top of page
  • Writer's pictureProject on Workforce Team

Navigating Public Job Training

Text on red background: Navigating Public Job Training

Navigating Public Job Training

Harvard Project on Workforce

David Deming, Alexis Gable, Rachel Lipson, and Arkādijs Zvaigzne

March 2023


Download a copy of the full report:

Download an overview of federal policy recommendations in this report:




Executive Summary

This report describes and analyzes the more than 75,000 “Eligible Training Provider” (ETP) programs in the United States. ETP programs are job training programs deemed eligible for funding under America’s primary federal workforce development law, the Workforce Innovation and Opportunity Act (WIOA).[1] Among other functions, WIOA funds vouchers for unemployed or underemployed workers to enroll in job training services. The vouchers are typically used to support enrollment in short-term, non-four-year-degree programs that connect to "in-demand employment” opportunities in a regional economy. Under the law, each state and territory must maintain a list of pre-approved programs that eligible individuals may select from. The programs on these lists (commonly known as “eligible training provider lists” – ETPLs) comprise our primary unit of analysis.

We analyze federal and state data sources to better understand the publicly-funded job training landscape in the United States. We combine training provider and program data from the Department of Labor (DOL) with individual performance records and occupational datasets to study the types of providers receiving WIOA funding and the kinds of jobs for which they are training. In addition, we look at state websites for all 50 U.S. states to understand how program information is made available to potential enrollees. Our analysis seeks to answer three primary research questions:

1) What are the most common characteristics of WIOA-eligible training providers and programs?

2) Which fields of study and occupations are most commonly supported by federal funding?

3) Is federal funding for workforce training directed towards good-paying and in-demand occupations?

Key Findings

First, federal funding for job training is small compared to traditional higher education, and the dollars are widely dispersed. We estimate that WIOA vouchers fund under $500mn in training annually, which is about one-fiftieth the size of the ~$25bn spent by the federal government on Pell Grants for undergraduate education.[2] Similarly, few people benefit from workforce training vouchers. In Program Year 2019, WIOA funded training vouchers for approximately 220,000 individuals in its adult and dislocated worker programs.[3] In contrast, over 6 million students receive Pell Grants each year.

However, despite the smaller reach, the landscape of ETP offerings is vast and hard to navigate. According to our analysis, there are over 7,000 eligible training providers and approximately 75,000 eligible programs in more than 700 occupational fields nationwide. Performance information at the provider level is very limited and of questionable accuracy. The net result is a highly fragmented system, where strong programs are not differentiated from weak ones, and where incentives for high-performing providers to participate in WIOA are limited. According to our estimates, nationally, the average ETP program enrolled just three WIOA-funded learners per year, which would correspond with around $6,000 total in training revenue for the provider. The implication is that WIOA is not a sustainable or scalable source of funding for most providers on these lists, as they must draw from other sources to fund most of their learners. This finding poses challenges to the consumer choice architecture the law is designed around.

Second, while federal policy intends to support training for good jobs, overall, our analysis of the ETP data implies that our public workforce development training dollars do not promote job quality. Many eligible programs train for low-wage occupations, especially in healthcare. For instance, training participants enrolled in the most common (medical assistant) and fourth-most common (nursing assistant) ETP programs make under $6,000 in a quarter, or $24,000 annually. We estimate that over 40 percent of WIOA training participants earn under $25,000 annually. Women and participants of color are especially likely to be enrolled in training programs for low-wage occupations. Ground transportation programs (e.g. Commercial Driver’s License programs for heavy truck driving) represented the largest share of eligible programs nationwide with at least 50 WIOA-funded participants. While these transportation jobs often pay above-average wages for workers without college degrees, they have limited potential for upward mobility. Overall, our analysis suggests WIOA training vouchers fall short of the economic advancement objectives of the legislation.

The system performs better on prioritizing in-demand occupations. Fifty-four percent of programs eligible to receive WIOA funds are in occupations that are expected to grow in the coming decade. However, only two of the ten most common programs correspond to the top quartile of projected fastest-growing U.S. occupations between 2020 and 2030. Some high-growth fields that pay well and don’t require a BA, like oil and gas drill operators or forest fire specialists, appear under-represented, with less than ten eligible programs nationally. In addition, we identify 10,000 eligible programs that target jobs that the Bureau of Labor Statistics expects will decline in the next ten years. Though many WIOA observers cite a perceived tradeoff between wages and employer demand, we find no relationship between training program earnings and occupational growth rates.

Finally, there is limited consumer-facing information about these programs, making it difficult to execute on the vision of “informed consumer choice” promised by the legislation. The Department of Labor’s website is the first national effort to show performance information at the program level, but it is missing critical information. In its first release, the database lacked information on completions, employment rates, median earnings, and credentials for over 75 percent of programs. Most state websites overwhelm potential users with thousands of potential programs and compliance-oriented information, but many do not include accessible information to help inform enrollment decisions, like program price, financial aid, and format.

We conclude with recommendations for state and federal policymakers. U.S. workforce development funding should prioritize upgrading technological infrastructure, expanding support for career navigation, and increasing overall funding for workforce training. In addition, we highlight needed areas of policy innovation to ensure that existing public dollars support high-quality training options. These include rethinking provider eligibility criteria, revamping data systems, and better supporting remote work and remote learning opportunities. Finally, we highlight quick wins that the Department of Labor and state partners could implement in the short-run to improve accessibility and efficacy in the current system.


The public discourse is filled with conversation about the importance of reskilling for the future of work. However, the flow of federal dollars to support job training is not well-understood. In this report, we take a deep dive into the programs intended as the key vehicle for U.S. federal investment in workforce development. The Workforce Innovation and Opportunity Act (WIOA) authorizes the primary federal programs dedicated to funding short-term, job-focused training for adults. The law aims to provide a “combination of education and training services to prepare individuals for work and to help them improve their prospects in the labor market.”[4]

Training provision through WIOA is highly decentralized, in part because most of the financial support for training comes through vouchers (known as Individual Training Accounts, or ITAs). According to our analysis of U.S. Department of Labor data, there are over 7,100 different training providers eligible for federal funding nationwide, and 75,442 different eligible programs in more than 700 occupational fields. Each state or territory is responsible for developing a list of training programs eligible to receive federal funds in its jurisdiction.

This report aims to better understand how well this system is performing relative to the goal of providing training services for in-demand jobs that boost participant earnings. To do this, we ask three primary questions:

1) What are the most common characteristics of WIOA-eligible training providers and programs?

