A method of assessing facial profile attractiveness and its application in comparing the aesthetic preferences of two samples of South Africans

Abstract

Objective: Numerous studies have evaluated the perception of facial attractiveness. However, many of the instruments previously used have limitations. This study introduces an improved tool and describes its application in the assessment of the preferred facial profile in two sample groups.

Design: Cross-sectional study.

Setting: Two sites were involved: a rural healthcare facility (Winterveldt, Northwest Province) and the campus of the University of the Witwatersrand (Johannesburg, Gauteng Province).

Participants: Adult females and males selected from amongst first, attendees at the healthcare facility, and second, staff of the University of the Witwatersrand.

Methods: Eight androgynous lateral facial profile images were created using a morphing software programme representing six transitions between two anchoring extremes in terms of lip retrusion/protrusion vs protrusion/retrusion. These images were presented to, and rated by, two mixed male/female groups of rural and of urban habitat using a pre-piloted form. Statistical analysis of the responses obtained established the preferred facial profile by gender in each group.

Results: The perception of facial attractiveness varied marginally between rural and urban black South Africans. There was no statistically significant difference between females and males in the rural group (P = 0·2353) and those in the urban sample (P = 0·1318) with respect to their choice of ideal facial profile. Females and males in both the rural and urban groups found extreme profile convexity unappealing. By contrast, a larger proportion of rural females, rural males and urban females demonstrated a preference for extreme profile concavity.

Conclusions: The research tool described is a useful instrument in the assessment of facial profile attractiveness.

Keywords:

• Profile
• aesthetics
• morphing
• preference

Acknowledgments

The authors wish to thank Professor Peter Cleaton-Jones and Professor Paul Fatti for their assistance with the statistical analyses and Professor Bill Evans for his advice during the preparation of this manuscript.