This report presents statistics on full-time federal law enforcement officers who were authorized to make arrests, carry firearms, or both. Specifically, this report describes job functions and demographic characteristics of federal law enforcement officers. Key findings include that 88 federal agencies reported employing 133,798 full-time federal law enforcement officers. In addition, the U.S. Department of Homeland Security employed 65,106 federal law enforcement officers, about half (49%) of all federal law enforcement officers. Criminal investigation was the primary function for about two-thirds (70%) of federal officers. Lastly, 15% of federal law enforcement officers and 14% of supervisory law enforcement personnel were female.

For decades, advocates, observers, and policymakers have debated how to address the problem of the federal Controlled Substances Act and drug guidelines recommending punishment based on formulas that emphasize drug quantity. This results in low-level participants, as opposed to leaders of drug trafficking organizations, often receiving hefty sentences, either because of the application of a lengthy statutory minimum or by virtue of being sentenced under a severe (yet advisory) guidelines range.. This essay offers two contributions. First, it uses U.S. Sentencing Commission data to provide a descriptive empirical account of the considerable disconnect between culpability and sentence lengths for federal defendants who perform low-level functions in drug-trafficking organizations that deal in large quantities of drugs. The author shows that judges appear to respond to this disconnect by routinely sentencing below the advisory guidelines range for drug-trafficking defendants convicted in the upper echelons of offense levels—a trend that has been increasing over the last twenty years. Second, the essay argues that these persistent downward reductions highlight a structural flaw in the drug guidelines’ reliance on quantity as a proxy for culpability. The author asserts that reducing base offense levels at the top of the drug guidelines would better align recommended sentences with judicial practice, legislative intent, and the actual conduct of low-level defendants in large-scale drug conspiracies.

Artificial intelligence (AI) is rapidly embedding itself into the core machinery of the criminal justice system, powering everyday decisions from police analysis of digital evidence and pattern detection in crime data to prosecutorial discovery management, charging recommendations, algorithmic risk assessments in courts, and large language models for summarizing records and drafting documents. These tools promise efficiency gains — processing vast data volumes, reducing backlogs, and optimizing scarce resources in an overburdened system — but they also carry profound risks: embedding bias, producing opaque or unreliable outputs, shifting unmonitored power to vendors, and influencing high-stakes liberty decisions like arrests, detention, sentencing, and release. The central problem is that AI capabilities are deploying without sufficient understanding of their mechanics, failure modes, or implications for constitutional rights, democratic accountability, and system legitimacy. Criminal-justice entities who encounter AI tools — thousands of under-resourced police departments, prosecutors’ offices, courts, and probation units — lack the technical expertise to evaluate these tools rigorously, while vendors market directly to practitioners. This creates a governance gap: even well-intentioned actors cannot reliably apply emerging standards amid rapid technological changes, risking uneven, superficial oversight that undermines public trust.

Over the last two decades, a wave of investment from philanthropy, government, and the private sector has produced a growing set of tools intended to help people demonstrate what they know, employers understand who they’re hiring, and educators understand whether their programs are preparing people for real opportunities. Digital credentials, learning records, competency frameworks, talent marketplaces: the tools exist, and in isolated pockets, they work. The problem is that these tools don’t work together. Skills data created in one system can’t be read by another. Credentials issued by one institution aren’t recognized by the next. Employers who want to hire based on skills don’t have the infrastructure to do it consistently. Workers and learners who move between jobs, programs, or states find that their records don’t follow them, leaving them to start over repeatedly, regardless of the skills they’ve already demonstrated. States have an opportunity that few other actors share: the ability to build and govern infrastructure at population scale. But "the state" government is not a single actor. Workforce agencies, higher education systems, social services departments, and corrections agencies each hold different pieces of the picture, operate different data systems, and answer to different rules and priorities. A genuinely functional state skills ecosystem requires coordination across all of them, which is itself a significant design and governance challenge, separate from the technical one.

Roughly one in five undergraduates—or 3.1 million students—are parents, and nearly half are enrolled at community or technical colleges. Nearly 23% of public two-year college students have dependent children, compared with 11% of undergraduates at public four-year colleges. However, many colleges lack services for student parents, or students don’t use them. For example, only 38% of public two-year colleges offer on-campus childcare. Sixty percent of students reported never using career counseling services, and 88% said they never used job placement assistance, despite most parenting students in community college citing career change obtaining or updating job skills as the main goal of attending college. In addition, 27% of students who are parents when they enroll at public two-year colleges earn a degree or certificate within six years, compared with 41% of non-parenting students. However, there are reasons to focus on increasing college completion among student parents. For example, mothers who reenroll earn on average nearly $6,800 more annually after completing an associate or bachelor’s degree and their children are 38% more likely to complete a college degree. Policy supports could include a focus on affordable, reliable childcare and services related to basic needs including food, housing, and transportation. Further, reliable data on student parents is sparse. Policymakers should prioritize more systematic data collection to identify student parents, track their academic progress, and determine what help they need to complete their programs.