2) Which fields of study and occupations are most commonly supported by federal funding?

3) Is federal funding for workforce training directed towards good-paying and in-demand occupations?


Administered by the U.S. Department of Labor’s Employment and Training Administration (ETA), WIOA is the primary source of federal funding for the U.S. workforce development system. WIOA has a mandate for workforce development, but its mission is broad in scope. Activities supported by the law can range from job search assistance to career counseling to on-the-job training to occupational training to the JobCorps residential program for youth. In addition, the law serves a number of different target population groups. For this report, we focus on two sub-programs that have a training mandate for adults: the Adult and Dislocated Worker programs. Given data availability, we focus on the landscape of training programs and participants during Program Year (PY) 2019.

In 2019, ETA received appropriations of $3.7 billion to support training and employment services.[5] Of that total, $1.3 billion went to “Dislocated Worker Employment and Training Activities” and $845 million to “Adult Employment and Training Activities.”[6] However, career services are more prevalent than training and account for the majority of the spending. According to WIOA performance reports, $427 million was spent on training activities for adults and dislocated workers.[7] In PY2019, over 220,000 people received training through WIOA’s adult, dislocated worker, and national dislocated worker grant programs.[8]

While these numbers may appear large, WIOA comprises only a small share of education and training dollars nationwide. The United States spends approximately $2.1 trillion per year on education and training, including $489 billion in estimated education-related expenditures on Title IV degree-granting institutions of higher education. Beyond the WIOA system, federal funding to postsecondary education-seekers comes in many forms, including Pell grants, direct loans, the Lifetime Learning Credit, and 529 plans. Pell grants alone provide over $25 billion in aid to approximately 6 million undergraduate students. [9]

However, funding from other federal sources tends to mandate that training takes place at an accredited higher education institution, be in a for-credit program, and meet a minimum length requirement (usually six months). For individuals seeking shorter training, WIOA funding can be only the eligible source of federal assistance. Still, in real terms, funding for WIOA has been declining over the last few decades.[10] Overall, the U.S. spends only 20 percent of the OECD average on active labor market policies as a share of Gross Domestic Product.[11]

Like Pell Grants, WIOA funding predominantly operates in a voucher system, where funding travels with an eligible individual rather than the provider. To receive a training voucher, individuals must meet a variety of eligibility criteria, including economic hardship and exhaustion of eligibility for other funding sources, like Pell grants. In addition, to access a voucher, individuals typically must visit a career center and meet with a counselor who certifies they meet the defined criteria and helps them develop an employment plan.[12] Like other educational voucher systems, WIOA is meant to provide consumer choice to participants. Individuals deemed eligible are provided with a local list of Eligible Training Provider (ETP) programs that have been certified through their local or state public workforce development system. Upon enrollment, payment is made to the training services provider through the Individual Training Account (ITA) voucher.

The law states that all programs on the ETPL must lead to “in-demand” jobs, be delivered by eligible training providers, and result in a recognized post-secondary credential. However, each state has final say on program eligibility. Many states’ eligibility requirements are primarily compliance-focused. For instance, the Massachusetts ETPL registration page lists prominently that providers must have no outstanding citations from state agencies, labor standards violations, and must be in good standing with the state’s unemployment and tax authorities.[13]

While most WIOA training dollars are spent through ITAs, not all federally-funded job training is accessed this way. WIOA does allow local entities to provide other means of supporting training, though the law’s consumer choice requirements must always be met. The Department of Labor also supports workforce training through other programs, including H-1B Skills Training Grants, registered apprenticeship funding, and competitive grants.

Past research on workforce development programs funded by WIOA is sparse. While the WIOA program overall has been evaluated in the past, most notably, the 2018 Gold Standard Evaluation, prior studies assess the program’s effect in aggregate.[14] They do not measure the impact of particular training providers or distinguish between the thousands of eligible programs or hundreds of occupations that participants can select from.

Looking more broadly at the U.S. workforce development literature, a growing body of training programs has been rigorously evaluated, most commonly known as sectoral employment programs. Programs like Year Up, Per Scholas, and Project QUEST have a proven track record of positive impacts on earnings and employment rates. However, the generalizability of those findings to WIOA’s eligible training programs is questionable. Year Up’s most recent evaluation notes that only two percent of its revenues come from government funding.[15] The vast majority of the 75,000 ETP programs have not been evaluated, and they may differ substantially from best-practice sectoral programs in program operations and pedagogy.

One important distinction is cost. The average cost per training participant served through WIOA was just $1,854 in 2019.[16] WIOA training vouchers typically have maximum caps between $5,000 and $10,000. In some cases, funding can be as low as $1,000 per participant. Reimbursement rates vary across regions and states.[17] In contrast, sectoral employment programs with strong evaluations tend to have all-in costs to operate ranging from $8,000 to $35,000 per participant. Among other features, these programs include in-house, intensive wraparound support services and dedicated employer relations teams that help contribute to their success. They also tend to feature cohort-based models where participants learn and advance through a course of study together in a group.


To understand the landscape of WIOA-funded training providers and programs, we utilize ETP programs as our unit of analysis. We analyze two publicly available sources of data on ETP programs (Figure 1). First, we utilize a dataset from the (TPR) website, a Federal Department of Labor initiative that went live in December 2020 and that aggregates information reported from states about their Eligible Training Programs according to a common reporting template approved by the White House Office of Management and Budget.[18] Its stated purpose is to help individuals make career training choices based on program completion and employment results. The public materials about TPR also cite a broader purpose of increasing data transparency and accountability to help job seekers, career counselors, employers, training providers, and policy leaders. In total, 75,442 ETP programs are represented in TPR, representing PY2019 and PY2020 data.

Second, we utilize state-level ETPLs, which all states are required by WIOA to display publicly. These lists are intended to be the primary resource for case managers, partners, and participants interested in enrolling in training activities funded by WIOA. States vary in how they publish these lists; for our analysis, we utilized either direct downloads or webscrapes from all 50 states, collected in the Summer of 2021 (see Appendix A).

We then combine these ETP data sources with other publicly-available data to answer our research questions. To understand WIOA training participation and outcomes, we use WIOA Individual Performance Records for 2019. These records use data from the major reporting system of the DOL, the Participant Individual Record Layout (PIRL). From these records, we aggregate training participant demographics and earnings to the occupation level, utilizing Standard Occupational Classification (SOC) codes. To analyze labor market demand, job growth, and educational requirements, we use occupational classifications and data from Federal Reserve Banks of Philadelphia and Cleveland’s Opportunity Occupations research.[19] Finally, to look more deeply at job growth, we use occupational data from the Bureau of Labor Statistics, including 2020-30 occupational projections and tags for “Bright Outlook Occupations,” which identify jobs that are expected to grow rapidly, have a large number of openings (100,000+), and are new and emerging.

