# EDCOM II early-childhood dashboard data, cleaned (version 2)

Data behind the NASH Reviewer series on the CRLA at the end of SY 2025-26. It was extracted from the public EDCOM II
early-childhood care and development dashboard (<https://edcom2.gov.ph/eccddashboard/>), checked and cleaned, and
joined to DepEd's public school data. NASH Reviewer is not affiliated with EDCOM II or the Department of Education.

- **Every file, with the report and charts:** [EDCOM II CRLA data](https://nash.com.ph/edcom-data)
- **How it was made:** [How the EDCOM Dashboard Data Was Cleaned](https://nash.com.ph/blog/edcom-eccd-dashboard-data-cleaned)
- **What it shows:** [CRLA Results at the End of SY 2025-26](https://nash.com.ph/blog/crla-end-of-year-results-sy-2025-2026)
- **Reading school results fairly:** [How to Read a School's CRLA Results](https://nash.com.ph/blog/how-to-read-school-crla-results)

## Files

| File | Contents | Rows | Size |
|---|---|---:|---:|
| [`edcom_raw_extract_csv.zip`](https://data.nash.com.ph/edcom-data/v2/edcom_raw_extract_csv.zip) | The six tables as recovered from the dashboard workbook, unchanged, with the field dictionary and metadata |  | 9.1 MB |
| [`schools_literacy_clean.csv.gz`](https://data.nash.com.ph/edcom-data/v2/schools_literacy_clean.csv.gz) | One row per school: CRLA results by grade and language, barangay link, coordinates, flags | 38,679 | 4.2 MB |
| [`school_locations_clean.csv.gz`](https://data.nash.com.ph/edcom-data/v2/school_locations_clean.csv.gz) | All school location records with geography and coordinate quality | 48,254 | 3.4 MB |
| [`barangays_clean.csv.gz`](https://data.nash.com.ph/edcom-data/v2/barangays_clean.csv.gz) | PSGC barangay reference with match keys and centroids | 42,010 | 1.6 MB |
| [`barangays_without_cdc_clean.csv.gz`](https://data.nash.com.ph/edcom-data/v2/barangays_without_cdc_clean.csv.gz) | Barangays with no child development center (DILG CBMS, 17 Sep 2025) | 4,604 | 35 KB |
| [`stunting_2025_clean.csv.gz`](https://data.nash.com.ph/edcom-data/v2/stunting_2025_clean.csv.gz) | OPT Plus 2025 stunting, children under 2 (0-23 months): normal, stunted and classified counts, by region, province and municipality (DOH/NNC, 2 Mar 2026) | 1,742 | 131 KB |
| [`barangay_programs_clean.csv.gz`](https://data.nash.com.ph/edcom-data/v2/barangay_programs_clean.csv.gz) | Literacy, nutrition and childcare program records (100) | 100 | 6 KB |
| [`overview_by_region_clean.csv.gz`](https://data.nash.com.ph/edcom-data/v2/overview_by_region_clean.csv.gz) | The dashboard's regional overview | 19 | 1 KB |
| [`region_crosswalk.csv.gz`](https://data.nash.com.ph/edcom-data/v2/region_crosswalk.csv.gz) | Region labels to PSGC regions | 18 | 1 KB |
| [`region_reconciliation.csv.gz`](https://data.nash.com.ph/edcom-data/v2/region_reconciliation.csv.gz) | Regional totals reconciled with the national totals | 19 | 1 KB |
| [`cleaning_log.csv`](https://data.nash.com.ph/edcom-data/v2/cleaning_log.csv) | Every change the cleaning made: table, record, rule, old and new value, reason | 1,524 | 231 KB |
| [`validation.json`](https://data.nash.com.ph/edcom-data/v2/validation.json) | Row counts, reconciliation checks and link summary |  | 2 KB |
| [`edcom_crla_2026.parquet`](https://data.nash.com.ph/edcom-data/v2/edcom_crla_2026.parquet) | Analysis table: one row per school, CRLA blocks and totals, barangay and municipality codes, SY 2023-24 school inputs (in_ columns), head type | 38,679 | 8.4 MB |
| [`division_estimates.csv`](https://data.nash.com.ph/edcom-data/v2/division_estimates.csv) | 232 divisions: share struggling, empirical Bayes estimate and 95% interval, rank interval, funnel and FDR flags | 232 | 46 KB |
| [`school_estimates.csv.gz`](https://data.nash.com.ph/edcom-data/v2/school_estimates.csv.gz) | Every school: share struggling, empirical Bayes and hierarchical Bayes estimates with 95% intervals, funnel flag (read the guide before ranking schools) | 38,679 | 2.9 MB |
| [`inputs_by_region.csv`](https://data.nash.com.ph/edcom-data/v2/inputs_by_region.csv) | School inputs and struggling readers: odds ratios with 95% intervals from a separate model for each region | 230 | 27 KB |
| [`municipalities.csv`](https://data.nash.com.ph/edcom-data/v2/municipalities.csv) | Municipality or city: learners, struggling readers, estimate, LISA and Gi* results, stunting and CDC gaps | 1,639 | 447 KB |

CSV files are UTF-8; `.csv.gz` files are gzip-compressed CSV. Checksums are in [`CHECKSUMS.sha256`](https://data.nash.com.ph/edcom-data/v2/CHECKSUMS.sha256).

