This web report provides statistics on the percentage of female recruits who started and completed basic training, by state and by type of academy. Data are from federal Bureau of Justice Statistics Census of Law Enforcement Training Academies, which was last administered in 2018 (data released November 2021). The census gathers information from state and local academies that are responsible for administering mandatory basic training to newly appointed or elected law enforcement officers. This report found that in 2018 females accounted for the highest percentage of recruits who completed basic training in Montana (34.3%), Idaho (28.3%), California (23.3%), and Oklahoma (23.2%). In Florida, 20.5% recruits who completed were female. Additionally, the report found that one in five recruits who completed basic training at academies operated by county police (21%), by 2-year colleges (21%), and by technical schools (20%) were female.

Integrating artificial intelligence (AI) technologies into law enforcement has become a concern of contemporary politics and public discourse. This study synthesizes qualitative analysis to (1) highlight potential societal benefits and the ethical concerns of AI technologies in policing, and (2) inform the responsible design and integration of AI technologies in law enforcement. Through conducting semi-structured interviews of law enforcement professionals in North Carolina, the authors investigate how integrating AI technologies, such as predictive policing and autonomous vehicle technology, impacts the relationships between communities and police jurisdictions. The evidence suggests that police officers maintain that AI plays a limited role in policing but believe the technologies will continue to expand, improving public safety and increasing policing capability. Conversely, police officers believe that AI will not necessarily increase trust between police and the community, citing ethical concerns and the potential to infringe on civil rights. It is thus argued that the trends toward integrating AI technologies into law enforcement are not without risk. Policymaking guided by public consensus and collaborative discussion with law enforcement professionals must aim to promote accountability through the application of responsible design of AI in policing with an end state of providing societal benefits and mitigating harm to the populace.

Roughly half of U.S. counties do not provide defense counsel at bail hearings, and few studies have documented the potential impacts of legal representation at this stage. This paper presents the results from a field experiment in Allegheny County, Pennsylvania, that provided a public defender at a defendant's initial bail hearing. The presence of a public defender decreased the use of monetary bail and pretrial detention without increasing failure to appear rates at the preliminary hearing. Specifically, the authors found that providing public defenders at bail hearings increased the probability of receiving release with no conditions or with nonmonetary release at bail hearings by 21%, reduced the probability an individual was in jail 3 days after their bail hearing by 10%, and had no impact on failure to appear rates or the probable cause determination at the preliminary hearing. The intervention did, however, result in a short-term increase in rearrests on theft charges, although a theft incident would have to be at least 8.5 times as costly as a day in detention for jurisdictions to find this tradeoff undesirable.

Jobs of the future will require all students to have some level of computer science education to be proficient in the workforce. A host of policy levers – spanning K-12, postsecondary and the workforce – are available to state leaders seeking to bolster their state’s work in this area. According to Code.org’s 2022 State of Computer Science Education report 53% of U.S. high schools offer foundational computer science, while 5.6% of high school students are enrolled in those courses,; Nationally, about 32% of female students participate in foundational computer science courses, with three states (Maryland, Mississippi, South Carolina) exceeding 40% participation. Further disparities exist among student enrollment in these courses. For example, schools that offer foundational computer science education are less likely to enroll Latino and Hispanic students than they are to enroll Asian or white students. This report also includes computer science policy actions by Iowa, New York, Tennessee, California, Washington, Kentucky, Missouri, and Louisiana. These actions include requiring a computer science course for all middle school students (TN), providing grant funding to develop programs that help to fill computer science teacher shortages such as by providing housing for qualified teachers (CA), and allowing a computer science course to count as a third-year credit for math or science to meet high school graduation requirements (WA).

This study focuses on the disparities between the educational aspirations and expectations of high school students in the U.S., and explores why students do not expect to earn a bachelor’s degree despite their aspirations. Using a national dataset and logistic/multinomial logistic regression analyses, the study identified the factors associated with this diminished expectation, such as family background, school experiences, and college preparedness activities. The results indicated that when the highest parental education level was a bachelor’s degree, subjects were 8.2% less likely to have low expectations for obtaining a bachelor’s degree. High school students who attended public school were 3.2% more likely to have lower expectations than college aspirants, and those who attended schools located in the Northeast compared to Midwest had 4.2% less lower expectations for college aspirants. Additionally, results indicated that students with low expectations of reaching their college aspirations are less likely to search for, apply to, and/or enroll in 4-year colleges. Students who had low expectations for obtaining a bachelor’s degree were 7% less likely than those with high expectations to search for college information, even after holding other variables such as personal, family, and high school backgrounds and factors constant. Also, students who had low expectations for obtaining a bachelor’s degree were 2.4% less likely to apply to colleges, as compared to the students with high expectations, even after controlling for whether the student searched for information.