Figure 1: Key Data Sources for ETP Program Analysis


Importantly, this data is imperfect. For more than 75 percent of the 75,000+ programs included in the first TPR release, data on program enrollment, completions, employment, earnings, and total credentials was not available at time of download (June 2021). Data could be suppressed through a waiver obtained by a state on reporting a specific performance metric to the Department of Labor. Additionally, to protect the personally identifiable information of participants, performance data is suppressed for programs serving fewer than 10 students, comprising 22 percent of all participants. Given the lack of participant data at the ETP program level, we focus our analysis on the prevalence of eligible programs, but this is an imperfect proxy.

Further, many data points that could be used in studying the causal impacts of programs are not collected at a national level. Programs are not required to report information on whom they serve. Only six percent of states list any participant demographic information by program. No information is listed about pre-enrollment earnings.

Finally, the information these datasets offer us is limited. Some performance completion data is self-reported by the training programs and is therefore subject to bias and miscoding. Because TPR is a novel dataset, we are limited to one point in time in our understanding of workforce development programs. In sum, with the current data available, it is nearly impossible to gauge relative effectiveness of ETPs or how they have changed over time.

I. The Landscape of Federally-Funded Training Providers

The landscape of federally-funded job training is very fragmented. This fragmentation is in part by design. WIOA is an open, consumer-choice system with fixed resources and a limited number of vouchers. Over 7,000 eligible providers offer more than 75,000 programs across the country (Figure 2). These programs vary in length, cost, and credential, and they prepare workers for over 700 unique occupations. We estimate that WIOA funding served roughly 220,000 training participants in PY2019. The fragmentation and small scale impede the targeting of particular industries, roles, or priorities at a state or national level. In addition, the fact that each program only enrolls a few WIOA participants per year makes it challenging to collect substantively useful program performance evaluations.

Figure 2: Landscape of WIOA Eligible Training Providers and Programs (PY 2019)

In terms of who WIOA serves, WIOA training participants are more likely to identify as Black than the average American (Table 1). Men and women are almost equally likely to participate in training through WIOA.

Table 1: WIOA Participant Demographics

Source: WIOA Individual Performance Records (PIRL) 2019.

ETPs are diverse in entity type. Most programs by share and by number of participants are offered at degree-granting institutions (Table 2). While individuals can also use funding from the Department of Education at many of these institutions (e.g. Pell Grants), they must have exhausted other grant assistance to receive WIOA funding. Thirty-five percent of WIOA recipients receive funding through formal higher education institutions. Fifteen percent of ETP programs are run by private providers, either for-profit or not-for-profit. Together, these providers serve almost as many WIOA recipients as higher education institutions, and private for-profit training providers serve nearly 20 percent of WIOA recipients. Apprenticeship programs appear to represent a very small share of ETP programs, though they are typically the most rigorous in terms of duration requirements. In general, WIOA dollars tend to go toward classroom learning environments rather than work-based learning environments.

Table 2: Eligible Training Provider Descriptive Statistics by Entity Type

Source: Project on Workforce analysis of

Note: ETP entity type data definitions are available in ETA-9171

ETP programs vary substantially in terms of credentials awarded. According to our analysis, the most common credential type associated with an ETP is an industry-recognized certificate or certification (21,000+ programs). In addition, almost 12,000 eligible programs produce an associate degree and another 10,000+ programs lead to a community college certificate. In the TPR data, nearly 10,000 eligible programs correspond only to employment or measurable skill gains, with no accompanying credential recorded.

ETP programs are also diverse in their training requirements. Median tuition fluctuates across entity type, with private providers tending to offer the most expensive program options. While national apprenticeship programs are the least expensive on average, they have the highest median program duration, requiring more than one year of training commitment. Private providers tend to offer programs that are expensive but short, while higher education institutions, public providers, and on-the-job learning providers offer programs that are longer and less expensive.

Programs are diverse in duration but tend to be shorter in length than traditional postsecondary degree programs. The median ETP program is 32 weeks long. Twenty-five percent of programs are 12 weeks or less. On the high end, 25 percent of programs are 64 weeks or longer.

ETPLs appear to vary in volume and diversity across different states (Figure 3). According to TPR, California offers the largest number of programs at nearly 5,000, and Florida, Illinois, Missouri, North Carolina, Ohio, Tennessee, and Washington offer over 3,000. New Jersey has the most concentrated program to participant ratio in TPR, with 152 reported eligible programs serving at least 15,000 participants. It is worth noting, however, that we observe at times substantial discrepancies between numbers of programs on the national TPR website and ETPLs on state websites. One potential explanation may be that states only decide to report to DOL program data for ETP programs with at least one WIOA participant enrolled in that program year, whereas their ETPL websites may list all approved programs eligible for enrollment.

Figure 3: WIOA Eligible Training Programs by State

Source: Project on Workforce analysis of

Though all states support many ETP programs, each individual training program tends to serve a relatively small number of WIOA participants (Table 3). Using the participant data available in TPR, the median program serves just 11 people funded by federal dollars. One-quarter of programs serve fewer than 7 people. However, participant enrollment data is missing for the majority of programs in TPR. An alternative metric is to compare the 75,000 TPR programs with almost 220,000 WIOA training participants nationwide in 2019. This calculation would suggest an even lower average number of participants served, about three WIOA participants per program. Figure 4 displays the full distribution of learners served in each program, suggesting that most WIOA-eligible training programs serve only a few participants funded by WIOA. While there is some variation across states in terms of WIOA enrollment per program figures, the trendline is fairly consistent. The small number of WIOA participants enrolled in each eligible program suggests that most training programs do not rely on WIOA as a primary source of funding. Other participants may be paying out of pocket or funded through other financing sources.

Table 3: Total WIOA Participants Served by ETP program

Source: Project on Workforce analysis of

Figure 4: Distribution of Eligible Training Programs by WIOA participants served

Source: Project on Workforce analysis of

II. What Jobs are We Training For?

While federal policy intends to support funding training for good, in-demand jobs, WIOA providers appear to have a mixed record. Overall, our analysis of the ETP program data implies that our public workforce development training dollars are prioritizing demand over job quality. Most (but not all) of the programs receiving WIOA funds are in occupations that are expected to grow in the next decade. We estimate that over half (54 percent) of ETP programs map to occupations that are expected to grow faster than the national average. However, many programs do not result in jobs that pay living wages. We find that the median ETP program corresponds to annual earnings of $29,388/year (extrapolating from quarterly wages).[20] Using individual data, we estimate that over 40 percent of WIOA training participants earn under $25,000 annually. This prompts consideration about whether and how policy should be revisited to better align with the legislation’s goals.

