electricsheepafrica/africa-food-consumption-score-fcs

Food Consumption Score (FCS)

Publisher: WFP - World Food Programme · Source: HDX · License: cc-by-igo · Updated: 2024-05-30

---

Abstract

The Food Consumption Score (FCS) dataset is based on the FCS indicator, which assigns a food security score based on food consumption and diets. This data is available sub-nationally for 38 countries, such as Nepal and Sierra Leone.

Each row in this dataset represents time-series observations. Data was last updated on HDX on 2024-05-30. Geographic scope: ARM, BGD, BEN, BFA, BDI, CMR, CAF, TCD, and 30 others.

Curated into ML-ready Parquet format by Electric Sheep Africa.

---

Dataset Characteristics

Domain	Food security and nutrition
Unit of observation	Time-series observations
Rows (total)	885
Columns	14 (6 numeric, 8 categorical, 0 datetime)
Train split	708 rows
Test split	177 rows
Geographic scope	ARM, BGD, BEN, BFA, BDI, CMR, CAF, TCD, and 30 others
Publisher	WFP - World Food Programme
HDX last updated	2024-05-30

---

Variables

Geographic — fcs_year (range 2006.0–2015.0), methodology (Raised Threshold: 28/42, Standard Threshold: 21/35, Non-Standard Threshold (please specify)), indicator_type (Baseline, Monitoring, Emergency).

Temporal — fcs_month (range 1.0–12.0).

Identifier / Metadata — adm0_name (Kyrgyzstan, Sudan, Tajikistan), adm1_name (Northern Darfur, Western Darfur, Southern Darfur), fcs_datasource (FSMS, WFP VAM Comprehensive food security and vulnerability analysis (CFSVA) and nutrition assessment, 2012, WFP VAM Comprehensive Food Security and Vulnerability Analysis and Nutrition Survey (CFSVA), 2012), esa_source (HDX), esa_processed (2026-04-15).

Other — fcs_mean (range 16.7–88.58), fcs_poor (range 0.0–73.5294), fcs_borderline (range 0.0–75.4601), fcs_acceptable (range 2.9412–100.0), target_group (Total Population by Administrative Units, Male Headed HHs, Female Headed HHs).

---

Quick Start

from datasets import load_dataset

ds    = load_dataset("electricsheepafrica/africa-food-consumption-score-fcs")
train = ds["train"].to_pandas()
test  = ds["test"].to_pandas()

print(train.shape)
train.head()

---

Schema

Column	Type	Null %	Range / Sample Values
adm0_name	object	0.0%	Kyrgyzstan, Sudan, Tajikistan
adm1_name	object	14.0%	Northern Darfur, Western Darfur, Southern Darfur
fcs_mean	float64	54.7%	16.7 – 88.58 (mean 60.0362)
fcs_poor	float64	0.0%	0.0 – 73.5294 (mean 8.109)
fcs_borderline	float64	0.0%	0.0 – 75.4601 (mean 18.8136)
fcs_acceptable	float64	0.0%	2.9412 – 100.0 (mean 73.0774)
target_group	object	0.0%	Total Population by Administrative Units, Male Headed HHs, Female Headed HHs
fcs_month	int64	0.0%	1.0 – 12.0 (mean 6.4712)
fcs_year	int64	0.0%	2006.0 – 2015.0 (mean 2010.4395)
methodology	object	0.0%	Raised Threshold: 28/42, Standard Threshold: 21/35, Non-Standard Threshold (please specify)
fcs_datasource	object	0.0%	FSMS, WFP VAM Comprehensive food security and vulnerability analysis (CFSVA) and nutrition assessment, 2012, WFP VAM Comprehensive Food Security and Vulnerability Analysis and Nutrition Survey (CFSVA), 2012
indicator_type	object	0.0%	Baseline, Monitoring, Emergency
esa_source	object	0.0%	HDX
esa_processed	object	0.0%	2026-04-15

---

Numeric Summary

Column	Min	Max	Mean	Median
fcs_mean	16.7	88.58	60.0362	59.44
fcs_poor	0.0	73.5294	8.109	4.447
fcs_borderline	0.0	75.4601	18.8136	17.2
fcs_acceptable	2.9412	100.0	73.0774	75.2252
fcs_month	1.0	12.0	6.4712	6.0
fcs_year	2006.0	2015.0	2010.4395	2010.0

---

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. 3 column(s) with >80% missing values were removed: adm2_name, fcs_lowerthreshold, fcs_upperthreshold. 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: fcs_mean.
• This dataset spans 38 countries; geographic and methodological inconsistencies across national boundaries may affect cross-country comparability.
• Refer to the original HDX dataset page for the publisher's own methodology notes and caveats.

---

Citation

@dataset{hdx_africa_food_consumption_score_fcs,
  title     = {Food Consumption Score (FCS)},
  author    = {WFP - World Food Programme},
  year      = {2024},
  url       = {https://data.humdata.org/dataset/food-consumption-score-fcs},
  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

Electric Sheep Africa — Africa's ML dataset infrastructure. Lagos, Nigeria.