Research indicates that perfectionism is on the rise among college students. This study updates and expands on this work in several ways. This included the research team investigating whether self-oriented and socially prescribed perfectionism continue to increase in tandem with personal standards, concerns about mistakes, and doubts about actions. Further, the research team examine generational differences in higher order dimensions of perfectionism (perfectionistic strivings and concerns). Cross-temporal meta-analyses of 307 samples encompassing 82,939 American, Canadian, and British college students revealed that self-oriented perfectionism, concerns over mistakes, and doubts about actions increased linearly. Socially prescribed perfectionism followed a quadratic trajectory, with a notable acceleration starting in the early 2000s. At the higher order level, perfectionistic strivings increased linearly, whereas perfectionistic concerns followed a quadratic trajectory. Overall, results reveal that college students increasingly perceive others as excessively demanding while becoming more demanding of themselves, accompanied by growing indecisiveness, uncertainty, and sensitivity about making mistakes.

Vacancy rates are an indicator of the health of housing markets. These rates measure the availability of vacant housing for sale or for rent, which are both a function of the demand for and supply of housing in a market. The U.S. Census Bureau’s American Community Survey measures vacancy rates annually down to the small geographies where housing demand and supply interact. The time period from the global financial crisis to the post-pandemic period saw a steady decline in vacancy rates in both the owner occupied and rental markets across all building types, home sizes, and county metropolitan status. This decline is consistent with a tightening housing market, in which demand for housing is greater than the supply of housing available. This decline also coincides with increasing housing costs, another outcome of high demand for and low supply of housing. Counties with among the highest homeowner vacancy rates included several counties across Texas and Florida: Bexar (1.5%), Broward (1.4%), Harris (1.4%), and Palm Beach (1.4%). The highest rental vacancy rates among these counties were in growing cities in the Sun Belt, in the states of Arizona, Florida, and Texas.

The federal Bureau of Labor Statistics’s (BLS) Employment Situation or jobs report provides key information on the nation’s economy based on data from two surveys—one of households (the household survey) and one of employers (the establishment survey). Stakeholders with jobs data expertise said the report generally meets users’ needs. However, they said occasional large revisions can make the data less useful for informing timely decisions, and the BLS faces risks to data quality due to lower survey response rates over time. The BLS and U.S. Census Bureau are planning to add an online response method to the household survey in 2027 to try to increase response rates and lower data collection costs. However, officials said recent funding constraints may delay full implementation. The Government Accountability Office recommended that BLS develop a plan to address gaps in its ability to obtain external input on the jobs report data and publish an assessment on the effects of survey nonresponse on the establishment survey.

According to the Federal Trade Commission, consumers lost $12.8 billion to fraud in 2024, with actual losses estimated to be as high as $196 billion. Research on financial fraud knowledge is scant, limiting development of effective protection tools. Using RAND's American Life Panel, 1,509 respondents (ages 21–90) listed up to five fraud schemes and tactics. The authors explore respondents' conceptualizations of fraud types through the free-listing technique and categorize the answers into five groups: threat-based (which include a clear reference to fear, panic, or urgency to avoid something bad), opportunity-based (which include an opportunity, often capitalizing on a feeling of excitement), consumer-based (which include purporting to sell a service or product, or collecting money for a product you have or would buy), imposter-based (which include a clear reference to impersonation of a company, entity, or person, other than the victim), and ID-based (which include a clear reference to using personal identifiable information or theft of the victim's identity) frauds. The report finds that women mention more imposter-based and fewer ID-based scams than men, while older adults mention more consumer-based scams than their younger counterparts. Higher financial literacy is associated with a greater likelihood of mentioning threat-, ID-, and imposter-based fraud. Mentioning ID- or consumer-based scams is tied to a greater likelihood of being targeted by a fraudster, whereas mentioning threat-based scams is tied to a smaller likelihood of losing money.

Both retail health clinics and urgent care centers have become mainstream access points for people seeking medical care. Clinics and care centers offer similar, yet distinct, health services. Health clinics address quick, uncomplicated health needs that can be served outside the doctor’s office or urgent care center. Retail health clinics are typically staffed by nurse practitioners or physician assistants and are housed in pharmacies, grocery stores, or supercenters. Urgent care centers are freestanding facilities designed to treat acute non-life-threatening injuries and illnesses, and most are staffed by a full-time, onsite licensed physician. This report found that in 2024, 6% of people had at least one visit to an urgent care center and 19% had at least one visit to a retail health clinic in the past 12 months. Among adults ages 18–64, the percentage who had at least one visit to an urgent care center was lower among those living in large central metropolitan (26.6%) and non-metropolitan (26.1%) areas compared with those living in large fringe metropolitan (31.5%) and medium and small metropolitan (30.6%) areas. Among all age groups, the percentage who had at least one visit to a retail health clinic was lower for those living in medium and small metropolitan and non-metropolitan areas compared with those living in large central metropolitan and large fringe metropolitan areas.

Cyberattacks on healthcare systems are increasing. This puts the federal electronic health record system, which supports health care for millions of service members and veterans, at risk. Four federal agencies, including the U.S. Departments of Defense and Veterans Affairs, use the system to store, share, and analyze patient information. The Federal Electronic Health Record Modernization (FEHRM) office—a joint department office—facilitates collaboration among these agencies. The Government Accountability Office made a recommendation to the departments to direct FEHRM to define common goals, outcomes, and associated performance measures, and monitor, assess, and communicate progress on collaboration efforts toward ensuring the cybersecurity and privacy of the federal enclave. The Department of Defense disagreed with the research team’s report and Veterans Affairs neither agreed nor disagreed with the recommendation.