To assess the national landscape of ETP training by occupation type, we prioritize two criteria: earnings (job quality proxy) and growth (demand proxy). Certainly, other factors are worthy of consideration by states and regions in determining the types of jobs public money should support for training. However, we select these two criteria because workers consistently cite them as important in making decisions about training. Individuals pursuing training programs want to know that if they complete the program, there will consistently be jobs available in the field where they can take advantage of newly-acquired human capital. They also want to know that this position will pay wages that will pay more than jobs they otherwise could have obtained without the time and resources required to pursue training. We argue that these two criteria should be bare-minimum considerations for public policy in determining allocation of scarce resources in job training.

Educational requirements are a secondary factor we consider in our analysis. Since the stated objective of WIOA is workforce development, it is a reasonable assumption that the occupations targeted should not require a bachelor’s degree. Unfortunately, the current ETP data does not allow us to analyze with precision whether the training program is a good fit for shorter-term training. However, we use occupation information to draw inferences about whether the potential job role associated with the training does not require a bachelor’s degree.

To measure ETP program performance against these criteria, we make use of existing methodologies.

To understand job quality and educational requirements, we draw heavily on the Federal Reserve’s “Opportunity Occupations” research.[21] In this work, Federal Reserve researchers measure and define “the extent to which the U.S. economy offers decent-paying jobs to workers without a four-year college degree.”[22] They define “opportunity occupations” as occupations that typically require less than a bachelor's degree while paying more than the national median wage adjusted for local cost of living.[23]

We use the Bureau of Labor Statistics’ “Bright Outlook” Occupations as an indicator of employer demand. The BLS utilizes 2020-2030 nationwide employment projections to assess whether an occupation is projected to grow faster than average (employment increase of 10 percent or more) and is projected to have 100,000 or more job openings nationwide over 2020-2030.

Finally, we map programs to providers using SOC and Classification of Instructional Program (CIP) codes for each provider and training program. While imperfect, this crosswalk allows us to match the training programs in the TPR database to specific occupations that require the skills and knowledge provided through the program’s curriculum.

What types of jobs do eligible training providers train for?

Table 4 displays the ten most frequent fields of study and corresponding occupations for which ETP programs provide training, based on TPR data. While there is wide diversity in program types represented, healthcare fields are prominent, with medical assistant programs representing the largest share of ETP programs nationally in the available data. The top ten fields represent 24 percent of all ETP programs, and 29 percent of all training participants represented in TPR.

Table 4 : Occupational Focus of Eligible Training Programs

Source: Project on Workforce analysis of

While many programs serve only a few WIOA participants, certain types of providers seem to more consistently utilize WIOA funding (Table 5). Programs training for ground transportation, such as programs that provide Commercial Truck Driving Licenses, represent the largest share of programs with more than 50 WIOA participants. These programs enroll more than double the average number of WIOA participants, suggesting that they might rely more heavily on WIOA funding to operate. Programs in nursing, business administration, and allied health also train relatively larger numbers of WIOA participants.

Table 5: Five Most Common CIP Codes Among Programs with 50+ WIOA Participants

Source: Project on Workforce analysis of

Less than half of the ETP programs train for jobs considered top Opportunity Occupations nationally, even though Opportunity Occupations do not require a BA. Combining TPR data with the Federal Reserve’s List of the top 100 Opportunity Occupations, we find that 39 percent of all ETP programs (29,028 programs) provide training for occupations that are among the top 100 Opportunity Occupations. The most common programs that train for Opportunity Occupations are for welders, cutters, solderers, and brazers, medical records and health information technicians, and general and operations managers. Other top Opportunity Occupations trained for include nursing fields and heating, air conditioning, and refrigeration mechanics and installers.

Several other Opportunity Occupations are prevalent in TPR program data. Programs that train heavy and tractor-trailer truck drivers—the second-largest Opportunity Occupation nationally—are very common.[24] While these programs represent only 1.4 percent of ETP programs, they appear to be larger in size and therefore serve a substantial share of participants. In fact, programs that train for heavy and tractor-trailer truck drivers comprise 23 percent of all WIOA participants trained for Opportunity Occupations and 12 percent of training participants overall in the TPR data. Licensed vocational nursing and registered nursing, both classified as Opportunity Occupations, also appear to be common destinations for WIOA training enrollment.

However, many Opportunity Occupations are under-represented in ETPLs. For instance, some top 100 occupations are almost absent from the data. According to TPR, there are fewer than ten approved programs in the country to train workers for some Opportunity Occupations, despite their wages and prevalence (Table 6).

Table 6: Underrepresented Opportunity Occupations in WIOA Training

Sources: Project on Workforce analysis of; Bureau of Labor Statistics (2021).

Notably, some of the most common occupations in the training data are not considered Opportunity Occupations by the Federal Reserve’s definition because of low wages. There are nearly 2,500 programs nationally that train for medical assistants and over 2,000 programs that train nursing assistants. Medical assistants and nursing assistants appear to be in-demand jobs; according to the Bureau of Labor Statistics, they are growing much faster than average and as fast as average, respectively. However, they are typically low-paying. The median hourly wage for U.S. direct care workers (including home care workers, residential care aides, and nursing assistants) is $13.56, with median annual earnings of $20,200.[25] A 2020 analysis from PHI found that the median wage for direct care workers was lower than the median wage for other occupations with similar entry-level requirements, such as janitors, retail salespeople, and customer service representatives, in all 50 states and the District of Columbia. In fact, in 23 states and the District, direct care worker wages were lower than those for occupations with lower entry-level requirements (like housekeepers, groundskeepers, and food preparation workers).[26]

Other common programs may not be considered Opportunity Occupations because of educational requirements. Our analysis finds that a number of jobs– computer and information systems managers (2,321 programs nationally), preschool teachers (737 programs), accountants and auditors (745 programs), and computer programmers (1,087 programs) – are commonly approved for WIOA eligibility even though these fields typically require a college degree, according to the Bureau of Labor Statistics.[27] We don’t have enough information to assess whether these programs have been successful in landing WIOA participants into the desired field after completion. However, states should monitor carefully whether participants are successfully employed in their field of study.

