Program Evaluation Journey

Sharpen the research question, choose the sampling strategy, and calculate the sample size before you collect anything. The five-stage design ends with an analysis plan and a pilot plan, which fix who you sample and which outcome measures every later step works from.

Enter three or more pre-treatment periods for your treatment group and your control group. You get a pre-trend slope gap with a t-test, a period-by-period event-study table, and an in-time placebo that reruns the estimate at a fake treatment date. The treatment and control series that pass here are the ones step 3 stratifies.

Paste your stratified 2x2 outcome table. You get per-stratum odds ratios, the crude pooled odds ratio, and the Mantel-Haenszel adjusted odds ratio, with an explicit reversal flag when the aggregate and stratified stories disagree. A pooled effect that survives stratification is the credible placement signal step 4 adjusts.

Enter the program's total cost, enrollment, and gross employed-at-exit count, then set a baseline counterfactual job-finding rate, a substitution rate, and a creaming discount. The output is a haircut-adjusted placement count and cost-per-outcome figure alongside the unadjusted gross numbers, piped straight into the cost ladder.

Enter cost, enrollment, placement, and retention data for up to three programs to produce the full DOL TEGL 6-13 cost ladder: Cost per Participant, Cost per Employed Exiter, Cost per Retained Employee, and Cost per Individual with Earnings Increase, plus benchmark flags against the Massachusetts $7,500 outcome cap and the roughly $26/employed-day national CPED benchmark, and a funder-facing one-pager.

Enter your program's cost and outcome data, including enrollment, exits, and the reported WIOA performance outcomes. The report benchmarks against WIOA PY 2023 national negotiated performance targets, calculates cost-per-outcome metrics, models ROI, and exports a WIOA-formatted Policy Mandate for the board packet.