Big Data or Small Stories? Exemplification and Interactivity Effects on Shaping Perception and Attitude of Social Issues
Open Access
- Author:
- Jia, Haiyan
- Graduate Program:
- Mass Communications
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- September 03, 2014
- Committee Members:
- S. Shyam Sundar, Dissertation Advisor/Co-Advisor
S. Shyam Sundar, Committee Chair/Co-Chair
Mary Beth Oliver, Committee Member
Michael Grant Schmierbach, Committee Member
Mary Beth Rosson, Committee Member - Keywords:
- Interactivity
exemplification theory
interactive visualization
baserate information
big data
user engagement
vividness - Abstract:
- With the rising popularity of “big data”, an unanswered question is the extent to which laypersons are ready, willing and able to make sense of the information embedded in large-scale data. In order to make this data more accessible to readers, journalists tend to present statistical content about social issues and problems by accompanying them with exemplars or case studies that illustrate the data pattern. But, these exemplars do not always fully capture the complexity of the data patterns, yet they end up playing an important role in shaping user perceptions of the issue because they are more vivid than baserate information. How can we counter this biasing effect of isolated, non-representative exemplars and make readers pay more attention to baserates? Journalists have tended to employ visualization techniques to attract greater reader attention to baserates, but with limited success. This dissertation proposes that interactive visualization of big data is the answer. Interactive visualization can increase information processing of the baserates through two potential mediators: Vividness and Engagement. Interactive information visualization can make the pallid baserate information more vivid and therefore easier to process. Alternatively, by requiring users to act on the visualization, it can draw their attention to the baserate information and therefore serve to engage them with the data-rich content. The study design involved a 2 (baserate-exemplar consistency) × 3 (interactivity of information visualization) × 2 (issue topic) mixed-design experiment. Participants were asked to read news reports about climate change and same-sex marriage depicted with a combination of exemplifying cases and information visualization of baserates. By empirically examining the interactivity effects in individuals’ issue perception formation and related outcomes, this study revealed distinctive mechanisms through which interactive visualization affects users’ information processing and emotional responses. The findings show that interactive visualization is positively associated with user psychology, and further leads to perceived vividness and user engagement with the content. Compared to static visualization, interactive visualization is effective in motivating systematic processing of the baserate information, which positively influences accuracy of issue perception as well as personal attitude. Results help uncover the psychological processes through which individual perceptions are influenced by journalistic evidence, baserates in particular, in a highly visual and interactive media environment. The study also generates useful findings to inform ethical journalistic practices and effective interface designs for communicating big-data information.