What kinds of wages do eligible training programs lead to?

Overall, WIOA ETP programs prepare participants for jobs that pay below the US median income and also below median income for individuals without college degrees. We combine the TPR data with earnings data by occupational (SOC) code from WIOA Individual Performance Records.[28] Figure 5 presents a histogram of programs by earnings in the third quarter after completion. Median program earnings in the 3rd quarter after completion were $7,513. If we assume attendees earn the same in subsequent quarters, this corresponds to $29,388 annual earnings for the median program. Over 25 percent of program completers made under $6,000 per quarter – the bottom 25 percent of programs correspond with median earnings of $23,308 or below. For comparison, the median earnings for U.S. high school graduates without college in 2019 were $38,792, according to the Bureau of Labor Statistics. The median WIOA training program participant also earned less than the median worker with less than a high school diploma ($31,512).

Figure 5: Eligible Training Programs by Earnings in 3rd Quarter After Training Completion


**WIOA Individual Performance Records (PIRL) 2019

In addition, we find demographic disparities in the distribution of occupational earnings in the WIOA program data. Table 7 shows again the ten most common ETP programs nationally. Within these common programs, the two lowest-ranking programs by participant earnings also have some of the highest shares of female and non-white participants. Nursing assistant programs have some of the lowest earnings of any ETP program; their WIOA completers are 93 percent female and 57 percent non-white. In contrast, business and computer and information systems programs have some of the highest earnings for completers ($58,314 and $46,004 in the first four quarters, respectively), and they also are disproportionately male (58 percent of business programs and 64 percent of computer and information system manager programs, respectively) and white (72 percent and 55 percent, respectively).

These figures suggest that publicly-funded training programs may perpetuate occupational segregation. We find a strong relationship between the racial composition of the occupation and associated earnings. Occupations with higher shares of white training enrollees have higher wages. We observe a similar relationship between gender and wages. Fields of study with higher shares of male participants are associated with higher earnings for WIOA participants. These patterns also reinforce the important role that job centers and career counselors can play in providing useful guidance in the selection of programs and providers.

Table 7: Most Prevalent ETP Programs by Quarterly Earnings, Race, and Gender

Source: Project on Workforce analysis of; WIOA individual performance records (PIRL), Bureau of Labor Statistics.

Do eligible training programs focus on in-demand jobs?

Compared to earnings, we find that ETP programs perform better on labor market demand. The majority of ETP programs are providing training for jobs that are expected to grow in the coming decade. Combining TPR with data from the Department of Labor on projected employment change from 2020-2030 for different occupations by occupation code, Figure 6 shows the distribution of programs by their projected employment change.

Figure 6: Projected Employment Change for Eligible Training Programs

Source: Project on Workforce analysis of; Bureau of Labor Statistics.

However, while most of the top ETP program occupations are projected to grow as fast or faster than the national average, only two are categorized as growing rapidly, meaning they are categorized among the top 25 percent of growing occupations nationally (Table 8). In fact, there are over 10,000 programs (11 percent of all ETP programs) on state program lists that are training for occupations BLS expects to decline. The largest declines are for programs that train word processors and typists, data entry keyers, and legal secretaries. In addition, there are a number of high-growth occupations that don’t require a bachelor’s degree that appear to be under-represented on state program lists. For instance, there are fewer than five approved programs nationally that train for derrick and rotary drill operator roles for oil and gas, for forest fire inspector and prevention specialists, or for hearing aid specialists.

​​Table 8. Prevalent Eligible Training Programs by Demand Growth

Source: Project on Workforce analysis of; Bureau of Labor Statistics.

Interestingly, there is no evidence of an implied tradeoff between job quality and demand growth in occupations. Utilizing regression analysis, we find no relationship between projected demand growth for the occupation and levels of earnings. In fact, this may be an area for optimism. We identify 8,700 programs nationally (12 percent of all ETP programs) that train for occupations that are both Opportunity Occupations and “Bright Outlook” (high growth, high volume) occupations. These occupations and corresponding programs of study provide promising avenues for state policymakers to consider as they target future improvements to their eligible training provider lists. They include general and operations managers (1,200 eligible programs nationally), registered nurses (1,100 eligible programs nationally), and front-line supervisors of office and administrative support workers (590 programs).

The metrics that we study in this paper only scratch the surface of potential ways policymakers could assess program eligibility. For instance, long-term labor market demand predictions are notoriously unreliable. States and localities may wish to embed more real-time labor market information sources, including the National Labor Exchange (NLx) or private data providers like Lightcast. On job quality, states may wish to consider factors beyond just wages. For instance, formulas that include consideration for career advancement could provide more nuance around occupations like heavy and tractor-trailer truck driving (1,100 eligible programs nationally) that are fast-growing and pay above-median wages but may have limited upward mobility potential.

III. Challenges with Consumer Choice in Public Workforce Training

Because federal workforce training funding is designed to maximize “informed consumer choice,” a well-functioning system depends on good access to information. However, our research highlights that this is often a challenge in the public workforce system. We analyze both Department of Labor databases and state public employment and training websites to try to understand what information individuals can access to make decisions about WIOA-funded training. We identify a number of design and implementation challenges. At minimum, our observations suggest it is unlikely that the system is delivering optimal matches between training program and trainee.

First, the digital infrastructure intended to help individuals navigate eligible training programs is outdated and not user-friendly. States are required by law to post a public, electronic list of all eligible training programs that includes “appropriate information to assist participants in choosing employment and training activities.”[29] In reality, the websites that host these lists are often poorly-designed, infrequently updated, and hard to use. Digital interfaces vary substantially across states in accessibility and sophistication, and bugs were common. For six states, our researchers could only find downloadable spreadsheets or pdf lists. In addition, most states overwhelm their ETPL websites with information, seemingly catered to different audiences. In many cases, the websites appear to be designed for compliance purposes or workforce system staff rather than to help a jobseeker select a training program. For instance, in most states, the ETPL websites provide static lists that do not allow users to search or compare different programs on variables like occupation, cost, or credentials.

Box A. Example: UX challenges on state ETPL websites

Second, public-facing lists often lack information that is important for workers making training decisions. Polling shows that Americans consider a number of factors in assessing education options, including relevance, duration, cost, convenience, stackability, value, and career advancement potential.[30] Yet, most state lists do not include enough information for potential learners to make this judgment. Table 9 displays the number of states displaying key information fields on their ETPLs. The lack of information included on public-facing websites implies that the “self-service” option recommended in the law is functionally impractical in most states. Instead, training enrollees must rely on other sources of information, like advice from career center counselors, word of mouth from friends or family, or referrals from providers themselves.

