electricsheepafrica/africa-zimbabwe-health-facilities

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- 1K<n<10K
source_datasets:
- original
task_categories:
- tabular-classification
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- facilities-infrastructure
- health
- health-facilities
- hxl
- zwe
pretty_name: "Zimbabwe: Health facilities"
dataset_info:
splits:
- name: train
num_examples: 1352
- name: test
num_examples: 338
---

# Zimbabwe: Health facilities

**Publisher:** OCHA Regional Office for Southern and Eastern Africa (ROSEA) · **Source:** [HDX](https://data.humdata.org/dataset/zimbabwe-health-facilities) · **License:** `cc-by` · **Updated:** 2025-11-13

---

## Abstract

List of Health facilities in Zimbabwe

Each row in this dataset represents subnational administrative unit observations. Data was last updated on HDX on 2025-11-13. Geographic scope: **ZWE**.

*Curated into ML-ready Parquet format by [Electric Sheep Africa](https://huggingface.co/electricsheepafrica).*

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Public health |
| **Unit of observation** | Subnational administrative unit observations |
| **Rows (total)** | 1,690 |
| **Columns** | 43 (37 numeric, 6 categorical, 0 datetime) |
| **Train split** | 1,352 rows |
| **Test split** | 338 rows |
| **Geographic scope** | ZWE |
| **Publisher** | OCHA Regional Office for Southern and Eastern Africa (ROSEA) |
| **HDX last updated** | 2025-11-13 |

---

## Variables

**Geographic** — `province` (Manicaland, Mashonaland East, Midlands), `district` (Mutasa, Chipinge, Mutare), `longitude` (range 25.8258–33.0345), `latitude` (range -22.3173–-15.7006), `yearbuilt` (range 0.0–2011.0) and 2 others.

**Temporal** — `updated` (range 1998.0–2013.0).

**Identifier / Metadata** — `id1` (range 1.0–1686.0), `id` (range 0.0–1685.0), `nameoffaci` (ZRP, Chivi, Shamva), `esa_source` (HDX), `esa_processed` (2026-04-18).

