Influence of Screen Transition Format on User's Article Browsing Behavior on Smartphones
Ryuki Onitsuka (AY 2023)
Despite the rise of text browsing on smartphones, the problem of print-oriented websites is becoming more serious. Recent survey results show that the average time spent on text-based sites has increased, with many people using their smartphones as their primary device for reading news. It also shows that many people want to increase their reading volume and are particularly interested in news apps. Second, previous research has shown that screen transition format affects comprehension, strategic backtracking, and workload, which would be essential to understanding text reading in a digital environment. In this study, we considered the experimental design of reading all specific texts as a limitation of previous research and used Figma UI design software to replicate a realistic environment. This allows the introduction of "scroll depth", which measures the amount of text viewed, and investigates the environment (UI/UX) for deeper reading of text on smartphones.
In this study, a user experiment was conducted with 20 students from the University of Tsukuba using a within-subjects design. Subjects first responded to a questionnaire that included basic information (gender, age, and affiliation), average time spent using a smartphone, and preferences for the 10 topics used in the experiment. Then, after reading the instructions that included the experimental setting, they had the opportunity to familiarize themselves with the experimental environment through a tutorial. Participants then spent 10 minutes freely browsing the articles in the experimental environment using different screen transition formats (scrolling, paging) and answering a comprehension and workload questionnaire. This procedure was repeated twice, and at the end of the session, participants answered which format they preferred and completed a questionnaire in which they freely stated the reasons for their preference. In previous studies, texts from science news were used because they have fewer emotional elements and have less impact on participants, and in this study, it was also decided to use texts from the science portal operated by the Japan Science and Technology Agency as the subject of the experiment. In addition, to unify the direction of the screen transition, we used up-and-down motion in the paging format, whereas left-right motion is common.
As a result, no significant differences were found in any of the dependent variables in the statistical hypothesis test at the 5% significance level. Therefore, considering the correlation between the content of the free text responses and specific dependent variables or significant differences due to the screen transition format, we examined the interaction effects of "familiarity" with the operation, "interest" in the subject matter, "number of characters," and "difficulty" of the text as additional independent variables, but none of the results were statistically significant. However, the additional analysis suggests that the experimental design with the science portal as the subject is appropriate. Another finding is that we were able to accurately capture the concept of "familiarity" and confirm that it did not confound the experiment. This is because the open-ended responses and the assessment of "habituation" within the questionnaire suggested unfamiliarity with the page format, but no interaction effect was observed. Future directions include re-examining the questions and sample size of each questionnaire, and re-examining the hypotheses in an environment similar to the one used in this study, since the experimental environment did not seem to present any problems.
(Translated by DeepL)