Race and Gender-Based Perceptions of Older Septuagenarian Adults

Abstract

Objectives: Older adults face racism, sexism, and ageism. As the U.S. population ages, it is important to understand how the current population views older adults.

Methods: Participants recruited through Amazon's Mechanical Turk provided perceptions of older Black and White models' photographs. Using mixed-effect models, we assessed interactions between race and gender of participants and models.

Results: Among Participants of Color and White participants (n = 712, 70% non-Hispanic White, 70% women, mean 37.81 years), Black models were perceived as more attractive, less threatening, and sadder than White models, but differences were greater for White participants (race-by-race interaction: attractive p = 0.003, threatening p = 0.009, sad p = 0.016). Each gender perceived their respective gender as more attractive (gender-by-gender interaction p < 0.0001). Male and female participants perceived male models as happier than female models, but differences were greater for male participants (p = 0.026). Irrespective of participant age group, women were perceived as more threatening (p = 0.012). Other perceptions were not significant.

Discussion: Participants had few biases toward older Black and White models, while gender biases favored men.

Keywords: bias; gender disparities; geriatric; racial disparities.

FIG. 1.

Photos of older adult subjects. These photos of subjects in their seventh decade were used in the MTurk survey. MTurk, Mechanical Turk.

FIG. 2.

Survey inclusion criteria. Some surveys met more than one exclusion criteria. The total number of surveys excluded on all criteria was n = 595. When a participant reported the wrong gender for only one photo, the participant's entire survey was not removed; rather their responses were excluded for that individual photo. From the final sample of 712 participants, responses for a photo were excluded for 82 of them. This does not change the total number of surveys included in the analysis. Outliers for completion were defined by Tukey's Fences (less than quartile 1 − 1.5 [interquartile range] = −23.5 seconds, or greater than quartile 3 + 1.5 [interquartile range] = 1636.5 seconds).

FIG. 3.

Interactions by race. Higher scores indicate higher ratings for photos, whereas lower scores indicate lower ratings for photos. *p-Value was <0.05 for attractive, sad, and threatening. BIPOC, Black, Indigenous and People of Color.

FIG. 4.

Interactions by age and race. Higher scores indicate higher ratings for photos, whereas lower scores indicate lower ratings for photos. No p-values reached statistical significance.

FIG. 5.

Interactions by gender. Higher scores indicate higher ratings for photos, whereas lower scores indicate lower ratings for photos. *p-Value was <0.05 for happy and attractive.

FIG. 6.

Interactions by age and gender. Higher scores indicate higher ratings for photos, whereas lower scores indicate lower ratings for photos. *p-Value was <0.05 for threatening.