A couple of blogs recently have highlighted user viewing times as evidence of the effectiveness of short videos for teaching in HE.
Alan Cann (http://scienceoftheinvisible.blogspot.co.uk/2013/11/whats-wrong-with-lecture-capture-this.html) uses viewing times on You Tube videos (which he suggests are around 45-85 seconds) to argue for short (< 5 mins) videos covering difficult to grasp or threshold concepts to supplement lecture material. Similarly, Philip Guo (https://www.edx.org/blog/optimal-video-length-student/1239) uses the viewing times of videos in edX MOOCs to show that the ‘optimal’ video length is 6 minutes. Both posts specifically mention students’ short attention spans.
I’m not sure that either of these explanations is persuasive in the HE setting. Firstly, the statistics upon which they are based are not context specific – You Tube videos cover pretty much everything from A (http://www.youtube.com/watch?v=cIRGGG8bvQk) to Z (http://www.youtube.com/watch?v=HcwTxRuq-uk) so the cited statistics are very general and, although the modules formats are not mentioned, the MOOC learning experience, and the student body so far as we know it, is not typical to that of the majority of HE (even distance learners). Secondly, I’m not sure that either justifies a claim that students using video online have a shortened attention span as the research seems to show a more complex picture. For example, Leadbeater et al (Computers and Education, 2013, 61:185-192), show students having different viewing patterns when using recorded lectures, with some viewing the whole thing and others moving between sections, and de Boer et al (Computers in Education, 2011, 56(3):727-735) identifying four viewing styles for instructional videos among a group of undergraduate students : linear (watching a video from beginning to end), elaboration (watching a video again after viewing it completely), maintenance (watching parts of a video repeatedly) and zapping (skipping through with relatively short viewing intervals).
Based on this I’m not sure that it is possible to make generalisations about optimal video lengths or short attentions spans.