For the past two decades, school shootings have been a constant looming threat for students across the U.S., with the number of shootings on campuses dramatically increasing in recent years. In response, schools have instituted preparedness and response measures, with one of the most common being school-shooter drills. Educators have raised concerns that these drills might negatively affect student well-being, and students have reported experiencing distress after participating in these drills. Although the data needed to measure the relationship between school-shooter drills and student mental health (and student well-being more broadly) are not readily available, analyzing the impact of drills on accountability outcomes in Arkansas—specifically, attendance rates and proficiency rates on statewide end-of-year tests—can help policymakers understand how they might affect students’ academics. Data from 2016 through 2019 show that in English and math, students in grades three through five who test on the school days immediately after an active-shooter drill have lower proficiency rates than their counterparts who test on the days and weeks before the drill. Proficiency rates return to typical levels as tests are administered in the weeks following the week of the drill. Attendance rates are slightly lower (a 0.09% decrease) during quarters when active-shooter drills occur compared with quarters without active-shooter drills. Though policymakers should continue their efforts to understand the impact that these drills have on student outcomes, the data on standardized test scores suggests that school leaders should consider avoiding scheduling active-shooter drills on the days before major tests.

More than 90% of potentially eligible individuals received pandemic-related stimulus payments but receipt varied across race and ethnic groups, according to recently published joint federal research. Among the four largest race/ethnic subgroups of the potentially eligible population (Non-Hispanic White, Non-Hispanic Black, Non-Hispanic Asian and Hispanic of any race), Hispanic individuals had the lowest receipt rate (87%) and non-Hispanic White individuals the highest (94%). In addition, lower-income individuals and families with children received payments earlier than higher-income individuals and families without children. Overall, 92% of potentially eligible individuals received payments. With respect to timing, 95% of recipients received payments in the first six weeks of the program and more than half (55%) in the first week of the program. In addition, 90% of every race/ethnic subgroup received payments within six weeks of when payment disbursement began, but Non-Hispanic White and Non-Hispanic Asian recipients were most likely to receive their payments within the first six weeks. Lower-income individuals and families with children received payments earlier, indicating that the way the Internal Revenue Service prioritized payments was effective.

Research has suggested that people living in higher poverty areas experience more acute systemic problems than people in lower poverty areas (e.g., limited access to medical services, healthy and affordable food, quality education, and civic engagement opportunities). Government agencies and researchers have previously identified counties with high rates of poverty over an extended period as targets for increased level of support. To identify counties in persistent poverty, this report incorporates poverty estimates from the 1990 and 2000 Censuses, the 2005–2009 American Community Survey (ACS) 5-year estimates, and the 2015–2019 ACS 5-year estimates. This report expands upon the persistent poverty literature by examining sub-county geographies and comparing those results to county results. Census tracts were more geographically precise in identifying persistent poverty populations than counties. From 1989 to 2015–2019, there were 341 counties (10.9% of the total) in persistent poverty; Florida had 4 persistent poverty counties (Alachua, Hardee, Madison, and Putnam ). From 1989 to 2015–2019, there were 8,238 tracts (11.3% of the total) in persistent poverty; Florida had 330 persistent poverty census tracts. In 2019, approximately 6.1% of the U.S. population lived in a persistent poverty county (1.8% of Florida’s population) and approximately 9% lived in a persistent poverty tract (6.6% of Florida’s population). Approximately 9.1 million more people lived in a persistent poverty tract than lived in a persistent poverty county. Over 74% of persistent poverty census tracts were not in a persistent poverty county.

