Religion Seminar Series 2021. How accurate are informant reports of their religious behaviour? A comparison of self-report and systematic observation in a rural Fijian Village

Speaker: Associate Professor John Shaver

Social scientists rely on self-reports for measuring behaviour despite ongoing criticism concerning informant inaccuracy. But are informant reports inaccurate? And are biases in self-report random? Here I compare self-reports of church attendance to observed attendance across 48 services in a rural Fijian village.

Findings suggest that:

1. self-report does not reliably predict observed attendance
2. women with several children are more likely to over-report their attendance than women with fewer children
3. self-report of religiosity is more reliably associated with observed church attendance than self-report of church attendance

Further, third-party judgments of church attendance by fellow villagers are more reliably associated with observed church attendance than self-report. Findings suggest that informants inaccurately report their religious behaviour, but that biases are culturally influenced. Researchers interested in estimating behavioural variation should consider third-party methods to avoid biases inherent to self-report.

Save this event

• Google Calendar
• Yahoo Calendar
• iCalendar