Evaluating the Impact of Project Diakonia’s Low Barrier Workforce Development Program on At-Risk Youth in Colorado

The intervention is operated by Project Diakonia, a workforce development organization in Colorado that provides construction skills training and comprehensive wraparound support to marginalized youth. The intervention consists of offering eligible individuals a place in Project Diakonia’s low barrier workforce development program, which is designed to “screen in” youth who are often excluded from similar training programs due to perceived instability or risk.

Participants assigned to the treatment group are offered access to an eight-week construction skills course that meets three times per week in four-hour sessions. In addition to technical skills training, treatment participants receive individualized case management, supportive services, mentoring, and career navigation support. After completing the skills course, they continue to receive employment preparation, job search assistance, and up to twelve months of post-program follow-up to help them stabilize, pursue longer-term goals, and secure employment. Some participants may also be matched to a paid work experience with Project Diakonia or Habitat for Humanity, depending on readiness and availability.

Participants assigned to the control group will not be offered access to the skills course or Project Diakonia’s wraparound services. Instead, they will receive referrals to community resources, including the Pikes Peak Workforce Center and other local services related to employment, food security, and basic needs. All other individuals not enrolled in the study will continue to move through Project Diakonia’s usual intake and enrollment process.

2026-02-02

2029-03-30

Using administrative data sources, the research team will examine earnings, employment, justice system involvement, and educational attainment as the primary outcomes for this study. The research team will explore and specify constructions of these outcomes over the course of the pilot period.

Precise measures of earnings, labor market attachment, justice system involvement, and educational attainment will be developed during the course of a pilot phase of the study.

For secondary outcomes, the research team will analyze housing stability, public benefits receipt, engagement with other support services, financial stability (e.g. credit access and well-being), and psychosocial measures (e.g. hope, grit, self-confidence), and overall well-being (e.g. mental and physical health). The research team will explore and specify constructions of these outcomes during the pilot period.

Precise measures of the above outcomes will be developed during the course of a pilot phase of the study.

This study uses a randomized controlled trial to evaluate the impact of Project Diakonia’s Low Barrier Workforce Development Program on youth ages 17 to 24 who face significant barriers to employment and stability. The main research questions are: (1) does participation in Project Diakonia reduce criminal justice involvement? (2) Does the program improve labor market outcomes, including employment and earnings?, and (3) does the program increase educational attainment? The research team will first conduct a pilot RCT with a sample size of 300 participants (150 treatment, 150 control). Once the pilot period has concluded, the research team will update this registry to specify the analysis for the main RCT, which will enroll up to 700 additional participants (350 treatment, 350 control).

Youth will be eligible for the study if they are between ages 17 and 24, are enrolled in an alternative high school or have dropped out of high school, are referred by a recognized community partner, and meet at least one risk factor such as homelessness, justice involvement, foster care involvement, or unemployment and underemployment. Eligible youth will be identified through Project Diakonia’s existing referral and intake process. After completing intake, youth will be randomly assigned to either a treatment group or a control group. Randomization will occur in regular batches aligned with program operations.

Youth assigned to the treatment group will be offered enrollment in Project Diakonia’s eight week construction skills training course and will have access to case management, mentoring, supportive services, and employment supports. Youth assigned to the control group will receive referrals to other community resources.

For the analysis, we will compare outcomes across treatment and control groups and explore heterogeneity across pre-specified sub-groups to be determined during the course of our pilot.

We will also examine heterogeneous treatment effects using baseline characteristics; pre-specified sub-groups will be determined during the course of the pilot phase based on preliminary results.

These subgroup analyses will help identify whether the program is differentially effective for youth facing varying levels and types of barriers at baseline. Together, the experimental design and planned heterogeneity analyses will provide a comprehensive understanding of how a low-barrier workforce model can best support at-risk young adults.

Not available

Participants will be randomized by the research team using a computer generated random assignment procedure. Project Diakonia will provide a list of all eligible youth who completed intake during the relevant intake window, and the research team will assign each individual to either the treatment group or the control group using a reproducible, pre-specified randomization code. Randomization will occur in regular batches that align with program operations.

Unit of randomization: Individual

No

Up to 1,000 individuals, 300 of which will be enrolled during the pilot period.

Up to 1,000 individuals, 300 of which will be enrolled during the pilot period.

The pilot RCT will enroll 300 individuals, assigning 150 to the treatment group and 150 to the control group. The primary RCT will enroll up to 700 individuals, assigning 350 to the treatment group and 350 to the control group. Combined, this means the study will enroll up to 1,000 individuals, assigning 500 to the treatment group and 500 to the control group.

Power calculations assume individual level randomization with no clustering and a total sample of 1,000 youth (500 assigned to treatment and 500 to control). Using a one sided 5 percent significance level and 80 percent power, and assuming a baseline quarterly employment rate of 45 percent with a standard deviation of 0.50, the study is powered to detect a minimum effect of 7.8 percentage points in quarterly employment, which corresponds to a 17.5 percent increase relative to the control mean. For quarterly earnings, assuming a control group mean of 1,165 dollars and a standard deviation of about 2,200 dollars, the design is powered to detect a minimum effect of 346 dollars per quarter, or about a 30 percent increase relative to the control mean. These values will be updated as baseline data on the Project Diakonia population become available.