This dataset is provided by Kasi Insight Inc. for non-commercial teaching, learning and academic research only. By requesting access you agree not to use it for commercial purposes, not to redistribute or re-share the data (in original or modified form), and not to attempt to re-identify any respondent.

Kasi Academia African Consumer Surveys 2024

Anonymised respondent-level data from three online consumer surveys run by Kasi Academia (Kasi Insight Inc.) across African markets in 2024.

The files are shared as exported from the survey platform, without cleaning, for teaching and practising data cleaning and data analysis, as well as for academic research. Expect to explore, assess and prepare the data before analysing it.

Survey	Fieldwork	File	Respondents	Language of questionnaire
Climate change & consumer sentiment	Feb–Mar 2024	`climate_change_feb2024/data_enfr.csv`	10,351	English / French
		`climate_change_feb2024/data_pt.csv`	1,101	Portuguese
Share of wallet (household spending)	May–Jun 2024	`share_of_wallet_may2024/data_en.csv`	6,177	English
		`share_of_wallet_may2024/data_fr.csv`	4,583	French
		`share_of_wallet_may2024/data_pt.csv`	1,026	Portuguese
Share of wallet (household spending)	Sep–Oct 2024	`share_of_wallet_sep2024/data_en.csv`	2,784	English
		`share_of_wallet_sep2024/data_fr.csv`	1,136	French

Load one file with:

from datasets import load_dataset
ds = load_dataset("KasiInsight/Kasi-Academia-Data", "share_of_wallet_may2024_en")

File layout

One row per respondent; columns are the survey questions (full question text as header).
respondent_id – random ID, unique within a file. It is not linkable across files or waves.
survey_date – date the response was started (YYYY-MM-DD).
duration_minutes – time taken to complete the survey.
For multi-select questions each answer option is its own column.
Questions are routed by country, so many country-specific columns (e.g. city, income bands in local currency) are blank for respondents from other countries.

Anonymisation

Applied before release:

Removed: IP address, panel/recruiter respondent code, platform response number, serial number.
Exact start/end timestamps replaced by survey_date (day only) and duration_minutes.
Rows shuffled and given new random IDs.
Email addresses and phone numbers typed into free-text answers replaced with [REDACTED_EMAIL] / [REDACTED_PHONE].

Demographics are kept only as categories (gender, birth-year bracket, occupation, household size, country, city/region, income band). Free-text "other (please specify)" answers are kept, with email addresses and phone numbers removed.

Terms of use

This dataset is not released under an open licence. See LICENSE. In short:

Non-commercial teaching, learning and academic research only. No commercial use of any kind.
No redistribution. Do not share, re-upload or publish the data, in original or modified form, including on GitHub, Kaggle or course websites. Classmates and collaborators should request their own access here.
No re-identification. Do not attempt to identify or contact any respondent.
Code, notebooks, assignments, and aggregate results, tables and figures derived from the data may be shared or published, with citation, as long as they do not contain respondent-level records.

For commercial licensing or other uses, contact Kasi Insight Inc. at info@kasiinsight.com.

Citation

If you use this dataset, please cite:

Kasi Insight Inc. (2026). Kasi Academia African Consumer Surveys 2024 [Data set]. Hugging Face. https://huggingface.co/datasets/KasiInsight/Kasi-Academia-Data

BibTeX:

@misc{kasiinsight2026consumer,
  author       = {{Kasi Insight Inc.}},
  title        = {Kasi Academia African Consumer Surveys 2024},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/KasiInsight/Kasi-Academia-Data}}
}

Downloads last month: 34

Total file size:

82.8 MB