Researchers and policymakers increasingly recognize the importance of job quality in contributing to worker well-being and labor market outcomes. One important form of data on job quality is administrative data—data collected directly from employment or earnings records as part of existing program operations, such as tax collection. However, a full and accurate assessment of the quality of jobs held by workers in the United States is hindered by limited data on and measures of both central features of work and important classes of jobs and workers. These limits not only constrain efforts to describe job quality trends and distributions, but also hamper the progress of research needed to understand both the determinants and importance of job quality, as well as hold back efforts to design policies and implement and operate programs that can improve job quality. The authors offer a range of potential directions for advancing wage record enhancement, with a focus on the potential for improving job quality measurement.

Since 2007, the National Health Interview Survey Early Release Program has regularly released preliminary estimates of the percentages of adults and children living in homes with only wireless telephones (also known as cellular telephones, cell phones, or mobile phones). These estimates are the most up-to-date estimates available from the federal government concerning the size and characteristics of this population. Estimates in this report are based on the second six months of 2022. During this time period, 72.6% of adults and 81.9% of children lived in wireless-only households. Survey data can also be used to estimate the percentage of adults who live in wireless-only households and have their own wireless telephone (wireless-only adults). For July-December 2022, 71.7% of adults were wireless-only adults. Demographic subgroups with the highest percentages of wireless-only adults include adults aged 25–29 (87.6%) and 30-34 (88.4%), and adults renting their homes (85.3%). Hispanic adults (80.0%) were more likely than non-Hispanic Asian (73.0%), non-Hispanic Black (69.5%), or non-Hispanic White (69.5%) adults to be wireless-only. Men (72.4%) were more likely than women (71.1%) to be wireless-only. Adults living in the Midwest (73.8%), South (74.1%), and West (76.0%) were more likely than those living in the Northeast (58.0%) to be wireless-only.

This brief has three goals: (1) describe the evidence supporting economic and concrete supports as a child welfare prevention mechanism and evidence-based service; (2) identify the programs that provide economic and concrete supports and that are already rated on the Title IV-E Prevention Services Clearinghouse, a resource for states that conducts objective and transparent reviews of research on programs and services that provide enhanced support to children and families and prevent foster care placement; and (3) discuss policy and practice changes in child welfare and other family-serving systems that could facilitate community-level change in child and family well-being. The authors conducted an extensive literature review to summarize and synthesize the extant of the literature supporting economic and concrete supports as a prevention mechanism for child welfare. The literature review informed the discussion and policy and practice implications presented in the brief. The authors recommended prioritization of services providing economic and concrete supports when designing and implementing Family First prevention plans; inclusion of economic and concrete supports as key components of behavioral health, substance use, and in-home family support program development; implementation of systematic screening for economic and concrete support needs within child welfare and in upstream systems; creation of programs that center family voice, are highly relational, and involve building community capacity and resources through increased communication and integration across family-serving systems; and ensuring that any policy designed to prevent or address child welfare involvement and out-of-home placement includes economic and concrete supports as a core resource and strategy.

The First Star Institute, a non-profit that promotes best practices serving abused and neglected children in education, research and the law, sought to build from the existing knowledge base to focus more in depth on defining and leveraging the important – and likely largely untapped – role that attorneys for children, including those acting as guardians ad litem (GALs) and court appointed special advocates (CASAs) play in such collaborative efforts. To learn more about the involvement of children’s attorneys, GALs and CASAs in cross-systems collaborative efforts to address child sex trafficking, particularly among the child welfare population, the institute conducted a set of targeted key stakeholder interviews with selected jurisdictions that have established collaborative child sex trafficking initiatives. The authors focused their efforts primarily on selected sites from the federal Children’s Bureau Grants to Address Trafficking Within the Child Welfare Population, representatives from select specialized child sex trafficking courts, and selected CASA associations focusing on this issue. The institute completed 11 interviews with 15 key stakeholders. Those interviewed represented attorneys for children, GALs, CASAs, judges and child welfare professionals spanning eight states (Arizona, California, Connecticut, Florida, Massachusetts, South Carolina, Washington and Washington DC). Interviewees identified several effective approaches (e.g., resolving the lack of appropriate and safe placements, increasing trafficking-specific services, enhancing widespread training and education). Several next steps were suggested, including the need for specialized training for child welfare attorneys and other advocates; the use of multidisciplinary team meetings; cross-systems training to improve communication and clarify roles and responsibilities of each partner; the importance of including survivor mentors/advocates as a key team partner; and the necessity of increasing the availability of appropriate and safe placements for child victims of sex trafficking.