electricsheepafrica/africa-economic-impact-of-covid-19-in-sub-saharan-africa

---
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
- covid-19
- demographics
- eastern-africa
- economics
- employment
- geodata
- indicators
- civ
- ken
- moz
- nga
- zaf
pretty_name: "Economic Impact of COVID-19 in Sub-Saharan Africa"
dataset_info:
splits:
- name: train
num_examples: 1940
- name: test
num_examples: 485
---

# Economic Impact of COVID-19 in Sub-Saharan Africa

**Publisher:** Mobile Accord, Inc. (GeoPoll) · **Source:** [HDX](https://data.humdata.org/dataset/economic-impact-of-covid-19-in-sub-saharan-africa) · **License:** `cc-by` · **Updated:** 2025-09-26

---

## Abstract

This data looks at the impact of COVID-19 on employment, income, ability to pay expenses, and more in Côte D'Ivoire, Kenya, Mozambique Nigeria, and South Africa. Data is nationally representative by age, gender, and location, and is broken down by job type and formal or informal workers. Please contact us for data broken down by province or more information on the methodology.

Each row in this dataset represents country-level aggregates. Data was last updated on HDX on 2025-09-26. Geographic scope: **CIV, KEN, MOZ, NGA, ZAF**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Humanitarian and development data |
| **Unit of observation** | Country-level aggregates |
| **Rows (total)** | 2,426 |
| **Columns** | 27 (2 numeric, 25 categorical, 0 datetime) |
| **Train split** | 1,940 rows |
| **Test split** | 485 rows |
| **Geographic scope** | CIV, KEN, MOZ, NGA, ZAF |
| **Publisher** | Mobile Accord, Inc. (GeoPoll) |
| **HDX last updated** | 2025-09-26 |

---

## Variables

**Geographic** — `country` (Ivory Coast (Cote D'Ivoire), Kenya, South Africa), `employmenttype` (Unemployed, Employed full time, Student), `jobtype` (Educator, Small business owner/employee, Large business), `informal_work_type` (Seller, Agriculture, Other), `monthlyincome` (0-100000, 0-150000, 0-15000) and 11 others.

**Demographic** — `age_group` (36+, 26-35, 18-25), `gender` (Male, Female).

**Outcome / Measurement** — `incomechange`.

**Identifier / Metadata** — `aid`, `covidloans`, `esa_source`, `esa_processed`.

**Other** — `informalworker` (Yes, No, Not sure), `jobloss` (Yes, No-I’m still able to work, Prefer not to say), `jobregain` (Yes, Don't know, No), `lengthsurvival`.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-economic-impact-of-covid-19-in-sub-saharan-africa")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `country` | object | 0.0% | Ivory Coast (Cote D'Ivoire), Kenya, South Africa |
| `age_group` | object | 0.0% | 36+, 26-35, 18-25 |
| `gender` | object | 0.0% | Male, Female |
| `employmenttype` | object | 0.0% | Unemployed, Employed full time, Student |
| `jobtype` | object | 65.6% | Educator, Small business owner/employee, Large business |
| `informalworker` | object | 49.5% | Yes, No, Not sure |
| `informal_work_type` | object | 73.5% | Seller, Agriculture, Other |
| `jobloss` | object | 46.9% | Yes, No-I’m still able to work, Prefer not to say |
| `jobregain` | object | 68.6% | Yes, Don't know, No |
| `monthlyincome` | object | 46.9% | 0-100000, 0-150000, 0-15000 |
| `monthlyincome_bracket` | float64 | 47.2% | 1.0 – 11.0 (mean 1.6846) |
| `incomechange` | object | 47.0% | |
| `expenseresponsibility` | object | 0.0% | |
| `lengthsurvival` | object | 21.8% | |
| `moneyforexpenses` | object | 21.8% | |
| `concernexpenses` | object | 21.8% | |
| `expense_concern_rating` | float64 | 21.8% | 1.0 – 5.0 (mean 3.9612) |
| `monthlyneed` | object | 20.5% | |
| `toppriority` | object | 19.0% | |
| `lowpriority` | object | 19.0% | |
| `aid` | object | 0.0% | |
| `covidloans` | object | 0.0% | |
| `mobilemoneyactivity` | object | 0.0% | |
| `mobilemoneydeposit` | object | 0.0% | |
| `governmentpriority` | object | 0.0% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `monthlyincome_bracket` | 1.0 | 11.0 | 1.6846 | 1.0 |
| `expense_concern_rating` | 1.0 | 5.0 | 3.9612 | 5.0 |

