice Statistics reports racial and ethnic data.

Finally, we encourage other researchers to explore this rich and unique dataset. There is much more potential for analysis of these data, beyond what we and the Bureau of Justice Statistics have published.

   

Acknowledgements

The authors wish to thank Alexi Jones for her significant contributions to the early data analysis for this report, and our colleagues Naila Awan, Wanda Bertram, Peter Wagner, and Mike Wessler for their helpful feedback. The Prison Policy Initiative also thanks Public Welfare Foundation for supporting our analysis of the Survey of Prison Inmates, as well as our broader work strengthening the justice movement by filling key data and messaging gaps. Lastly, we thank our individual donors, who give us the resources and the flexibility to quickly turn our insights into new movement resources.

   

About the authors

Leah Wang is a Research Analyst at the Prison Policy Initiative. Her most recent major report is Building exits off the highway to mass incarceration, a guide to understanding criminal justice “diversion” programs designed for local policymakers and the general public. Her other recent work includes reports on how incarceration affects people’s experiences in the job market and the importance of family contact for incarcerated people.

Wendy Sawyer is the Research Director at the Prison Policy Initiative. She is co-author of the organization’s most well-known report, Mass Incarceration: The Whole Pie; as well as Youth Confinement: The Whole Pie and Arrest, Release, Repeat: How police and jails are misused to respond to social problems. In addition to these reports, Wendy frequently contributes briefings on recent data releases, academic research, women’s incarceration, pretrial detention, probation, and more.

Tiana Herring is a Research Associate at the Prison Policy Initiative. In addition to several short reports on issues including felony theft thresholds and money bail, Tiana is co-author of the organization’s recent reports States of Incarceration: The Global Context 2021, which compares U.S. states to countries around the world on incarceration rates; and States of Emergency: The Failure of Prison System Responses to COVID-19.

Emily Widra is the Senior Research Analyst at the Prison Policy Initiative. She is the organization’s expert on health issues behind bars; over the past year, she has curated the Prison Policy Initiative’s virus response page, which tracks the policy changes that prisons have made in response to the COVID-19 pandemic. She is also co-author of the organization’s recent major report States of Incarceration: The Global Context 2021.

   

About the Prison Policy Initiative

The non-profit, non-partisan Prison Policy Initiative was founded in 2001 to expose the broader harm of mass criminalization and spark advocacy campaigns to create a more just society. The organization is known for its visual breakdown of mass incarceration in the U.S., as well as its data-rich analyses of how U.S. states vary in their use of punishment. Alongside reports like these, the organization leads the nation’s fight to keep the prison system from exerting undue influence on the political process (a.k.a. prison gerrymandering) and plays a leading role in protecting the families of incarcerated people from the predatory prison and jail telephone industry and the video calling industry.

   

Footnotes

  1. A small number of people in state prisons (around 2%) are unsentenced — awaiting trial, sentencing, or a parole or probation revocation hearing. Much of the data we analyzed includes these unsentenced people in the total population. Although we can’t be certain, because the survey data can’t be broken down by state, it’s likely that just a handful of states hold most of this unsentenced population: States with “unified” (combined) prison and jail systems, which include Alaska, Connecticut, Delaware, Hawaii, Rhode Island, and Vermont.  ↩

  2. Importantly, many people in prison exist at the intersection of a number of oppressed identities, which often compounds the effects of discrimination and marginalization, and puts these individuals at greater risk of criminalization and incarceration. For example, lesbian, gay, and bisexual people are overrepresented in prisons, compared to their share of the overall U.S. population, which suggests that discrimination based on sexual orientation may — directly or indirectly — impact the risk of arrest and incarceration. But, as discussed in a later section, this disparity is much more dramatic among women who identify as lesbian or bisexual. This difference suggests that these women face discrimination related to the combination of their gender and sexual orientation that impacts the risk of arrest and incarceration more than discrimination based on sexual orientation alone.  ↩