Table 9: ETPL Information Fields by State

Source: Project on Workforce analysis of state ETPLs.

Third, earnings, employment, and other program performance and outcome data are rarely available. The Department of Labor’s TPR website, launched in 2020, is the first national data source that displays information at the provider and program levels. However, many states received waivers and therefore did not have to report program-level information.[31] In other cases, information about participants in programs serving fewer than 10 WIOA-sponsored participants is hidden in order to protect confidentiality.[32] The net effect in the first release is that metrics like participants served, completion rates, employment rates, and quarterly earnings are missing for between 75 and 99 percent of programs. For earnings metrics in particular, the share missing was over 95 percent.

Some state websites have more information about program and provider performance, but it is hard to discern whether the information is useful for potential training enrollees. Table 10 below displays the number of states displaying the WIOA performance indicators on their ETPL websites. It is worth noting, however, that performance information quality varies from state to state. Where some states connect earnings data from administrative tax records to WIOA participants, others rely on participants’ self-reported earnings. The methodology for these metrics is often hidden on ETPL websites. In addition, experts note that the diversity of populations served and program types represented in ETPLs makes it difficult to make apples to apples comparisons. Since approximately 95 percent of states do not include any information about the population groups the program serves, it is hard to evaluate programs’ relative success rates.

Table 10: Performance information available on state ETPL websites

*In the median earnings category, we have also included 12 states that reported on average wages rather than median wages.

Source: Project on Workforce analysis of state ETPLs.

Conclusion & Recommendations

Past research confirms that training investments for U.S. adults can work. Prior studies find substantial earnings impacts of sectoral employment programs like Year Up, Per Scholas, and Project QUEST, and occupational programs for high-demand fields at community colleges.[33] Many have also shown substantial return-on-investment for the public. In fact, research from the last decade has substantially strengthened the case for training as a source of economic mobility for U.S. workers.

However, the evidence on public workforce development training is mixed. Part of the challenge is that all previous evaluations of WIOA intended to measure the impact of the program in aggregate. Yet, we know that not all training providers and training programs are created equal – there is wide variation in instructional quality, organizational capacity, and corresponding employment opportunities. The true effect of the public investment probably depends largely on which training provider and program the participant selects.

ETP programs vary substantially by provider type, size, duration, and occupation. Our analysis also strongly suggests that the WIOA system alone does not provide a sustainable or scalable source of financing for most of the country’s job training providers, given the small numbers of eligible participants enrolled per program. In addition, we use descriptive data to provide examples of differences in training outcomes by gender, race, and field of study. Finally, we demonstrate that it is difficult or near-impossible for individuals to make informed decisions about enrollment in our current system. The plethora of options, outdated technology platforms, and lack of outcomes information makes it hard to choose training that will actually deliver on the promise of a good job.

Investment Priorities

First, we desperately need new technology investments to support the workforce development system. After the COVID-19 pandemic, there has been substantial focus and funding for technical upgrades to unemployment systems. Yet, few federal or state resources have supported technology that would improve training navigation. However, if the aspirations for individual “self-service” and “consumer choice” in WIOA are ever to be realized, we will need front-end and back-end investments to more accurately and accessibly present information about training providers. The subsequent section on implementation presents some priority areas for focus.

Second, we need greater human investments in career navigation. Given the constraints in the current information ecosystem, most workers rely on support from people to make decisions about training and job transitions. However, there is no dedicated funding in WIOA for professional development and career coaches in job centers are often woefully underpaid. Job seekers must first pass several eligibility hurdles to qualify for WIOA-funded individualized services rather than it being an entitlement. Funding for the Wagner-Peyser system, which serves the most people with career services but at very high volumes, has declined substantially in real terms in recent decades. We need more funding and capacity building for personnel if we are to have the best human capital to help Americans get access to good jobs.

Third, our review highlights the need for different public funding models for short-term training that go beyond consumer choice programs. This paper highlights the shortcomings of the current WIOA voucher system, where the flow of resources for human capital investment is often entirely dictated by individual enrollment decisions. This is suboptimal from an economic development perspective, both nationally and locally. With over 75,000 programs eligible, vouchers alone do not facilitate strategic investments by the government in priority sectors with fast-growing, good-paying jobs.

Federal job training policy should focus on making more funding available to directly support high-quality sectoral training programs, including wraparound supports. Policymakers should prioritize options that would boost federal funding for cohort-based sectoral training programs, including through WIOA reauthorization.

Complementary models could also include:

  • Competitive grant competitions that encourage cross-sector partnerships and support training investments for high priority roles (e.g. the Department of Commerce’s Good Jobs Challenge)

  • Direct appropriations to high-performing workforce development providers that allow them to recruit participants and scale their operations (e.g. Commonwealth of Massachusetts Legislature earmarks appropriations to JVS Boston, which has proven previous efficacy through a randomized control trial)

  • More financial aid for short-term, non-degree credentials from accredited, non-profit higher education providers (e.g. JOBS Act /expanded Pell grant eligibility). Virginia's FastForward program is a promising state model to fund enrollment in short-term community college programs in a select number of high-priority, high-growth fields, with a non-profit partner, VA Ready, that provides stipends.

  • Social impact bonds and other forms of performance-based financing (e.g. Massachusetts Pathways to Economic Advancement Pay for Success Project)

  • More flexibility to state and local entities to support direct contracting of strong providers (e.g. larger dislocated worker grants; H-1B grants, utilizing more flexible options in WIOA like the Governor’s Reserve, on-the-job training and customized training)

The U.S. workforce development system is persistently underfunded. If Congress is only willing to fund 220,000 learners per year at $2,000 per learner, training through the public workforce system will remain stuck in a low-resource, low-efficacy equilibrium.[34] At these funding levels, WIOA's vouchers won't come close to meeting the full scope of America’s human capital development and workforce needs.

Implementation Priorities

The need for new resources is substantial, but there are also ways to spend dollars more effectively in the current system. One clear area for improvement is data on workforce development provider performance. The WIOA legislation includes requirements for states to collect data about both individual participants (wage and employment metrics) and providers (numbers and types of learners, costs, and completion rates), but implementation has proved difficult, especially data collection at the provider level. As noted, performance information is missing for many programs due to state waivers and data suppression. In addition, some data we do have may be unreliable if it relies on providers or participants to self-report. Some states report that the process is burdensome for organizations responsible for the outreach.

