Observational data were gathered from a large university cafeteria for
a period of 22 days, in 1-hr periods per day, over one semester. Obse
rvations were made of the frequency with which Black and white cashier
s were selected. Chi-square analyses showed a significant association
between a cashier's being Black and increased likelihood that she woul
d not be selected. Some comments and comparisons are made with other r
esearch using similar measures.