  3. In 2016, the total U.S. population was 62% white (Not Hispanic), 12.3% Black (Not Hispanic), 17.3% Hispanic, 5.2% Asian (Not Hispanic), 2.3% two or more races (Not Hispanic), 0.7% American Indian or Alaska Native, and 0.2% Native Hawaiian or other Pacific Islander (the Census Bureau collects separate data on Asian and Native Hawaiian/Pacific Islander people). Of course, these proportions look different at the state level — both in the incarcerated and general population — and there are some states where Asian and Native Hawaiian/Pacific Islander people actually are overrepresented in prisons, such as in Hawaii.  ↩

  4. The linked article, which describes “sexual minorities” within U.S. incarcerated populations based on the Bureau of Justice Statistics’ 2011-2012 National Inmate Survey, analyzed over 80,000 respondents’ answers to questions about sexual orientation. Compared to the Survey of Prison Inmates, that survey found a higher prevalence of LGB people: 5.5% of men, and 33% of women in prisons (both state and federal), identifying as gay, lesbian or bisexual. One possible reason these numbers are higher is that they include people who identify as transgender, which was not a separate option for respondents in the National Inmate Survey.  ↩

  5. The Survey of Prison Inmates sample included just 29 people in state prisons who identify as transgender, so we caution against generalizing the findings about this population too broadly. However, 17 of the 29 trans respondents identified as lesbian, gay, bisexual or something else other than straight. See our separate analysis of this small sample of imprisoned trans people for more details.  ↩

  6. “Joblessness” and “unemployment” are distinct terms, although they are closely related. Anyone who reports not having a job may be counted as “jobless,” but only people who are both jobless and are actively part of the labor force are counted in “unemployment” measures. Many people without jobs are not actively in the labor force, such as people who are in school, retired, or unable to work, and are therefore not considered “unemployed.” People who are jobless but are available to work and seeking work, on the other hand, are considered “unemployed.”  ↩

  7. We’ve also analyzed past survey data from people who were formerly incarcerated in state prisons, and found that they had a similarly shocking 27% unemployment rate, higher than any other recorded unemployment rate in U.S. history. And overall, joblessness (whether someone was seeking work or not) might be even more pervasive than the Survey of Prison Inmates data show: A recent Bureau of Justice Statistics study revealed that in a group of over 50,000 people who served time in federal prisons, over 60% were jobless three years prior to their admission to prison.  ↩

  8. This figure might even be higher were it not for a wide range of exclusionary policies that disqualify people with felony convictions from governmental assistance programs (though the Survey of Prison Inmates data don’t tell us how many people have previous felony convictions). Some states have opted out of these federal bans on benefits, but at least half of states have upheld or modified these bans (such as for SNAP, formerly known as food stamps), making it extremely difficult for formerly incarcerated people to access critical financial relief.  ↩

  9. We decided that the income data from the 2016 Survey of Prison Inmates was not generalizable because of the “skip patterns” that excluded many respondents — in other words, not everyone was even asked the income-related questions because of how the survey was designed (and this design represents changes since the last survey, in 2004). Only those who reported a source of income in the 30 days before arrest were asked to estimate that income in one number, and those responses were not made available for public data analysis. If the respondent did not know or refused to provide one number for their monthly income, only then were they asked to choose from a range, like “Less than $200,” or “$500 to $999.” Because so few people were asked this income range question — about 6% of survey respondents were asked, and fewer than 5% provided a usable response — we concluded that the data are not representative enough to discuss here. We may revisit these data in a later analysis where we can provide more context.  ↩

  10. Of those living without children before their arrest, we found that 66% reported that their income represented at least half of their household’s finances.  ↩

  11. We prefer to expand the concept of housing insecurity beyond homelessness, and include these measures of “marginal housing.” For an analysis of housing insecurity and risk factors for homelessness among formerly incarcerated people, see our groundbreaking report Nowhere to Go.  ↩

