The Impact of Expertise on Evaluator Bias and Variance

Abstract:

Humans are prone to various biases when making decisions, including demographic biases like sexism and racism, and cognitive biases like framing effects, beauty effects, and information effects. We investigate how expertise affects the prevalence of these evaluation biases when assessing others. Unless demographic evaluation biases are extreme, they’re unlikely to be detected. Our idea is that as expertise increases, it reduces evaluation bias and decreases evaluation variance as the evaluator is less reliant on their biases. We use a behavioral lab with a financial literacy knowledge treatment midway through the experiment to change the level of expertise.