electricsheepafrica/africa-wfp-food-security-indicators-for-sierra-leone

---
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-regression
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- food-security
- hxl
- indicators
- sle
pretty_name: "Sierra Leone - Food Security Indicators"
dataset_info:
splits:
- name: train
num_examples: 3396
- name: test
num_examples: 849
---

# Sierra Leone - Food Security Indicators

**Publisher:** WFP - World Food Programme · **Source:** [HDX](https://data.humdata.org/dataset/wfp-food-security-indicators-for-sierra-leone) · **License:** `cc-by-igo` · **Updated:** 2024-09-13

---

## Abstract

The World Food Programme (WFP) launched the mobile Vulnerability Analysis and Mapping (mVAM) project in 2013, beginning in DRC and Somalia. mVAM uses mobile technology to track food security trends in real-time, providing high-frequency data that supports humanitarian decision-making. Data collection methods are tailored to the needs of each country that mVAM operates in. This dataset contains data from the [mVAM databank](http://vam.wfp.org/sites/mvam_monitoring/) covering various indicators (one per resource).

Each row in this dataset represents time-series observations. Temporal coverage is indicated by the `svydate` column(s). Geographic scope: **SLE**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Food security and nutrition |
| **Unit of observation** | Time-series observations |
| **Rows (total)** | 4,246 |
| **Columns** | 9 (1 numeric, 7 categorical, 1 datetime) |
| **Train split** | 3,396 rows |
| **Test split** | 849 rows |
| **Geographic scope** | SLE |
| **Publisher** | WFP - World Food Programme |
| **HDX last updated** | 2024-09-13 |

---

## Variables

**Geographic** — `svydate`, `adminstrata` (Sierra Leone, Bo, Bombali-Koinadugu-Tonkolili).

**Identifier / Metadata** — `adm0_name` (Sierra Leone, #country+name), `esa_source` (HDX), `esa_processed` (2026-04-08).

**Other** — `variable` (rCSI>=1, RestrictConsumption>=1, BorrowOrHelp>=1), `variabledescription` (prevalence-->equals to 1 if household uses this strategy 1 or more times per week, # of days household using this coping strategy per week, reduced coping strategy), `demographic` (Shared flush toilet, Own flush toilet, Cement pit latrine), `mean` (range -0.1522–14423.3333).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-wfp-food-security-indicators-for-sierra-leone")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `svydate` | datetime64[ns] | 0.0% | |
| `adm0_name` | object | 0.0% | Sierra Leone, #country+name |
| `adminstrata` | object | 43.4% | Sierra Leone, Bo, Bombali-Koinadugu-Tonkolili |
| `variable` | object | 0.0% | rCSI>=1, RestrictConsumption>=1, BorrowOrHelp>=1 |
| `variabledescription` | object | 5.7% | prevalence-->equals to 1 if household uses this strategy 1 or more times per week, # of days household using this coping strategy per week, reduced coping strategy |
| `demographic` | object | 56.6% | Shared flush toilet, Own flush toilet, Cement pit latrine |
| `mean` | float64 | 0.0% | -0.1522 – 14423.3333 (mean 865.57) |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-08 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `mean` | -0.1522 | 14423.3333 | 865.57 | 1.3833 |

---

## 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: `adm1_name`, `adm2_name`. 2 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 WFP - World Food Programme and has not been independently validated by ESA.
- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
- The following columns have >20% missing values and should be treated with caution in modelling: `adminstrata`, `demographic`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/wfp-food-security-indicators-for-sierra-leone) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_wfp_food_security_indicators_for_sierra_leone,
title = {Sierra Leone - Food Security Indicators},
author = {WFP - World Food Programme},
year = {2024},
url = {https://data.humdata.org/dataset/wfp-food-security-indicators-for-sierra-leone},
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.*