The Effects of Affective Framing and Interactivity on Persuasive e-Learning
Jacob Oury (Ph.D Candidate); Srishti Gupta (Ph.D Candidate)
Dr. S. Shyam Sundar
This paper was based on a project as part of the “COMM517 Psychological Aspects of Communication Technology” course.
In this study we explore how psychological mechanisms of persuasive design are affected by interactivity and message framing for information presented to users through an interactive article about user experience and usability. We used a 3 (Message Interactivity: High/Medium/Low) x 2 (Framing: Gain/Loss) factorial experiment (N=126) to determine whether interactivity and framing would have differential effects on the value that a person placed on user experience and usability after reading the interactive article. We found that changing interactivity was associated with changes to value proposition, but there was no effect found for framing on the subjects’ value proposition for user experience and usability.
RESEARCH QUESTION / HYPOTHESES
H1: Positive framing will show greater value proposition than negative framing.
H2: Higher interactivity will be correlated with higher value proposition, particularly for low-involvement users.
H3: Interactivity will be correlated with higher interface assessment scores.
H4: Elaboration will differ across interactivity conditions and higher elaboration will be associated with a higher value proposition.
RQ1: How do interactivity and framing affect value proposition and self-reported elaboration?
We developed a 3 (Message Interactivity: High/Medium/Low) x 2 (Framing: Gain/Loss) fully factorial, between-subjects experiment on Mechanical Turk to find out how interactivity and framing would affect the value proposition of users of a simple, interactive website to learn about the basics of usability. The participants were directed to an interactive website where they were instructed to read through a brief article about the basic of usability. After this, they completed a post-survey.
We found that the framing manipulation did not results in significant differences for any outcomes variables, including the primary dependent variable perceived value. We could not support any findings about framing which led us to dismiss H1. We also were unable to establish an effective manipulation with framing using our manipulation check questions. We also analyzed interactivity’s effects on perceived value of usability. We found that H2 was partially supported; the high interactivity condition had the highest perceived value, though low interactivity was second highest. Our framing manipulation check for interactivity, perceived interactivity, did not show significant group differences however which tempers these results. Our comparisons for interactivity condition on elaboration and interface assessment also did not show significant differences which led us to dismiss H3 and H4. There was no significant interaction found between interactivity and framing, which was our research question. Finally, we used the PROCESS model and found significant mediating effects of elaboration, interface assessment, and issue involvement on perceived value though the model was overall not significant.
Proposed mediation effects from the Process Model
One plausible explanation for these findings on interactivity could be related to how interactivity is applied. The study framework was based on Oh & Sundar’s (2015) study on persuasive design and interactivity for smoking cessation messages. Their study operationalized interactivity through two methods: website depth and interaction modality in the form of a slider. Our study similarly used website depth, but we did not directly apply the modality operationalization for our use of the accordion-style text boxes. Both the medium and high interactivity conditions used this feature, but the high interactivity condition combined it with website depth as well. The lack of a consistent trend for perceived interactivity may have been due to inconsistent interface design for interactive features.
Framing seemed to be essentially unsuccessful across the board. While interactivity may have some justification for why perceived interactivity and true interactivity can be difficult to correlate, framing did not capture any significant differences in outcomes for perceived value or other outcomes. Daamgard and Nielsen’s (2018) review concludes that framing needs to be strongly operationalized and detected by the subject to be effective; this experiment may not have supported strong framing effects between groups. This project is consistent with their overall findings that framing tends to have mixed results in educational contexts. Future work could try to identify how to measure the strength of given framing effects. Our study relied on a manipulation check developed by the authors. This method should be tested under conditions with known framing effects before assessing its performance.
Oh, J., & Sundar, S. S. (2015). How does interactivity persuade? An experimental test of interactivity on cognitive absorption, elaboration, and attitudes. Journal of Communication, 65(2), 213–236. https://doi.org/10.1111/jcom.12147
Damgaard, M. T., & Nielsen, H. S. (2018). Nudging in education. Economics of Education Review, 64(October 2017), 313–342. https://doi.org/10.1016/j.econedurev.2018.03.008
For more details regarding the study contact
Dr. S. Shyam Sundar by e-mail at firstname.lastname@example.org or by telephone at (814) 865-2173