---

## 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`. 8 column(s) with >80% missing values were removed: `otherjob`, `aidsource_aid_organisations`, `aidsource_charities_donations`, `aidsource_friends_family`, `aidsource_government`, `aidsource_not_sure`.... 74 exact duplicate rows were removed. 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 Mobile Accord, Inc. (GeoPoll) 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: `jobtype`, `informalworker`, `informal_work_type`, `jobloss`, `jobregain`, `monthlyincome`, `monthlyincome_bracket`, `incomechange`....
- This dataset spans 5 countries; geographic and methodological inconsistencies across national boundaries may affect cross-country comparability.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/economic-impact-of-covid-19-in-sub-saharan-africa) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_economic_impact_of_covid_19_in_sub_saharan_africa,
title = {Economic Impact of COVID-19 in Sub-Saharan Africa},
author = {Mobile Accord, Inc. (GeoPoll)},
year = {2025},
url = {https://data.humdata.org/dataset/economic-impact-of-covid-19-in-sub-saharan-africa},
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.*

annotations_creators:
- 无标注（no-annotation）
language_creators:
- 公开采集数据（found）
language:
- 英语（en）
license: 知识共享署名4.0国际许可协议（CC BY 4.0）
multilinguality:
- 单语言（monolingual）
size_categories:
- 1000<n<10000条数据
source_datasets:
- 原创数据集（original）
task_categories:
- 表格回归（tabular-regression）
- 其他（other）
task_ids: []
tags:
- 非洲（africa）
- 人道主义（humanitarian）
- HDX
- electric-sheep-africa
- COVID-19
- 人口统计（demographics）
- 东非（eastern-africa）
- 经济（economics）
- 就业（employment）
- 地理数据（geodata）
- 指标（indicators）
- civ
- ken
- moz
- nga
- zaf
pretty_name: "撒哈拉以南非洲地区COVID-19的经济影响"
dataset_info:
splits:
- name: train
num_examples: 1940
- name: test
num_examples: 485

---

# 撒哈拉以南非洲地区COVID-19的经济影响

**发布方：** Mobile Accord, Inc.（GeoPoll） · **来源：** [HDX（人道主义数据交换平台）](https://data.humdata.org/dataset/economic-impact-of-covid-19-in-sub-saharan-africa) · **许可协议：** `CC BY` · **最后更新：** 2025年9月26日

---

## 摘要

本数据集聚焦COVID-19对科特迪瓦、肯尼亚、莫桑比克、尼日利亚及南非的就业、收入、支付开支能力等多维度的影响。数据按年龄、性别与地域实现全国代表性抽样，并按职业类型、正式/非正式就业群体进行细分。如需获取按省份细分的数据或更多方法论细节，请联系我们。

本数据集每一行代表国家级汇总数据。数据最后一次在HDX平台更新于2025年9月26日。地理覆盖范围：**CIV、KEN、MOZ、NGA、ZAF**。

*本数据集经[Electric Sheep Africa](https://huggingface.co/electricsheepafrica)整理为适配机器学习的Parquet格式。*

---

## 数据集特征

| | |
|---|---|
| **领域** | 人道主义与发展数据 |
| **观测单元** | 国家级汇总数据 |
| **总数据行数** | 2426 |
| **列数** | 27列（2个数值型、25个分类型、0个日期时间型） |
| **训练集划分** | 1940行 |
| **测试集划分** | 485行 |
| **地理覆盖范围** | CIV、KEN、MOZ、NGA、ZAF |
| **发布方** | Mobile Accord, Inc.（GeoPoll） |
| **HDX平台最后更新时间** | 2025年9月26日 |

---

## 变量

**地理变量** — `country`（科特迪瓦（Cote D'Ivoire）、肯尼亚、南非）、`employmenttype`（失业、全职就业、学生）、`jobtype`（教育工作者、小企业主/雇员、大型企业）、`informal_work_type`（商贩、农业从业者、其他）、`monthlyincome`（0-100000、0-150000、0-15000）等共11个附加变量。

**人口统计变量** — `age_group`（36岁及以上、26-35岁、18-25岁）、`gender`（男性、女性）。

**结果/测量变量** — `incomechange`（收入变化）。

**标识符/元数据变量** — `aid`、`covidloans`、`esa_source`、`esa_processed`。

**其他变量** — `informalworker`（是、否、不确定）、`jobloss`（是、否——我仍能工作、不愿透露）、`jobregain`（是、不知道、否）、`lengthsurvival`（生存时长）。

---

## 快速上手

python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-economic-impact-of-covid-19-in-sub-saharan-africa")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()