  12. For example, research on “gendered pathways” points to the involvement of women in unlawful activities through problematic romantic relationships and/or engagement in criminalized survival behaviors, like sex work, shoplifting, and selling drugs. Incarcerated women are also extremely likely to report a history of abuse and trauma.  ↩

  13. According to the survey data, about 27% of people in state prison spent less than a month in jail before coming to prison; 20% spent 1 to 3 months in jail; 25% spent 4 to 6 months in jail; and the remaining 28% were in jail for 7 months or more before transferring to prison to serve the remainder of their carceral sentence.  ↩

  14. About half of one percent (0.6%) of people in state prison report serving over 30 separate jail sentences. This may seem like a small number, but it equates to about 7,000 people and hundreds of thousands of jail stints. For more information about the repeated cycling of people in and out of jails, see our report Arrest, Release, Repeat.  ↩

  15. For a significant number of people in prison, homelessness has been an ongoing problem. About one-third of people in state prisons who were homeless before turning 18 also reported being homeless in the 12 months before the arrest that led to their current incarceration.  ↩

  16. The Survey of Prison Inmates questionnaire did not ask respondents about their family or household income during their childhood. As a proxy for low family income or poverty, we rely on the survey’s questions related to social safety nets like living in public housing and receiving governmental assistance.  ↩

  17. Additionally, 3 in 5 people in state prison reported having any immediate family member incarcerated at some point in their lives. However, the survey did not ask how old respondents were when family members were incarcerated, so parental incarceration did not necessarily occur before they turned 18.  ↩

  18. People whose first arrest was when they were under 10 or between ages 11-15 were more likely than those first arrested at any older age to have reported these disadvantages.  ↩

  19. By mid-2023, the ban on access to federal Pell grants will be fully lifted, thanks to an act of Congress at the end of 2020. Meanwhile, as of the drafting of this report, New York is poised to lift its own ban that blocked incarcerated students’ access to the Tuition Assistance Program (TAP).  ↩

  20. This figure doesn’t include alcohol; alcohol use in the 30 days before arrest was not an available datapoint. For more detail on drug and alcohol use and disorders, see the Bureau of Justice Statistics report Alcohol and Drug Use and Treatment Reported by Prisoners, based on the same Survey of Prison Inmates data.  ↩

  21. These statistics were reported by the Bureau of Justice Statistics, and therefore “men” and “women” here refer to the binary administrative sex variable, and not self-reported gender identity.  ↩

Read the footnotes



Appendices

Appendix A: Demographics

Race and ethnicity

Gender identity (self‑reported)
Total state prison population Male Female
White (Not Hispanic) 31.4% 30.2% 47.5%
Black (Not Hispanic) 33.3% 34.4% 18.9%
Hispanic 20.4% 20.7% 15.9%
American Indian/Alaska Native (Not Hispanic) 1.4% 1.4% 1.7%
Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) 0.9% 0.9% 0.7%
Two or more races (Not Hispanic) 11.1% 10.8% 13.9%
Other (Not Hispanic) 0.1% 0.1% 0.0%

Gender Identity

* Due to the survey's small number of transgender respondents, we analyzed these data separately in What the Survey of Prison Inmates tells us about trans people in state prison.
Race/ethnicity (self‑reported)
Total state prison population White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Male 92.7% 89.1% 95.8% 94.1% 91.3% 93.7% 90.7% 100.0%
Female 7.1% 10.7% 4.0% 5.5% 8.6% 5.5% 8.9% 0.0%
Transgender* 0.2%


Sexual orientation

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Straight 95.4% 97.1% 76.1% 94.4% 97.2% 95.3% 94.7% 95.5% 93.0% 100.0%
Lesbian or Gay 1.4% 0.9% 6.9% 1.8% 1.1% 1.1% 1.9% 1.4% 1.9% 0.0%
Bisexual 2.5% 1.4% 15.6% 3.2% 1.3% 2.4% 2.9% 1.6% 4.0% 0.0%
Something  else 0.2% 0.2% 0.3% 0.2% 0.2% 0.2% 0.0% 0.7% 0.4% 0.0%

