Background: YouTube is widely used for learning about scientific issues in and out of school. However, much of the scientific information on YouTube is inaccurate. Prior studies have mostly focused on how students evaluate textual online information sources and have not yet systematically examined how they evaluate authentic scientific YouTube videos. Objectives: The purpose of this study was to examine how students evaluate authentic scientific YouTube videos in comparison to scientific webpages. Methods: Eighth grade students ranked six YouTube videos on the topic of climate change, justified their rankings, and responded to metacognitive prompts designed to elicit their evaluation criteria and strategies. Students also responded to a parallel webpage evaluation measure. We analysed and compared students' evaluation strategies and criteria. We also examined which evaluation criteria predict students' judgments of YouTube video quality. Results and Conclusions: Students predominantly relied on evaluation of video content, focusing mainly on criteria of communicative quality and explanation quality. Students tended to neglect source expertise and information validity criteria. Students evaluated videos mostly similarly to how they evaluated webpages. However, affective experiences played a greater role in video evaluation; whereas task relevance and verbal quality played a smaller role. Students' evaluation criteria predicted their judgments of video quality. Implications: The findings suggest that students are unprepared for critically evaluating scientific YouTube videos and that digital information literacy instruction should address this gap. The study also identifies some unique challenges of video evaluation that educators should attend to.