To address this challenge, the U.S. needs better data systems to understand the impacts for participants in workforce development programs. While represents a valiant first effort, in the long-run, we should aim to create a national data source for workforce development that is comparable to higher education’s College Scorecard. Matching participant information to tax data from the IRS would create a more unified and trustworthy source of information to understand if training programs lead to higher wages and better career outcomes. It would also remove the administrative burden, allowing institutions that deliver training more time and resources to focus on what they do best. In the long-run, the federal or state government may also wish to pursue innovative new methods of automating data collection. For instance, government may be able to follow the money to improve performance information. Payments to training providers through the ITA vouchers could be used to understand how long participants stay in a training program and then aggregated publicly to communicate completion rates.

In the short-to-medium run, states can continue to prioritize and advance inter-agency data-sharing and data infrastructure initiatives. These agreements, which often involve partnerships with the K-12, higher education, tax agencies and/or unemployment agencies, can help us to better understand economic and educational outcomes for WIOA participants over time. Public-private initiatives like the Chamber of Commerce’s Jobs and Employment Data Exchange, which would improve data quality in employer reporting, also have promise.

Better program performance data matters because it can improve resource allocation. In a voucher system where money follows individual enrollment decisions, though, data can’t create better results on its own. Government must either provide better information (by making program performance information widely accessible and easy to understand for jobseekers) or improve the option set (by reforming eligibility criteria).

State websites share common challenges currently in conveying information. To date, most states rely primarily on third-party vendors. However, state agencies would greatly benefit from in-house technology talent, like product managers who could prioritize understanding the user journey from the worker perspective and improve front-end interfaces. In the future, we could also imagine the development of a common Application Programming Interface (API) that states could optionally adopt across state lines, to avoid duplication of technology development resources. New Jersey's Training Explorer provides a promising potential model.

Relatively simple improvements in data accessibility could also substantially reduce information barriers for workers and their families. For instance, states could consider showing median wages for occupations or programs in layman’s terms, like hourly or annual rather than quarterly. At the training program level, they could calculate and show completion rates, rather than stating the raw “number of individuals who exit.” They could consider using more inclusive and accessible language on public-facing pages to describe groups of workers, rather than relying on “barriers to employment” language from the WIOA statute. They could also collect and include information about whether the program is online or in-person, whether supports like childcare or transportation are available, and all-in costs to participate. “Better data” shouldn’t just mean more data for researchers or policymakers. Data needs to be translated in ways that are accessible or useful for the end user and ultimate customer.

On creating better lists, some states have already started the ambitious task of revisiting how they define and measure the “in-demand” requirement for eligible training provider lists. For instance, Virginia, Kentucky, New Jersey, and Pennsylvania have created formulas that quantify not just whether an occupation is truly “in-demand,” but also whether it pays good wages. Virginia has also incorporated whether the position has career advancement potential.

States should take seriously their responsibility to determine which occupations deserve public investment, and more states should revisit their criteria and/or reduce the number of eligible providers on their lists. However, the technical task of defining which types of jobs we want to support may actually be easier than implementing policy changes. It would require substantial political will to remove occupations like Certified Nursing Assistants and Tractor-Trailer Truck Drivers, which are in high demand from employers and have long been supported by WIOA dollars. Indeed, changes to eligible training provider lists will need to be part of a broader discussion about job quality in the country.

In any case, implementing reforms to ETPLs will likely require resources to build the capacity of state workforce agencies. Ideally, maintenance and updating of the lists would go beyond a quick website refresh. Ensuring quality means more frequent communication with local providers, including site visits, and is resource-intensive for states and localities. At minimum, the federal government could weigh in with new technical assistance on how states and local workforce development boards could improve their lists to prioritize job quality criteria. Workers would be better-served if more states treated the vetting of training partners as an important economic development decision, rather than as a compliance exercise.

In addition, the changing geography of work also poses challenges for how government decides which training is “in-demand.” Traditionally, the public workforce system has been entirely governed by the frame of the local or regional labor market. However, changing conditions on both the supply and demand side test the continued relevance of this approach. On the demand side, the rise of remote work means that individuals may wish to train for good-paying jobs that may not show up as “in-demand” locally but in fact are in-demand nationally. Current regulations do not support funding for this kind of worker.

On the supply side, the decentralization of the public workforce system poses unnecessary logistical challenges for good providers with a national footprint and potential to achieve scale. Getting approval across 50 state lists is administratively onerous and may actually dissuade proven, high-quality providers from participating in the system altogether. Similarly, at present, if an online community college program has good results, it has no way to easily get approval for residents living in another state to enroll. Given these changes in the economy, Congress and the Department of Labor should think seriously about new solutions that allow high-quality providers to become eligible in multiple states. States should also consider entering reciprocity agreements that can reduce bureaucratic hurdles for providers with proven quality results.

Finally, technocratic fixes alone should not overshadow the importance of focusing on the worker experience. Our analysis highlights that many design choices in WIOA, though likely intended to enhance accountability, have in practice created many barriers for workers to access the funding, choose a program with a high return, enroll in training, and successfully transition into a job in field of study. More resources are absolutely important, but so is a focus on how participants experience the system. As Congress considers WIOA reauthorization, they should seek out feedback from past users of training services and incorporate their experiences in new reforms.


Appendix A: ETPL Sources by State

Source: Project on Workforce analysis, June 2021

Appendix B: How do states list information about credentials on their eligible training provider list websites?

Source: Project on Workforce analysis of state websites

Appendix C: How do states list information about occupations on their eligible training provider list websites?

Information about target occupations is not required to be displayed on state ETPLs but is required to be reported by states to the Department of Labor.



Former Project on Workforce researchers Rachel Krust, Balaji Alwar, and Britteny Okorom-Achuonye provided critical contributions to this work, including collation of state eligible training provider lists and state website UI/UX analysis. Thank you to Nathalie Gazzaneo, Julian Hayes, and Jorge Encinas for their thoughtful questions and fact-checking support and to Isaiah Baldissera, for his design expertise. We are grateful for the support of Tessa Forshaw, our fearless co-leader of the Workforce Almanac initiative. Kellen Grode from the U.S. Department of Labor provided invaluable guidance on accessing and understanding ETP data. Finally, thank you to Harry Holzer, Marina Zhavoronkova, and Annelies Goger for their very helpful feedback and comments on earlier drafts.

The WorkRise Network has helped fund the Project on Workforce in conducting research on the U.S. workforce development sector.

