Students in CS like videos of Java coding

May 4, 2011 at 10:17 am 10 comments

This semester, I did a couple of new things in my Media Computation data structures class (as mentioned in a previous blog post). I started using Ubiquitous Presenter so that I could ask in-class questions, let students see the results, discuss, and re-vote.  I also created out-of-class video quizzes.  Each week, I created one or two short screencasts (almost all less than six minutes), then created a quiz where students had to answer questions about the videos: What was going on when I typed X, what would happen if I typed Y, and what data structure was created?  Only a couple students in my class are CS majors.

I ended up creating 30 videos.  To be frank, these are crappy videos. It’s just me using DrJava, simply capturing the screen and my voice.  I did them in the basement (to avoid kid, cat, and dog noises) on Sunday afternoons in one “take” (okay, I started over if it was really bad, but I kept in my mistakes and confusions) with no editing.  My pedagogical point was to give them a set of questions (an advance organizer), and a video where they could read the code and rewind/pause/fast-forward as they needed. I wanted to give them more worked examples, with the opportunity to review at their pace.  I wanted audio+code because of the Modality Principle.

First piece of data: They used it.  I hooked up Google Analytics.  Visits showed that everyone visited, but not for every quiz.  Quiz grades aren’t great, and as you can see, there was a time during the semester when many students ignored the quizzes.  The semester got hard, and the quizzes weren’t worth that many points.

Second: They re-watched the videos. Average time per visit over the semester is particularly interesting.  Remember that these are 4-6 minute videos on average.  Every spike over 5 minutes means that students are reviewing the video.  I took this graph 1 Jan through 1 May. Our final exam was 2 May. Students decided to review the videos a lot on the night before as a way of preparing for the final.

One of the latter quizzes included some questions about the videos themselves:

How useful are these out-of-class video-based quizzes? Do they help you learn?

Do they help give you insight on lectures? On homework assignments?

Is the video good enough?

Third piece of data: The students really liked the videos!  I was shocked!  I honestly expected answers like, “Well, the videos are awful, but it’s useful to be able to review, and I think I’m learning something from them.”  Instead, I got things like these:

These videos are extremely useful and supplement class very well. It is very helpful to see the code being worked on in Dr. Java after seeing the concepts and more static implementations in class. They have come in handy on homework assignments when I want to refresh my memory on how a certain class or method was altered or created. The video quality and content is good enough and has definitely helped cement the knowledge from the book and lectures.

They do help me learn and I do find them useful on homeworks. I like doing quizzes before the actual class because then I can ask questions about it. The videos are short and give plenty of info about the lectures.

They are very helpful. They go into more actual coding detail than in class discussions of topics, and they are easier to follow, since there are no interruptions. They are also good because they can be watched again at any time. They are useful when completing homework assignments, because they are more focused on one topic than a class lecture, which has to give a lot more background information before we see any real implementation of what we’ve just learned. The video quality is good; it is easy to hear and see.

I think the videos are very useful. I know sometimes when I read the book, I am still unsure about a topic, but being able to watch a video about the topic with the professor talking is a lot more engaging. Being able to get the material from three good resources (class, book, quizzes) really helps me get a grasp on the concepts.

Not a single student said that he or she disliked the videos.  Not a single student said that the quality was bad.  Would you have guessed that of a class of 30 students, watching videos of Java coding?!?

Now, these responses were on a quiz.  The answers were not anonymous, so they might have been telling me what they thought I wanted to hear.  We also did an anonymous end-of-class survey, which we are just transcribing now.  From a brief skim of those, the videos again turned up really high on the list of useful interventions. Students perceive the videos as being one of the most valuable features of the class in terms of helping them learn.

I have lots of follow-up questions.

  • Did the videos really help with learning?  There are lots of ways in which they might have helped with learning, e.g., more learning can result from just spending more time-on-task, trying to make sense of code and interactions and program behavior.  But I can also imagine a student, used to watching YouTube, who finds videos more “natural” than reading or coding, and so perceiving more learning, when it didn’t really happen.
  • I’m curious about the interaction between the videos and the lecture and textbook.  Notice in the quotes above that the videos are seen as providing more than the “more static implementations in class.”  Really?  In class, I use Powerpoint and DrJava.  In videos, I just use DrJava. Why is DrJava in class more “static” than DrJava in videos?  Even more interesting is the student who perceived that the video (watching somebody code) added more than the textbook, that s/he used the video when “unsure about a topic.”  How are they using the videos to supplement the textbook?
  • I’m wondering what the role of the quiz questions are.  I asked them on the survey if the quiz questions changed how they watched the videos (e.g., directed their attention), or if the quiz questions helped the students actually learn from the videos, rather than just let the videos wash over them.
Already, the new intervention is a success for me, in the sense of raising questions, getting results I didn’t expect, and having students engaging with the intervention. I want to try this again, and really study what’s going on.
Important caveats:   I have the students’ permission to share what they wrote.  These are not research results.  These are the results of a teacher evaluating his new intervention and sharing how it worked.  As a researcher, these are pilot results that inform later studies.

