The Effects of Facial Close-Ups and Viewers' Sex on Empathy and Intentions to Help People in Need

Abstract

This study examines the interaction effects of camera perspectives used to portray a victim of a social problem and the sex of audience members on empathy and intentions to help people sharing the victim's problem. As predicted, the study found interaction effects for camera perspectives and viewers' sex on empathy and care for the victim group. The findings corroborate not only the role played by empathy in inducing care for people in need but also the parallels between one's real and mediated experiences. Most importantly, the findings contribute to our understanding of how media portrayals of human suffering can increase empathy for those in need.

ACKNOWLEDGMENTS

I thank Diana C. Mutz for guidance as well as Shawnika J. Hull, Paul R. Brewer, the editor, and three anonymous reviewers for many helpful comments.

Notes

¹This study examines the interaction effects of camera perspectives used to portray a victim of a social problem and the sex of audience members using American participants. As a result, the research presented here speaks only to sex differences in the United States.

²Even though women are more accustomed to intimacy with another woman than with another man, whereas men's demand for distance is more pronounced in interaction with another man than with another woman (Leibman, 1970), this does not necessarily challenge the notion that women in general prefer closer interaction distance than men. Indeed, studies have found that women in general are more comfortable at closer interaction distance than men regardless of the sex of the person with whom they interact (Aiello, 1977; Camperio & Malaman, 2002).

³The first video was from the documentary Domestic Violence, the second from Off to War: From Rural Arkansas to Iraq. A pilot study was conducted to test victim likeability and effects of the two videos on problem attributions given that victim likeability and problem attribution can affect participants' empathic responses (Zillmann, 2006). The pilot study found that the two videos were comparable in terms of victim likeability and problem attributions.

⁴When changing to facial close-ups of the victim, the videos cut to the close-ups directly rather than showing the process of the face becoming closer.

⁵The original perspective taking index is a reliable and valid measure of individuals' perspective taking abilities (Davis, 1980; Mutz, 2002). Given that the original measure was created to assess individuals' perspective taking abilities in interpersonal communication, I modified the measure to capture participants' perspective taking experiences during the video viewing.

⁶Research has shown that the measure is highly reliable and valid (e.g., Batson et al., 2002; Batson et al., 1997).

⁷A factor analysis was conducted on the nine items used to measure empathy with oblique rotation (direct oblimin) because the two subscales used to capture perspective taking and the emotional experience were expected to be correlated. The analysis showed two components with eigenvalues greater than one. As expected, all five items measuring one's emotional experience clustered on Component 1 and all four items tapping perspective taking clustered on Component 2. There was a strong positive correlation between the two components (r = .41), which supported my decision to combine the perspective taking and the emotional experience subscales into one composite measure of empathy.

⁸ANOVAs for mixed-design were also conducted with camera perspectives, participants' sex and the order in which the two videos were presented as between-subject factors and victims' sex as a within-subject factor. The analyses did not show a significant main effect of video order or significant interaction effects involving video order, so video order was dropped from the analyses presented here.

⁹All the p values reported in this article were based upon one-tailed tests because all hypotheses are directional.

¹⁰Given that the abused woman (Joyce) is an African American and the disabled war veteran (Wayne) is a White American, analyses were also conducted to examine the potential interaction effects of participants' race and victims' race on empathy, care for the victim group, and intentions to donate. The analyses did not show any statistically significant interactions.

¹¹Here is the equation for the Sobel test: Z-value = a*b/SQRT (), where a is the regression coefficient for the association between the IV and the mediator, S_a is the standard error of a, b is the regression coefficient for the association between the mediator and the DV (when the IV is also included in the model), and S_b is the standard error of b.

Additional information

Notes on contributors

Xiaoxia Cao

Xiaoxia Cao (Ph.D., University of Pennsylvania, 2010) is an Assistant Professor in the Department of Journalism, Advertising, and Media Studies at University of Wisconsin–Milwaukee. Her research interests include media effects, media psychology, persuasion, and political communication.