Please direct inquiries to: Nathalie Gazzaneo

Suggested citation: David Deming, Alexis Gable, Rachel Lipson, and Arkādijs Zvaigzne. (March 2023). “Navigating Public Job Training.” Published by Harvard Kennedy School.

About the Project on Workforce at Harvard

The Project on Workforce is an interdisciplinary, collaborative project between the Harvard Kennedy School’s Malcolm Wiener Center for Social Policy, the Harvard Business School Managing the Future of Work Project, and the Harvard Graduate School of Education. The Project produces and catalyzes basic and applied research at the intersection of education and labor markets for leaders in business, education, and policy. The Project’s research aims to help shape a postsecondary system of the future that creates more and better pathways to economic mobility and forges smoother transitions between education and careers. Learn more at

About the Workforce Almanac

The Project on Workforce’s Workforce Almanac is a new effort to build a comprehensive new national data source and evidence base about the workforce development sector, including job training organizations and programs.

The views expressed in this report are the sole responsibility of the authors and are not meant to represent the views of Harvard University, the Harvard Kennedy School, or the Harvard Graduate School of Education.



[1] Congressional Research Service. "The Workforce Innovation and Opportunity Act and the One-Stop Delivery System." Updated September 26, 2022.

[2] "Highlights of the Trends in Student Aid 2021." College Board.

[4] Congressional Research Service. "The Workforce Innovation and Opportunity Act and the One-Stop Delivery System." Updated September 26, 2022.

[5] See Summary of Appropriation Budget Authority, Fiscal Year 2019, for the Employment and Training Administration here:

[6] Ibid.

[7] Calculations from the U.S. Department of Labor Employment and Training Administration “WIOA by the Numbers Interactive Data Analysis Tool.” In PY 2019, 147,286 people received WIOA assistance through the Adult Program, 61,926 people received assistance through the Dislocated Worker Program and 12,142 received training services through the National Dislocated Worker Grants. Calculations for training services from PY2019 WIOA Performance Reports, found

[8] Calculations from the U.S. Department of Labor Employment and Training Administration “WIOA by the Numbers Interactive Data Analysis Tool.” In PY 2019, 147,286 people received WIOA assistance through the Adult Program, 61,926 people received assistance through the Dislocated Worker Program and 12,142 received training services through the National Dislocated Worker Grants. Calculations for training services from PY2019 WIOA Performance Reports, found

[9] Credential Engine. 2022. Education and training expenditures in the U.S. Washington, DC: Credential Engine;

Congressional Research Service. "Federal Pell Grant Program of the Higher Education Act: Primer." Updated January 24, 2023.

[10] Chocolaad, Y. and Wandner, S. “Evidence-Building Capacity in State Workforce Agencies: Insights from a National Scan and Two State Site Visits.” NASWA 2017.

[11] Organisation for Economic Co-operation and Development, "Activation Policies."

[12] Eberts, Randall W. 2019. "Individual Training Accounts and Nonstandard Work Arrangements." Upjohn Institute Technical Report No. 19-037. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research.

[13] “Requirements to qualify on Mass ETPL: Massachusetts Eligibility Training Provider List Qualification.” MassHire Department of Career Services. Accessed February 10, 2023.

[14] Fortson, J., Navarro, J., & Thomas, M. (2018). WIOA Gold Standard Evaluation: Final Report (OPRE Report 2018-35). Office of Planning, Research, and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services. Retrieved from

[15] Fein, David and Samuel Dastrup. 2022. Benefits that Last: Long-Term Impact and Cost-Benefit Findings for Year Up. OPRE Report 2022-77. Washington, DC: Office of Planning, Research, and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services.

[16] Calculations from the U.S. Department of Labor Employment and Training Administration “WIOA by the Numbers Interactive Data Analysis Tool.” In PY 2019, 147,286 people received WIOA assistance through the Adult Program, and 61,926 people received assistance through the Dislocated Worker Program. Calculations for training services from PY2019 WIOA Performance Reports, found here.

[17] Eberts, Randall W. 2019. "Individual Training Accounts and Nonstandard Work Arrangements." Upjohn Institute Technical Report No. 19-037. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research.

[18] ETA-9171: Data Element Definitions/Instructions (Washington, D.C.: U.S. Department of Labor, Employment and Training Administration, 2022).

[19] Pitts, D. W., & Smith, K. 2019. Opportunity Occupations Revisited: Exploring Employment for Sub-Baccalaureate Workers Across Metro Areas and Over Time. Federal Reserve Bank of Cleveland. Retrieved from

[20] To calculate this rough metric, we estimated the annual earnings based on annualizing the average Q4 earnings for WIOA training participants who enrolled in programs that mapped to this occupational code, utilizing the Individual Performance records data.

[21] Wardrip, Keith, Kyle Fee, Lisa Nelson, and Stuart Andreason. 2015. "Identifying Opportunity Occupations in the Nation’s Largest Metropolitan Economies." Federal Reserve Banks of Philadelphia, Cleveland, and Atlanta.

Fee, Kyle, Keith Wardrip Lisa Nelson. 2019. “Opportunity Occupations Revisited: Exploring Employment for Sub-Baccalaureate Workers Across Metro Areas and Over Time.” Federal Reserve Banks of Philadelphia, Cleveland, and Atlanta.

[22] Ibid.

[23] Ibid. The wage requirement is calculated using median wage data from the Occupational Employment Statistics Program, which is then adjusted using Regional Price Parities. The educational requirement is estimated using data from the BLS’s Employment Projections Program, O*NET, and Burning Glass.

[25] "Direct Care Workers in the United States: Key Facts," PHI National, September 7, 2021.

[26] "Competitive Disadvantage: Direct Care Wages Are Lagging Behind." PHI National. October 10, 2020.

[27] "Occupational Projections and Characteristics," U.S. Bureau of Labor Statistics,

[28] This approach allowed us to overcome the suppression of data on a program level.

[29] For a full list of Eligible Training Provider List Requirements and Responsibilities, see:

[30] Strada Education Network. "Public Viewpoint: COVID-19 Work and Education Survey." Indianapolis, IN: Strada Education Network, August 26, 2020.

[33] Katz LF, Roth J, Hendra R, Schaberg K. Why Do Sectoral Employment Programs Work? Lessons from WorkAdvance. Journal of Labor Economics. 2022;40 (S1) :S249-S291.

[34] Holzer, Harry. “What Works In Workforce Development - And Making It Work Better” Forbes. February 27, 2023.


bottom of page