**Other** — `elevation` (range 0.0–1569.0), `ownership` (range 0.0–9.0), `numofdocto` (range 0.0–3.0), `numofnurse` (range 0.0–41.0), `numofnur_1` (range 0.0–21.0) and 25 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-zimbabwe-health-facilities")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()
```

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `id1` | float64 | 0.1% | 1.0 – 1686.0 (mean 845.1445) |
| `id` | float64 | 2.0% | 0.0 – 1685.0 (mean 662.1087) |
| `province` | object | 0.0% | Manicaland, Mashonaland East, Midlands |
| `district` | object | 0.0% | Mutasa, Chipinge, Mutare |
| `longitude` | float64 | 0.1% | 25.8258 – 33.0345 (mean 30.5756) |
| `latitude` | float64 | 0.1% | -22.3173 – -15.7006 (mean -18.8716) |
| `elevation` | float64 | 0.9% | 0.0 – 1569.0 (mean 87.397) |
| `updated` | float64 | 0.9% | 1998.0 – 2013.0 (mean 2001.2042) |
| `nameoffaci` | object | 0.1% | ZRP, Chivi, Shamva |
| `ownership` | float64 | 10.7% | 0.0 – 9.0 (mean 5.0199) |
| `yearbuilt` | float64 | 4.7% | 0.0 – 2011.0 (mean 297.0981) |
| `typeoffaci` | object | 1.0% | Clinic, Rural Health Centre, Council Clinic |
| `numofdocto` | float64 | 4.0% | 0.0 – 3.0 (mean 0.0265) |
| `numofnurse` | float64 | 4.0% | 0.0 – 41.0 (mean 0.4418) |
| `numofnur_1` | float64 | 4.0% | 0.0 – 21.0 (mean 0.244) |
| `numofpcn` | float64 | 4.0% | 0.0 – 39.0 (mean 0.1041) |
| `numofehts` | float64 | 4.0% | 0.0 – 4.0 (mean 0.0444) |
| `numofpharm` | float64 | 4.0% | 0.0 – 2.0 (mean 0.0105) |
| `numoflabte` | float64 | 4.0% | 0.0 – 5.0 (mean 0.008) |
| `numofbeds` | float64 | 4.0% | 0.0 – 2500.0 (mean 2.6402) |
| `numofmater` | float64 | 4.0% | 0.0 – 20.0 (mean 0.1763) |
| `numofgener` | float64 | 4.0% | 0.0 – 140.0 (mean 0.6245) |
| `cathmentpo` | float64 | 4.0% | 0.0 – 1616300.0 (mean 2154.8712) |
| `distneares` | float64 | 4.0% | 0.0 – 565.0 (mean 5.4855) |
| `hascommuni` | float64 | 4.0% | |
| `hascommu_1` | float64 | 4.0% | |
| `haswaterpi` | float64 | 4.0% | |
| `haswaterun` | float64 | 4.0% | |
| `haselectri` | float64 | 4.0% | |
| `haselect_1` | float64 | 4.0% | |
| `distnear_1` | float64 | 4.0% | |
| `hassanitat` | float64 | 4.0% | |
| `hassanit_1` | float64 | 4.0% | |
| `hassanit_2` | float64 | 4.0% | |
| `hassecurit` | float64 | 4.0% | |
| `hassecur_1` | float64 | 4.0% | |
| `hasroadtar` | float64 | 4.0% | |
| `hasroadgra` | float64 | 4.0% | |
| `hasinciner` | float64 | 4.0% | |
| `hasautoway` | float64 | 4.1% | |
| `hasdental` | float64 | 4.0% | |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-18 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `id1` | 1.0 | 1686.0 | 845.1445 | 845.0 |
| `id` | 0.0 | 1685.0 | 662.1087 | 652.5 |
| `longitude` | 25.8258 | 33.0345 | 30.5756 | 30.8722 |
| `latitude` | -22.3173 | -15.7006 | -18.8716 | -18.7445 |
| `elevation` | 0.0 | 1569.0 | 87.397 | 0.0 |
| `updated` | 1998.0 | 2013.0 | 2001.2042 | 1998.0 |
| `ownership` | 0.0 | 9.0 | 5.0199 | 5.0 |
| `yearbuilt` | 0.0 | 2011.0 | 297.0981 | 0.0 |
| `numofdocto` | 0.0 | 3.0 | 0.0265 | 0.0 |
| `numofnurse` | 0.0 | 41.0 | 0.4418 | 0.0 |
| `numofnur_1` | 0.0 | 21.0 | 0.244 | 0.0 |
| `numofpcn` | 0.0 | 39.0 | 0.1041 | 0.0 |
| `numofehts` | 0.0 | 4.0 | 0.0444 | 0.0 |
| `numofpharm` | 0.0 | 2.0 | 0.0105 | 0.0 |
| `numoflabte` | 0.0 | 5.0 | 0.008 | 0.0 |

---

## Curation

Raw data was downloaded from HDX via the CKAN API and converted to Parquet. Column names were lowercased and standardised to snake_case. Common missing-value markers (`N/A`, `null`, `none`, `-`, `unknown`, `no data`, `#N/A`) were unified to `NaN`. 2 column(s) with >80% missing values were removed: `comments`, `type_edite`. 4 column(s) were cast from string to numeric or datetime based on parse-success rate (>85% threshold). The dataset was split 80/20 into train and test partitions using a fixed random seed (42) and saved as Snappy-compressed Parquet.

---

## Limitations

- Data originates from OCHA Regional Office for Southern and Eastern Africa (ROSEA) and has not been independently validated by ESA.
- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/zimbabwe-health-facilities) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_zimbabwe_health_facilities,
title = {Zimbabwe: Health facilities},
author = {OCHA Regional Office for Southern and Eastern Africa (ROSEA)},
year = {2025},
url = {https://data.humdata.org/dataset/zimbabwe-health-facilities},
note = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}
```

---

*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — Africa's ML dataset infrastructure. Lagos, Nigeria.*