---

## 数据架构

| 列名 | 数据类型 | 缺失率 | 取值范围/示例值 |
|---|---|---|---|
| `country` | 字符串（object） | 0.0% | 科特迪瓦（Cote D'Ivoire）、肯尼亚、南非 |
| `age_group` | 字符串 | 0.0% | 36岁及以上、26-35岁、18-25岁 |
| `gender` | 字符串 | 0.0% | 男性、女性 |
| `employmenttype` | 字符串 | 0.0% | 失业、全职就业、学生 |
| `jobtype` | 字符串 | 65.6% | 教育工作者、小企业主/雇员、大型企业 |
| `informalworker` | 字符串 | 49.5% | 是、否、不确定 |
| `informal_work_type` | 字符串 | 73.5% | 商贩、农业从业者、其他 |
| `jobloss` | 字符串 | 46.9% | 是、否——我仍能工作、不愿透露 |
| `jobregain` | 字符串 | 68.6% | 是、不知道、否 |
| `monthlyincome` | 字符串 | 46.9% | 0-100000、0-150000、0-15000 |
| `monthlyincome_bracket` | 浮点数（float64） | 47.2% | 1.0 – 11.0（均值1.6846） |
| `incomechange` | 字符串 | 47.0% | 无 |
| `expenseresponsibility` | 字符串 | 0.0% | 无 |
| `lengthsurvival` | 字符串 | 21.8% | 无 |
| `moneyforexpenses` | 字符串 | 21.8% | 无 |
| `concernexpenses` | 字符串 | 21.8% | 无 |
| `expense_concern_rating` | 浮点数 | 21.8% | 1.0 – 5.0（均值3.9612） |
| `monthlyneed` | 字符串 | 20.5% | 无 |
| `toppriority` | 字符串 | 19.0% | 无 |
| `lowpriority` | 字符串 | 19.0% | 无 |
| `aid` | 字符串 | 0.0% | 无 |
| `covidloans` | 字符串 | 0.0% | 无 |
| `mobilemoneyactivity` | 字符串 | 0.0% | 无 |
| `mobilemoneydeposit` | 字符串 | 0.0% | 无 |
| `governmentpriority` | 字符串 | 0.0% | 无 |
| `esa_source` | 字符串 | 0.0% | 无 |
| `esa_processed` | 字符串 | 0.0% | 无 |

---

## 数值型变量摘要

| 列名 | 最小值 | 最大值 | 均值 | 中位数 |
|---|---|---|---|---|
| `monthlyincome_bracket` | 1.0 | 11.0 | 1.6846 | 1.0 |
| `expense_concern_rating` | 1.0 | 5.0 | 3.9612 | 5.0 |

---

## 数据整理

原始数据通过CKAN API从HDX平台下载，并转换为Parquet格式。列名统一转换为小写并标准化为蛇形命名法（snake_case）。将常见缺失值标记（`N/A`、`null`、`none`、`-`、`unknown`、`no data`、`#N/A`）统一替换为`NaN`。删除了8个缺失值占比超过80%的列：`otherjob`、`aidsource_aid_organisations`、`aidsource_charities_donations`、`aidsource_friends_family`、`aidsource_government`、`aidsource_not_sure`……删除了74条完全重复的行。使用固定随机种子（42）以80/20的比例划分为训练集与测试集，并保存为Snappy压缩的Parquet格式。

---

## 局限性

- 数据源自Mobile Accord, Inc.（GeoPoll），未经过Electric Sheep Africa的独立验证。
- 自动化清洗流程无法修正原始数据收集中的错报值、定义不一致或抽样偏差问题。
- 以下列的缺失值占比超过20%，在建模时需谨慎使用：`jobtype`、`informalworker`、`informal_work_type`、`jobloss`、`jobregain`、`monthlyincome`、`monthlyincome_bracket`、`incomechange`……
- 本数据集覆盖5个国家，各国间的地理与方法学差异可能影响跨国数据的可比性。
- 如需获取发布方提供的方法论说明与注意事项，请参阅[原始HDX数据集页面](https://data.humdata.org/dataset/economic-impact-of-covid-19-in-sub-saharan-africa)。

---

## 引用

以下为该数据集的标准引用格式：
bibtex
@dataset{hdx_africa_economic_impact_of_covid_19_in_sub_saharan_africa,
title = {Economic Impact of COVID-19 in Sub-Saharan Africa},
author = {Mobile Accord, Inc. (GeoPoll)},
year = {2025},
url = {https://data.humdata.org/dataset/economic-impact-of-covid-19-in-sub-saharan-africa},
note = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — 非洲的机器学习数据集基础设施。尼日利亚拉各斯。*