Health, attractiveness, and marriageability among Aka hunter-gatherer and Ngandu farmer adolescents and young adults in the Central African Republic

Hewlett, Bonnie 1 ; Hewlett, Barry1; Agrawal, Harshita 1

Published Aug 28, 2025; Updated Nov 10, 2025 on Dryad. https://doi.org/10.5061/dryad.r4xgxd2r9

Data files

Aug 28, 2025 version files 387.93 KB

README.md

8.20 KB
SubmissionKey.zip

379.73 KB

Nov 10, 2025 version files 251.93 KB

README.md

8.41 KB
SubmissionKey.zip

243.52 KB

Abstract

This study investigates how adolescents from two culturally distinct Central African communities—the Aka hunter-gatherers and Ngandu farmers—perceive health, attractiveness, and marriageability, focusing on how local socio-ecological contexts shape perceptions of health, attraction, and mate preferences. The research examines how body mass index and parasite load relate to perceived health and attractiveness. Seventy-five adolescents (39 Aka and 36 Ngandu) evaluated photographs of peers for health, attractiveness, and marriagability, while physical health was objectively measured using anthropometric and parasitological data. Regression analyses revealed that BMI was a consistent and significant predictor of both perceived health and attractiveness, whereas total parasite load did not significantly influence either outcome. Notably, perceptions of health strongly predicted attractiveness and marriageability, but not vice versa. Ethnic and gender differences in perception were also found: Aka adolescents prioritized social traits like kindness and cooperation, while Ngandu participants emphasized physical cleanliness and robustness. Male raters showed more variability and stricter standards, particularly when evaluating females. These findings underscore the role of robust phenotypic features, such as BMI, in determining how health, beauty, and reproductive value are interpreted. This research contributes to understanding how evolutionarily evolved preferences and cultural contexts shape perceptions of health, attraction, and marriageability.

Dataset DOI: 10.5061/dryad.r4xgxd2r9

Description of the data and file structure

This study employed a mixed-methods design to explore perceptions of health, attractiveness, and marriageability among Aka hunter-gatherer and Ngandu farmer adolescents in the Central African Republic. A sample of 75 participants completed pile-sorting tasks involving photographs of same- and opposite-sex peers, categorizing them by perceived health, attractiveness, and marriageability. These evaluations were supplemented by open-ended interviews, free-listing exercises, and anthropometric measures, including BMI and body fat percentage. Parasitological data were collected via stool samples and analyzed for prevalence of 13 intestinal parasites. The goal was to examine the relative influence of physical health indicators—particularly BMI and parasite load—on local mate preferences across distinct cultural contexts.

Files and variables

Key to: Health, Attractiveness, and Marriageability Variable Description

Unless explicitly stated, all the variables below are used in the `Final_Data_for_HealthAttraction`sheet of the data file.

1. Demographic & Biological Variables

Variable	Description	Type
ID	Unique participant identifier.	Numeric
Eth	Ethnicity (Aka or Ngandu Tribe).	Categorical (Binary: Aka / Ngandu)
Sex	Biological sex of the participant	Binary (Male/Female).
BMI	Body Mass Index (weight in kg / height in m²).	Continuous

(The above variables and the PathNP variable are used in all sheets of the data file.)

2. Pathogenic Load Variables

Variable	Description	Type
Path	Pathogenic exposure	Numeric
PathNP	Total pathogenic load	Numeric
PathEtc	Union of pathogenic parasites and ectoparasites	Numeric
NP	Non-pathogenic	Numeric

3. Perceived Health & Attractiveness (Rated by Others)

Variable	Description	Type
PHeal	Proportion of health perception (rated by both sexes toward an individual).	Numeric (0–1)
Pattr	Proportion of attractiveness perception (rated by both sexes toward an individual).	Numeric (0–1)
FHeal	Proportion of health perception (rated toward a female individual by both sexes).	Numeric (0–1)
MHeal	Proportion of health perception (rated toward a male individual by both sexes).	Numeric (0–1)
Fattr	Proportion of attractiveness perception (rated toward a female individual by both sexes).	Numeric (0–1)
Mattr	Proportion of attractiveness perception (rated toward a male individual by both sexes).	Numeric (0–1)

4. Marriageability Perceptions (Rated by Opposite Sex)

Variable	Description	Type
Mmarr	Proportion of marriageability perception (rated by females toward a male individual). (This variable can be found in the `MarriageM` sheet of the data file)	Numeric (0–1)
Fmarr	Proportion of marriageability perception (rated by males toward a female individual). (This variable can be found in the `MarriageF` sheet of the data file)	Numeric (0–1)

5. Transformed Variables (S&V Transformation)

Note: S&V - (Smithson & Verkuilen, 2006)

Variable	Description	Type
PhealT	Transformed (S&V) version of PHeal.	Numeric (0–1)
PattrT	Transformed (S&V) version of Pattr.	Numeric (0–1)
FhealT	Transformed (S&V) version of FHeal.	Numeric (0–1)
MhealT	Transformed (S&V) version of MHeal.	Numeric (0–1)
FattrT	Transformed (S&V) version of Fattr.	Numeric (0–1)
MattrT	Transformed (S&V) version of Mattrr.	Numeric (0–1)
PMarrM	Transformed (S&V) version of Mmarr. (This variable can be found in the `MarriageM`sheet of the data file)	Numeric (0–1)
PMarrF	Transformed (S&V) version of Fmarr. (This variable can be found in the `MarriageF` sheet of the data file)	Numeric (0–1)

The 0-1 value indicates the following:

PhealT: A value of 0 represents less healthiness, whereas a value of 1 indicates the healthiest condition. This applies to FhealT and MhealT as well.
PattrT: A value of 0 signifies less attractiveness, while a value of 1 represents the highest level of attractiveness. The same applies to FattrT and MattrT.
PMarrM: A value of 0 indicates a lower likelihood of being considered for marriage, whereas a value of 1 suggests a higher likelihood of being preferred for marriage. This also applies to PMarrF.

Code/software

Analysis Workflow

All statistical analyses were conducted in RStudio using R Markdown for reproducibility.

Code Location: First, unzip the SubmissionKey.zip folder. The full analysis scripts are available in the SubmissionKey/Code/ folder within the SubmissionKey directory.

File 1: Health_and_Attraction_Analysis.Rmd - Analysis of the Final_Data_for_HealthAttractionsheet. A PDF version of the code is also available.
File 2: Marriage_by_Sex_BR.Rmd - Analysis of MarriageM and MarriageF sheets. A PDF version of the code is also available.

Data Location: Datasets are stored in SubmissionKey/Data/Data_Marriage_Health_and_Attraction.xlsx.

Other publicly accessible locations of the data: NA

Data was derived from the following sources: NA

Human subjects data

Due to the sensitive nature of the research, strict protocols were implemented. Participation was voluntary, with participants informed of the study's aims, procedures, and risks before giving consent, including parental consent for minors. Participants could skip questions or withdraw at any time without consequences. To ensure confidentiality, all data were anonymized, and no identifiable information was included. Ethical approval was obtained from local and institutional review boards, adhering to international research guidelines.