WIOA training provider outcomes scraper - US ETP programs
Pricing
from $5.00 / 1,000 program records
WIOA training provider outcomes scraper - US ETP programs
WIOA training provider outcomes for 70,000+ US programs from the Department of Labor's Training Provider Results index (ETA-9171): provider, program, credential, SOC codes, completion and Q2 employment rates, median earnings, costs. Filter by state, SOC code and employment rate. $5 per 1,000 rows.
Pricing
from $5.00 / 1,000 program records
Rating
0.0
(0)
Developer
RetrainMap Team
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
5 hours ago
Last modified
Categories
Share
Returns: one row per US training program from the DOL Training Provider Results index (70,000+): provider, program, credential, SOC codes, completion and Q2 employment rates, median earnings, cost, length — nulls where DOL suppressed a value.
Input: { "states": ["NC"], "soc_codes": ["31-9092"], "min_q2_employment_pct": 60, "max_records": 500 } — every field optional; empty {} reads the whole index (~2.5 minutes).
Price: $5.00 per 1,000 rows ($0.005 per program row) plus $0.10 per run. Pay per event; no subscription.
One row per training program from the U.S. Department of Labor's Training Provider Results index: the states' WIOA Eligible Training Provider performance reports (form ETA-9171, OMB 1205-0526) that power https://www.trainingproviderresults.gov/. 70,088 programs on 2026-09-02. Filter by state, SOC occupation code and the reported second-quarter employment rate; get provider, program, credential, occupation codes, length, exits, completion and employment rates, median earnings and costs — normalised, with DOL's suppression markers turned into nulls and listed per row.
Nothing in this dataset decides anyone's eligibility. Eligibility for WIOA-funded training (and for a given program) is decided locally by the state and the local workforce development board through an American Job Center. Use the rows to compare programs, then confirm locally.
Data source and status
| Publisher | U.S. Department of Labor, Employment and Training Administration |
| Site | https://www.trainingproviderresults.gov/ (About page: "Data on this website reflect training programs approved to be on states' ETP list as of June 30, 2025, covering the period from July 1, 2021, through June 30, 2025.") |
| Bulk file of record | https://www.trainingproviderresults.gov/data/DownloadPrograms.xlsx (data dictionary v4.0: https://www.trainingproviderresults.gov/assets/ETP_Data_Dictionary.pdf) |
| What this Actor reads | The site's search index, https://cxsearch.dol.gov/etp/etp_scorecard_programs — plain GET requests, keyset-paged on the document id (nid), 500 documents per request, at most one request per second |
| Refresh cadence | DOL publishes the ETP reports once a year (program-year data; the current release is PY 2024, posted spring 2026). The Actor reads live on every run, so a new release appears in the rows automatically; the row field snapshot_timestamp is the fetch time |
| Copyright | Works of the United States Government are not subject to copyright in the U.S. (17 U.S.C. §105). The data are public; please cite DOL as the source |
| Known caveats (DOL's own) | "States with known data quality concerns include: Alabama, Colorado, District of Columbia, Illinois, Louisiana, New Jersey, New York, and Puerto Rico." Oklahoma had not submitted PY 2024 data. Values reported as -1 are suppressed (small sample, not reported or quality issue) — the Actor writes them as null and lists the field names in suppressed_fields |
Identification: the DOL web application firewall answers 403 to any User-Agent that
carries a product token, so the request uses the plain Python-urllib client token and
carries the operator's contact address in the standard From: header.
Input
| Field | Type | Meaning |
|---|---|---|
states | array of 2-letter codes | Optional. Empty = the whole index (~145 requests, ~2.5 minutes) |
soc_codes | array of strings | Optional. 31-9092 (also 31-9092.00, 319092); a prefix such as 31- matches the major group. Matched against all three occupation codes a program reports |
min_q2_employment_pct | integer 0–100 | Optional. Drops programs below this Q2 employment rate (all participants) and programs whose rate is suppressed |
max_records | integer | Default 1000. Stop after this many matching rows |
request_interval_ms | integer | Default 1000 (floor 500) |
contact_email | string | Sent in the From: header |
use_local_snapshot_path | string | Testing only: read a local programs_<date>.jsonl snapshot instead of the index (no network) |
Example — Medical Assistants in North Carolina with at least a 60 % Q2 employment rate:
{ "states": ["NC"], "soc_codes": ["31-9092"], "min_q2_employment_pct": 60, "max_records": 500 }
Output (dataset row)
provider, provider_ref, program, program_id, program_uuid, city, state, zip,
address, lat, lon, reporting_state, credential, entity_type, program_format,
has_prerequisites, soc_codes (O*NET-SOC spelling, e.g. 31-9092.00), cip_code,
length_weeks, length_hours, total_served, exited, completed, credential_attained,
employed_q2, employed_q4, completed_pct, employed_q2_pct, median_earnings_q2,
avg_ita_cost (average cost per WIOA participant), non_wioa_tuition, non_wioa_supplies,
wioa_served, wioa_exiters, wioa_completed_pct, wioa_employed_q2_pct,
wioa_credential_pct, program_url, description (first 400 characters),
suppressed_fields, snapshot_timestamp, source.
Percentages are 0–100 (the index stores fractions). A run summary (counts, filters, request
count, the index's own count for the requested states) is stored as RUN_SUMMARY in the
run's key-value store.
Pricing (pay per event)
| Event | Price |
|---|---|
run-start — once per run | $0.10 |
program-record — per row written | $0.005 |
A 500-row state/occupation pull costs $2.60; the whole index (70,088 rows) $350.54. Rows stop when your run's maximum charge is reached; the run summary says so.
Operator
Steelyard Ventures LLC (RetrainMap) — info@steelyardclinical.com. The Actor writes only to its own dataset and key-value store; it stores no credentials and sends nothing else.