Saturday, August 11, 2012

The Lack of Video Captions

Update: it looks like Coursera is now taking the video captions seriously; almost all of the videos now posted have captions. Strangely, there is no announcement on the homepage to let us know about that. I found out just by accident. The latest announcement on the homepage as of August 26 is an announcement dated August 14.


As I mentioned in my earlier post about course communication problems, people have been complaining - to no avail - about the lack of video captions for the latest videos. During the first week, when we were reading the fairy tales of the Brothers Grimm, transcripts were very helpfully supplied for the numerous videos. Then, starting in Week 2 - Alice, there were no transcripts (although for one "extra" video about the Week 2 content, a transcript did appear this morning, oddly enough), and there is no transcript for the first video for Week 3 - Dracula, the one we are supposed to watch before we start reading for Week 3 (reading that people are doing this weekend, presumably).

As a result of the missing transcripts, people have been making requests at the discussion forums - I've seen requests from deaf and hard-of-hearing students, and also from many non-native speakers of English who find the transcripts essential in being able to follow what the professor is saying accurately. I'm someone who would like the transcripts just for sheer convenience; it is faster, easier and more accurate for me to access the information in text form. I really don't have time to listen to all the videos and, just as a general rule, I would prefer to read.

So, the complete lack of a response from the Coursera staff for the course got me curious about this. I remembered that the contract between Coursera and Michigan had been leaked and reprinted at the Chronicle of Higher Ed. I figured ADA had to be discussed in there, and sure enough it is. Here is what it says about transcripts/captions for videos:

Company will provide an "Audio Text Transcript" for the audio stream (i.e. captions), as follows:
  • For all University Courses offered to the public under the Coursera Monetization Model whose initial enrollment is above 10,000 End Users, the audio will be proactively captioned within seven days of the time that the Instructor uploads the video onto the Website.
  • For all University Courses offered to the public under the Coursera Monetization Model whose initial enrollment is fewer than 10,000, the audio will be captioned upon the request of an End User, who has a disability, in a timely manner, as specified below.
  • For any University Courses under the University Monetization Model or the Registered Students Model for which University requests such captions, at an agreed-upon fee.
For the people who are under the impression that Coursera is some kind of charity because we are not charged a fee for this course, think again: reading through this contract is a great reminder that they are very much a business venture. We are operating, I believe, under the "Coursera Monetization Model" - and I think that our enrollment is surely over 10,000 for this class (even though not all enrolled students are turning in written assignments). So, based on this contract, I would assume that we should be seeing transcripts within one week from the date when the video is uploaded. That does not seem to be the case, though, because the "Alice: Before You Read" video was uploaded on Thursday, August 2, and there is still no transcript for it (and we are now completely done with the Alice material, of course, having moved on to Dracula). Based on what I understand about this contract (don't quote me, though - eegad, lawyer-ese is hard to understand), that means we should have had the "Alice: Before You Read" audio transcript by now.

Worse, though, is that the one-week delay allowed for in the contract is really not feasible for this class. If the "before you read" video is released only on Thursday, and our writing assignment is due on the subsequent Tuesday, five days later, that means deaf students who are dependent on the video captions will not be able to access the "before you read" video before they do the reading and writing assignments for the unit.

One of the reasons I am not interested in developing video materials for my own courses is exactly because of the time-consuming need to supply a transcript - it's faster and easier for me to supply the content to the students in written form which does not have the same accessibility problems. I know that video appeals to a lot of people, of course... and if Coursera is committed to the use of video, as they are, then this question of timely availability of transcripts is a very important issue for them to resolve. Even if they did follow the terms of the contract and supplied the transcripts one week after the video is made available to the students, the deaf students in this class would not be getting the video transcripts - esp. the "before you read" video - in time to access that video before they do the week's reading.


  1. Laura - Has anyone actually done any research into whether video is actually more effective for learners in general? (IE a larger percentage of students actually learn more) I think people like the idea of 'just watch a video and learn", but I don't personally find it that effective for most subjects (although some people are wonderful lecturers and worth watching and listening to.)

    BTW, I signed up for this course as well and left after the 2nd week because I wasn't learning anything (beyond what I could learn by reading the books on my own) but the essays and trying to provide peer evaluations were painful and took up way to much time.

    1. Hi Julie, I'm not a big fan of video myself (I read the transcripts rather than watched the video which is why the initial lack of transcripts was so frustrating for me, completely aside from ADA and accessibility, etc.). I'm not even sure you could really measure video effectiveness, at least not directly, given the enormous number of factors involved, variability of content, from student to student, short-term v. long-term retention, etc. - Michael Feldstein actually had a great article about that at eliterate this weekend:
      Why Big Data (Mostly) Can’t Help Improve Teaching



Note: Only a member of this blog may post a comment.