As science communication increasingly transitions from traditional text-based formats to multimedia content, short videos have emerged as a dominant form, particularly on IoT-enabled platforms that facilitate rapid information dissemination. However, the relative effectiveness of text versus short videos in communicating scientific knowledge remains underexplored. In this empirical study, we investigated how text and short videos perform in conveying information about climate change within IoT platform environments. A mixed-method experiment was conducted with 63 participants (32 men, 31 women), who were randomly assigned to receive content in either text or video format. A comprehensive, multidimensional evaluation framework was developed, grounded in dual-channel theory, cognitive load theory, and the technology acceptance model, and implemented via structured questionnaires. Descriptive statistics were analyzed using IBM Statistical Package for the Social Sciences (version 25), employing t-tests, one-way analysis of variance, Bartlett’s test of sphericity, and the Kaiser–Meyer–Olkin test. Results indicate that short videos outperform text in terms of memory retention, reduced cognitive load (p < 0.05), perceived readability, usefulness, and viewer satisfaction. Text-based content, however, was rated higher in perceived academic rigor and credibility. Willingness to share content showed no significant difference between the two formats (p = 0.134 > 0.05). These findings underscore the effectiveness of short videos as powerful tools for science communication on IoT-enabled platforms, while also highlighting the continued value of text in contexts requiring academic depth and trust.