When is Fiction as Good as Fact? Comparing the Influence of Documentary and Historical Reenactment Films on Engagement, Affect, Issue Interest, and Learning

Abstract

This study advances documentary film effects research by comparing the influence of a political documentary with a historical reenactment film on narrative engagement, affect, learning, and interest. Using the Rwandan genocide as a context of study, a documentary film, The Triumph of Evil, and the historical reenactment fictional film, Hotel Rwanda, were examined. Results revealed significant differences between documentary and historical reenactment film exposure for affective responses and issue knowledge gain. However, increased issue interest and narrative engagement were not significantly different for the two stimulus groups, suggesting that dramatic fictional reenactments of socio-political events lead to increased issue interest as much or more than the live footage and factual account of events offered in a documentary. The results of this study are discussed in terms of their importance for key democratic outcome variables (e.g., knowledge and participation), along with suggestions for future documentary film research.

Notes

¹Although prior viewing was controlled in each regression, we wanted to ensure that prior viewing was not a moderator of any of the relationships between condition and the dependent variables. We conducted Fisher's z transformations to more formally test for this possibility. This test involves finding the difference between the two groups' unstandardized beta coefficients on the key variable of interest and then dividing by the square root of the total squared standard errors (Cohen & Cohen, 1983). The test provides a z score for which significance can then be assessed. Values of 1.96 or greater are determined to be significantly different at the p < .05 level, indicating a significant difference between groups (i.e., moderation). We included condition as the sole independent variable to predict the various dependent variables we assessed. Then, we selected cases so that only previous viewers of Hotel Rwanda were in the initial set of regression equations, and we then ran the same models (i.e., condition predicting the dependent variable) for those who had not seen the film in the past. We recorded the unstandardized beta coefficients and respective standard errors for each regression equation. Next, we computed the Fisher z transformation score to assess significant differences. Results revealed there were no significant differences between the two groups for any of the five dependent variables (guilt, disgust, interest, knowledge change, narrative engagement), thus prior viewing of Hotel Rwanda was not a significant moderator at the p < .05 level.

Note. N = 97. Unstandardized ordinary least squares regression Bs are reported (standard errors in parentheses).

∗p < .05.

Note. N = 97. Unstandardized ordinary least squares regression Bs are reported (standard errors in parentheses).

†p < .10.

Note. N = 95. Unstandardized ordinary least squares regression Bs are reported (standard errors in parentheses).

∗p < .05. ∗∗p < .01. ∗∗∗p < .001.

Note. N = 94. Unstandardized ordinary least squares regression Bs are reported (standard errors in parentheses).

∗p < .05. ∗∗p < .01. ∗∗∗p < .001.

Additional information

Notes on contributors

Heather L. LaMarre

Heather L. LaMarre (Ph.D., The Ohio State University, 2009) is an assistant professor in the School of Journalism and Mass Communication at the University of Minnesota. Her research examines the social-psychological processes and effects of strategic communications and campaigns in news and entertainment media.

Kristen D. Landreville

Kristen D. Landreville (M.A., University of Florida, 2006) is a doctoral candidate in the School of Communication at The Ohio State University. Her research interests include the processes and effects of political entertainment use with a focus on emotion and discussion.