Image

Program Evaluation

Program evaluation is a way of assessing a program to better understand its implementation and measure outcomes, and ultimately improve its operation or outcomes. There are many different types of or methods for program evaluation used in workforce programs:

  • Experimental evaluation, often called randomized controlled trial evaluation, compares outcomes of “treatment” groups (those in the program or receiving services) and “control” groups (often those receiving whatever services would exist in the absence of the program) to analyze the impact of the program. This often requires “randomly” selecting whether a person participates in the program or not. This type of evaluation is considered the “gold standard,” as one can fully attribute effects on participants to the program. However, it can be resource-intensive to implement and may not be appropriate for all programs. 
  • Quasi-experimental evaluation uses statistical methods to assess the effects of programs, sometimes using methods to select a plausible comparison group not in the program (for example, synthetic comparison groups or propensity score matching). This type of evaluation may not be as resource-intensive as experimental design, but it may not be as rigorous. 
  • Outcomes evaluation describes the outcomes of program participants, without a control or comparison group. While not as rigorous as experimental or quasi-experimental evaluation, it can be useful for understanding how well participants did in the program, especially if it is a newer program. 
  • Implementation evaluation uses qualitative methods (including interviews, focus groups, and observational study) to understand the operation of a program. This is sometimes used to assess the fidelity of the program implementation to the intended program model.
  • Return on investment or cost-benefit analyses are ways of using the results of evaluation with information on costs to compare the value of program outcomes relative to the cost of producing those outcomes.