Age group

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
18‑24 10.3% 10.4% 8.7% 6.9% 12.0% 12.3% 7.8% 8.2% 10.4% 44.0%
25‑34 32.1% 31.7% 37.1% 30.2% 32.6% 36.0% 25.2% 33.4% 29.5% 27.3%
35‑44 26.3% 26.1% 28.8% 26.2% 25.5% 27.8% 28.1% 33.3% 26.8% 12.2%
45‑54 18.5% 18.5% 18.1% 20.3% 18.9% 14.6% 26.0% 19.9% 18.2% 16.5%
55‑64 9.7% 10.0% 6.0% 11.6% 9.1% 6.9% 8.9% 5.1% 11.8% 0.0%
65 or older 3.1% 3.2% 1.4% 4.8% 1.9% 2.3% 4.0% 0.1% 3.4% 0.0%

Age group (row totals)

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
18‑24 21.2% 38.8% 24.5% 1.1% 0.7% 11.2% 0.2%
25‑34 29.5% 33.8% 22.9% 1.1% 1.0% 10.2% 0.0%
35‑44 31.2% 32.2% 21.5% 1.5% 1.2% 11.3% 0.0%
45‑54 34.5% 34.0% 16.1% 2.0% 1.0% 10.9% 0.0%
55‑64 37.5% 31.3% 14.6% 1.3% 0.5% 13.6% 0.0%
65 or older 48.6% 20.0% 15.6% 1.9% 0.0% 12.4% 0.0%



Appendix B: Childhood and family experiences

Highest grade completed before prison admission

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Less than high school 61.7% 62.6% 52.8% 51.5% 68.1% 68.5% 58.0% 54.9% 60.1% 64.8%
High school graduate 22.5% 22.6% 23.1% 26.0% 20.5% 19.5% 24.7% 23.1% 24.2% 0.0%
Some college 11.2% 10.7% 18.1% 16.1% 8.5% 7.2% 14.7% 13.1% 11.7% 35.2%
College degree or more 3.6% 3.4% 5.8% 5.8% 2.1% 2.5% 2.1% 6.0% 3.3% 0.0%

Homeless before age 18

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 11.8% 11.7% 12.4% 12.2% 10.1% 11.2% 15.7% 10.7% 16.3% 46.0%
No 88.2% 88.2% 87.6% 87.7% 89.9% 88.8% 84.3% 89.1% 83.6% 54.0%

Lived in public housing before age 18

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 18.7% 18.9% 16.8% 9.0% 30.6% 14.4% 13.8% 13.3% 20.0% 60.9%
No 79.6% 79.4% 82.2% 89.7% 67.6% 84.1% 82.9% 86.2% 77.6% 39.1%

Family received government assistance before age 18

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 41.7% 41.6% 42.8% 30.4% 52.3% 39.1% 44.5% 44.5% 47.0% 64.8%
No 55.3% 55.3% 54.7% 66.8% 44.7% 57.9% 51.0% 59.6% 49.7% 35.2%

In foster care before age 18

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 17.9% 17.8% 19.3% 18.6% 16.8% 14.4% 27.6% 13.4% 24.9% 40.0%
No 82.0% 82.2% 80.7% 81.4% 83.2% 85.5% 72.4% 86.6% 75.1% 60.0%

Either parent was ever incarcerated

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 32.6% 32.4% 34.8% 27.3% 37.4% 31.5% 36.4% 14.1% 35.9% 96.1%
No 67.4% 67.6% 65.2% 72.7% 62.6% 68.5% 63.6% 85.9% 64.1% 3.9%



Appendix C: Employment and housing before arrest

Employed in the 30 days before arrest

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 60.8% 61.9% 47.0% 65.5% 54.3% 66.3% 60.0% 61.1% 57.8% 48.0%
No 39.1% 38.0% 53.0% 34.4% 45.7% 33.7% 39.6% 38.9% 42.0% 52.0%

