This study explored multiple biases-the possibility that different biases would concurrently occur in a given situation, and each would exert its influence independently on people's judgments. The study focused on media bias through nonverbal (NV) behavior, where viewers judged an interviewed politician after they viewed the interview with a nonverbally friendly or hostile interviewer. In a meta-analysis of several replications, 2 independent biases were found: media bias (viewers rated the interviewee more favorably when the interviewer's NV behavior was friendlier); and halo effect (viewers rated the interviewee according to the degree that they personally liked him). Regression analyses indicated that these 2 biases operated independently and additively on viewers' judgments. Implications for the study of multiple biases are discussed.