## Definitions

- **Period.** CRLA results at the end of SY 2025-26, as of 10 May 2026, for 38,679 public schools.
- **Learners.** The dashboard's totals are Grade 1 Mother Tongue + Grade 2 Filipino + Grade 3 English: one result per
  learner, 5,127,670 learners in all.
- **Struggling readers.** Low Emerging plus High Emerging: 511,753 learners (9.98%).
- **Regions.** The 18 PSGC regions (`region_clean`), with Sulu in Region IX and the Negros Island Region as its own region.
- **Stunting.** Stunted and severely stunted children under 2 (0-23 months) as a share of all children under 2 classified
  as normal, stunted or severely stunted: `stunted_age_0_23m / classified_age_0_23m`. Nationally that is 4.1%: 129,823
  of 3,138,473 children in all 1,642 cities and municipalities, or 129,690 of 3,137,385 in the 1,639 that the blog
  analyzes. The age columns for children found tall are not in the dashboard's extract, so they are not in
  `classified_age_0_23m`. `normal_age_0_23m` is the EDCOM dashboard's own denominator ("Actual_Children (0-2)"); its
  4.3% is stunted / normal, a ratio rather than a share.

## Changes in version 2 (13 November 2026)

**Stunting.** Version 1 named the count of children of normal height `children_assessed_age_0_2` ("Children assessed
age 0-2") and divided stunted children by it, as the EDCOM dashboard does. That count leaves out the stunted children,
so the share was stunted / normal. In version 2:

- `stunting_2025_clean.csv.gz`: `children_assessed_age_0_2` becomes `normal_age_0_23m`, `stunted_age_0_2` becomes
  `stunted_age_0_23m`, and `classified_age_0_23m` (normal + stunted + severely stunted) is new. The file holds counts.
- `overview_by_region_clean.csv.gz`: "Children assessed age 0-2", "Stunted children age 0-2" and "Stunting % age 0-2"
  become "Normal-height children age 0-23m", "Stunted children age 0-23m" and "Stunting % age 0-23m" (stunted /
  classified); "Children classified age 0-23m" and "Dashboard stunting ratio % (stunted / normal)" are new.
- `region_reconciliation.csv.gz`: the "Children assessed age 0-2" and "Stunted children age 0-2" columns are renamed
  to their 0-23m names; their values are unchanged.
- `municipalities.csv`: the `stunting` column is now stunted / classified, with the new count columns beside it.
- `edcom_raw_extract_csv.zip`: the reconstructed `overview_by_region.csv` and the README carry the same names.

National stunting moves from 4.3% to 4.1%; every area keeps its rank.

**Barangay links.** Version 2 also carries the hand-checked barangay links of 28 October 2026, which version 1
predates: 22 assessed schools (38 location records) change from `link_method` `name` to `manual` in
`schools_literacy_clean.csv.gz` and `school_locations_clean.csv.gz`, with their linked barangay, municipality and
province, distance to the barangay center and coordinate quality. `cleaning_log.csv` gains 38 rows (1,486 to 1,524)
and `validation.json` updates its link counts. The analysis files (`edcom_crla_2026.parquet`, `municipalities.csv`,
`school_estimates.csv.gz`, `division_estimates.csv`, `inputs_by_region.csv`) were built before these links and keep
the version 1 links, as the blog's analyses do.

No other value changed. Version 1 stays at <https://data.nash.com.ph/edcom-data/v1/>; see
[the correction](https://nash.com.ph/blog/stunting-day-care-and-reading#correction).

## Sources and credit

The data come from the EDCOM II dashboard, which compiles DepEd's CRLA results, DOH/NNC's Operation Timbang Plus 2025,
DILG's Community-Based Monitoring System and a list of program records. School inputs come from DepEd's public school
files for SY 2023-24. Municipal boundaries used in the analysis are from PSA and NAMRIA via OCHA HDX (CC BY-IGO).

Please cite: Llego, M. A. (2026). *EDCOM II early-childhood dashboard data, cleaned* (Version 2) [Data set]. NASH Reviewer.
https://nash.com.ph/edcom-data

## Cautions

- School-level shares from small schools are noisy. Use the estimates and intervals in `school_estimates.csv.gz`,
  not raw shares, and do not rank individual schools from them.
- Links and coordinates added by research are marked in `cleaning_log.csv`; researched coordinates are less precise
  than DepEd's (a median of 5.2 km from the barangay center against 0.73 km).
- Stunting (children under 2, 2025) and reading (Grades 1-3, SY 2025-26) describe different children. Comparisons between them
  are about places, not individual children.