Living space in the 30 days before arrest

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Family-owned house, apartment, etc. 82.8% 83.0% 80.4% 81.3% 85.0% 84.7% 78.0% 87.5% 78.2% 75.4%
Someone else's house, apartment, etc. 29.0% 28.7% 32.9% 26.7% 29.6% 29.3% 32.6% 24.9% 32.8% 49.8%
Transitional housing 2.0% 2.0% 1.9% 2.3% 1.6% 1.9% 2.4% 3.8% 2.2% 0.0%
Residential treatment facility 1.4% 1.4% 2.0% 1.7% 1.1% 1.3% 1.8% 2.1% 1.1% 0.0%
Rooming house, hotel, or motel 6.5% 6.4% 8.6% 7.4% 5.2% 6.5% 13.6% 5.8% 7.7% 0.0%
Car, truck, or motor vehicle 3.1% 3.0% 4.5% 3.7% 1.8% 3.2% 5.3% 5.1% 4.0% 0.0%
Homeless shelter or the streets 4.9% 4.8% 5.8% 5.5% 3.2% 4.8% 10.3% 5.7% 7.2% 0.0%
Prison, jail, or correctional facility 3.7% 3.7% 3.5% 3.5% 3.7% 3.3% 2.9% 1.8% 5.0% 0.0%
Other 1.9% 1.9% 1.7% 2.0% 1.6% 2.1% 1.9% 2.4% 2.0% 0.0%

Homeless in year prior to arrest

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 10.1% 9.6% 16.1% 11.8% 8.9% 8.4% 14.6% 8.6% 11.9% 12.4%
No 89.8% 90.3% 83.9% 88.0% 91.0% 91.6% 85.2% 91.4% 88.2% 87.7%



Appendix D: Substance use disorder

Ever received substance use disorder (SUD) treatment?

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 46.3% 45.6% 55.3% 55.8% 38.1% 41.5% 61.5% 44.3% 51.3% 28.5%
No 53.7% 54.3% 44.7% 44.1% 61.9% 58.5% 38.5% 55.7% 48.7% 71.5%

Met criteria for SUD* at the time of the survey

*These figures come from Bureau of Justice Statistics (BJS) publication, Alcohol and Drug Use and Treatment Reported by Prisoners: Survey of Prison Inmates, 2016. The BJS used administrative sex rather than self-reported gender identity for this variable.
Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 49.0% 48.2% 58.5% 58.4% 38.5% 45.8% 61.8% 39.2% 53.9% n/a



Appendix E: Criminal history

Age at first arrest

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
10 or under 3.9% 4.1% 1.0% 3.3% 4.4% 3.5% 3.4% 2.7% 4.7% 0.0%
11 to 15 33.7% 34.7% 21.1% 26.8% 39.0% 33.9% 38.3% 33.9% 36.3% 62.9%
16 to 18 30.0% 30.6% 23.3% 28.9% 33.1% 27.7% 28.4% 22.2% 29.0% 33.2%
19 to 21 12.0% 11.7% 15.6% 13.2% 10.9% 11.8% 13.5% 11.1% 11.4% 0.0%
22 to 25 7.3% 6.9% 12.5% 8.7% 5.9% 8.2% 3.3% 7.2% 6.9% 0.0%
26 to 35 7.3% 6.6% 16.4% 9.4% 4.7% 8.2% 8.6% 12.9% 7.6% 3.9%
36 to 45 3.4% 3.2% 6.1% 4.9% 1.4% 4.4% 2.4% 9.0% 2.6% 0.0%
46 to 55 1.5% 1.3% 3.2% 2.9% 0.4% 1.4% 1.3% 1.1% 0.8% 0.0%
56 to 65 0.8% 0.8% 0.8% 1.5% 0.1% 0.8% 0.0% 0.0% 0.7% 0.0%
66 and older 0.3% 0.3% 0.1% 0.6% 0.0% 0.2% 0.8% 0.0% 0.1% 0.0%

