Media Effects Research Lab - Research Archive
Looking at the web 2.0 world: How modality and bandwagon cues impact internet user attitudes
Student Researcher(s)
Justin Walden (Ph.D Candidate);
Mun-Young Chung (Ph.D Candidate);
George Christo-Baker (Ph.D Candidate);
Faculty Supervisor
INTRODUCTION
As the news industry continues its digital evolution, news editors have increasingly relied upon an array of media formats to reach audience members. While a wave of industry studies have recently investigated consumer behavior and preference when it comes to online content, academic research into the area of modality goes back almost four decades. Furthermore, recent research has also considered the ‘bandwagon’ phenomenon that may shape perceptions of content in today’s digital culture. The problem is that this literature is complex and findings are occasionally contradictory. This study attempts to fill this literature gap and clarify some of the questions involved with consumer perceptions of online content. Specifically, this study points to the continued need for refined study measurements when studying modality and bandwagon related issues. This study also sheds new light on some of the factors that may influence the production of user-generated responses to professionally-produced news stories and, unexpectedly, how time spent online is strong a predictor of user production of online content.
RESEARCH QUESTION / HYPOTHESES:
RQ1: Controlling for story, what is the relationship between level of modality in a health news story and perceptions of story content?
RQ2: Controlling for story, what is the relationship between level of modality in a health news story and likelihood that a person will seek to engage in discussions about the story content?
RQ3: Controlling for story, what is the relationship between perceived bandwagon effects and perceptions of story content?
RQ4: Controlling for story, what is the relationship between perceived bandwagon effects and likelihood that a person will seek to engage in discussions about the story content?
H1A: There will be a relationship between modality and perceptions of story quality, such that the greater the number of modality presentations, the higher the perception of story quality.
H1B: There will be a relationship between modality and perceptions of website quality, such that the greater the number of modality presentations, the higher the perception of website quality.
H1C: There will be a relationship between modality and behavioral intentions, such that the greater the number of modality presentations the greater the person will have learned from the story.
H2A: There will be a relationship between level of bandwagon cue and perceptions of website quality such that the greater the number of Tweets, Diggs and Facebook links appearing with a story, the higher perception of story quality.
H2B: There will be a relationship between level of bandwagon cue and perceptions of website quality such that the greater the number of Tweets, Diggs and Facebook links appearing with a story, the higher perception of website quality.
H3A: There will be a relationship between modality and behavioral intentions, such that the greater the number of modality presentations, the greater the likelihood a person will seek to contribute online comments of their own about the article.
H3B: There will be a relationship between perceived bandwagon cue and behavioral intentions, such that the greater the number of interactive features, the greater the likelihood a person will seek to contribute online comments of their own about the article.
METHOD
This study was originally designed as a 4 (modality level) x 3 (bandwagon level) mixed-design factorial experiment, controlling for three story conditions. The independent variable modality was operationalized as the level of visual and audio dimension format of story that was presented to participants. Also, the bandwagon effect was operationalized as the different level of sharing of health information on the social network sites.
Four different types of modalities were used for the study: text only, text with picture, text with audio, and text, picture, and audio combined. For the bandwagon condition, this study used the number of sharing or contribution with public sharing or social media icons from Facebook, Twitter, and Digg. For examining story effects, this study used stories with different topics (puberty, obesity, and microbes) to help minimize the risk that modality and bandwagon-related results would be confounded by particular story-related factor such as relevance or salience. Appendix 1 shows a screen capture of sample stimulus of the microbes story in the medium bandwagon, text/picture/audio condition. A total of 231 undergraduate students voluntarily participated in the study in exchange for a small amount of extra course credit.
RESULTS
A significant relationship between modality and learning emerged, such that participants in the text with audio condition (M=3.24, SD=.35) reported that they learned more than participants who were exposed to the other three conditions (See Table 1, Figure 1).
There were no significant main effects of bandwagon cues; however, this study found a significant interaction effect between bandwagon and story toward the contribution the health information, F (2, 206) = 3.69, p < .05. (See Table 3).
For additional findings, the hour of blogs was a significant predictor of learning, F(1, 228) = 6.11, Adjusted R2 = .02, β = .05, p < .05. The hour of reading news sites, F(1, 228) = 7.57, Adjusted R2 = .03, β = .05, p < .01, reading blogs per week, F(1, 228) = 15.10, Adjusted R2 = .06, β = .09, p < .01, and surfing the Internet, F(1, 221) = 4.97, Adjusted R2 = .02, β = .01, p < .05, were significant predictors of the contribution to leave comments for everyone to read or to post a blog about the health information (See Table 4).
DISCUSSION/ CONCLUSION
Even though this study could not find any significant effect toward the perception of the quality of story or website for health information, this study examined the significant modality effect toward learning from health information and the bandwagon with story interaction effect toward the contribution of the health information. These findings may be fruitful for the future research which is related to the health education and the public health campaign through media technology. Also, this study found that the hour of blogs can be a predictor of learning and contributing both, and the hour of reading news sites and internet surfing can be a predictor of the contribution to leave a comment for everyone to read or to post a blog about the health information.
For some limitations of this study, it was a dilemma to build optimized stimuli webpages concerning both the experiment and the ecological validity because most health information websites contain various visual cues that make hard to simplify for the stimuli designs. Another limitation was a slight story effect. The study speculated that the story content was more relevant to the sample demographic. For the future studies, researchers may use pre-test to find the most valuable health issues for the sample, or use the between-subject design for the experiment to minimize the story effect. Also, more approaches for the user generate content about the contribution of health issues will be requested for exploring the online users’ attitude and behavior changes from health information in the interactive media environment.
For more details regarding the study contact
Dr. S. Shyam Sundar by e-mail at sss12@psu.edu or by telephone at (814) 865-2173