Ever in a juvenile placement facility

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Yes 33.7% 34.6% 22.1% 26.3% 39.6% 36.0% 35.4% 27.7% 32.8% 50.6%
No 66.3% 65.4% 77.9% 73.8% 60.4% 64.0% 64.6% 72.3% 67.2% 49.4%

Number of lifetime arrests

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Zero 1.1% 1.1% 1.1% 0.9% 0.8% 1.8% 0.2% 0.3% 0.9% 0.0%
1 to 2 24.2% 23.6% 31.4% 25.3% 21.0% 29.1% 14.3% 43.6% 21.1% 12.2%
3 to 5 29.8% 29.8% 29.4% 26.5% 33.6% 28.2% 30.5% 31.0% 30.4% 60.2%
6 to 10 23.3% 23.5% 20.8% 22.3% 25.5% 21.7% 25.1% 15.6% 23.1% 27.6%
11 to 20 14.0% 14.2% 11.0% 15.7% 12.4% 13.4% 16.4% 6.7% 14.4% 0.0%
21 to 40 5.5% 5.6% 4.6% 6.5% 4.7% 4.5% 9.1% 0.6% 7.2% 0.0%
41 to 60 1.3% 1.3% 1.1% 1.8% 0.9% 0.8% 3.7% 0.8% 1.7% 0.0%
61 to 100 0.7% 0.7% 0.5% 0.7% 0.7% 0.4% 0.1% 0.0% 1.0% 0.0%
101 to 250 0.3% 0.3% 0.3% 0.2% 0.4% 0.2% 0.6% 0.7% 0.1% 0.0%
Over 251 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.0% 0.8% 0.0% 0.0%

Number of times ever placed on probation

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Zero 21.4% 21.2% 23.2% 21.3% 19.1% 26.2% 16.5% 29.0% 19.7% 33.2%
1 to 2 53.5% 53.5% 54.2% 48.9% 59.9% 51.3% 45.6% 60.1% 52.0% 51.8%
3 to 5 19.5% 19.7% 17.3% 22.2% 17.3% 17.1% 27.8% 10.4% 22.9% 15.0%
6 to 10 4.4% 4.4% 4.3% 5.9% 3.0% 4.5% 6.1% 0.5% 3.9% 0.0%
11 to 20 0.9% 0.9% 0.8% 1.3% 0.5% 0.8% 4.0% 0.0% 1.2% 0.0%
21 to 50 0.3% 0.3% 0.2% 0.4% 0.2% 0.1% 0.0% 0.0% 0.4% 0.0%
More than 51 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0%

Number of prior jail incarcerations

Gender identity (self‑reported) Race/ethnicity (self‑reported)
Total state prison population Male Female White (Not Hispanic) Black (Not Hispanic) Hispanic American Indian/Alaska Native (Not Hispanic) Asian/Native Hawaiian/other Pacific Islander (Not Hispanic) Two or more races (Not Hispanic) Other (Not Hispanic)
Zero times 42.0% 40.1% 40.8% 40.7% 42.6% 37.0% 25.7% 39.2% 38.3% 28.9%
Once 20.8% 19.8% 19.9% 18.1% 20.4% 21.0% 22.1% 31.0% 19.4% 71.2%
2 to 10 times 37.1% 35.5% 34.6% 35.3% 33.5% 37.6% 44.1% 28.5% 37.5% 0.0%
11 to 30 times 0.1% 4.0% 4.1% 5.0% 3.1% 3.9% 7.5% 1.4% 4.2% 0.0%
31 to 50 times 0.0% 0.4% 0.4% 0.6% 0.3% 0.3% 0.0% 0.0% 0.3% 0.0%
51 to 100 times 0.1% 0.2% 0.2% 0.2% 0.1% 0.1% 0.6% 0.0% 0.3% 0.0%
More than 101 times 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%


Stay Informed


Get the latest updates:



Tweet this page Donate