Entry filed under: Uncategorized. Tags: , .

Bill to establish a STEM Master Teacher Corps Not “Gamification” — it’s “Exploitationware”

10 Comments Add your own

  • 1. Matt Glickman  |  May 4, 2011 at 10:38 am

    Q: Why is DrJava in class more “static” than DrJava in videos?

    A: Perhaps (1) because Powerpoint is more static, so DrJava without Powerpoint is more dynamic, and (2) the ability to rewind/pause/fast-forward is more “dynamic” (read: interactive).

    Exciting observations, Mark!

    • 2. BKM  |  May 5, 2011 at 1:50 pm

      Why would you be using Powerpoint in class to show Dr Java? Why not just run it directly? Am I missing something here?

      I am actually surprised by this post, because in my experience, students never look at the supplementary videos. Maybe mine just aren’t any good.

      • 3. Mark Guzdial  |  May 5, 2011 at 8:26 pm

        I don’t use Powerpoint to show DrJava. I use Powerpoint to show diagrams of data structures, to show visualizations, to define concepts, etc. I show code and execute it in DrJava.

        • 4. BKM  |  May 7, 2011 at 11:29 am

          That makes more sense. I was reacting to this quote “In class, I use Powerpoint and DrJava. In videos, I just use DrJava. Why is DrJava in class more “static” than DrJava in videos? ” which I interpreted to mean that you had screenshots of Dr Java on Powerpoints, or something like that. That would certainly be more static.
          Is it possible that the students simply focus better when they see the video? In some classrooms, it is really hard to see the screens upfront, and the other students may be distracting. That is certainly a problem for us. Still, my students don’t care for supplementary videos.

        • 5. chaikens  |  April 9, 2012 at 6:18 pm

          Here are some tips about how I tried live coding this semester:
          (1) Blow up the font in DrJava a lot, like 50 pt, to focus on the detail being discussed or demonstrated.
          (2) Use a version history to both capture the content and teach the practice of versioning. Here’s what I’m trying:
          First, make several version directories, V1, V2, V3, etc. Start by saving the file(s) in V1.
          Second, after a final compile and demo of V1 or other version, “save-as” the .java files in the next version directory to start work on the next version. Make sure it compiles and tests ok, to foster the habit of doing this with any source code one receives.
          Third, after the lecture, upload the all the directories to the course web site.

          This practice of versioning can then be required in future lab and/or projects. It introduces and fosters an important practice with very little conceptual overhead, beyond the basics of multiple directories which we find necessary to demand of many students anyway. Further, we require the student turn in a say .zip archive of their version history. That way, we can grade the work on whether the student at least successfully compiled each version before moving on to the next, by checking for the .class files in the submission.


          Click to access L10.pdf

  • 6. Beth Simon  |  May 4, 2011 at 10:38 am


    According to this:
    Improving Classroom Learning by Collaboratively Observing Human Tutoring Videos While Problem Solving
    Scotty D. Craig, Michelene T. H. Chi, and Kurt VanLehn
    Journal of Educational Psychology 2009.

    You might get even better learning by sitting down with someone and tutoring them. I’m not sure if you could “fake” it with someone playing the role of the tutee, since you don’t have people in the video, just voices. It’s not the exact same setup (since you would want to have them be working on a similar problem with a partner….), but it really is interesting to think how the back and forth you would have with the tutee (what questions you would ask, etc.) might benefit students.

    Back at ITiCSE in Dundee, Sue Fitzgerald led a bunch of us in making debugging videos for CS1. We came up with a list of “key” issues (infinite loop, etc.) and worked up scripts in pairs where we recorded us pretending to be students and acting out how we were confused and how to debug these things.

    They were not used (except by Brian Hanks’ students who were required to), likely because they were not directly tied to the specific course content.

    beth simon

    • 7. Mark Guzdial  |  May 4, 2011 at 10:51 am

      Hi Beth,
      Thanks for the reference — I’ll look into it. I TOTALLY believe that I could make the videos much better, e.g., focus on debugging, creating a “tutoring” setting, addressing common misconceptions, etc.. That’s why I find these results so intriguing. I did probably the least that I could. And yet, I got a positive response, on something so small. Interesting!

  • […] — it’s finals week here at Georgia Tech, and grades are due Monday at noon.  My TA for Media Computation data structures had to leave the semester a couple weeks early, so I just finished catching up on all the grading […]

  • […] have now uploaded all the “video quizzes” from the data structures class I taught this last Spring, with links to the relevant videos […]

  • […] so much effort.  I do require students to do a weekly out-of-class quiz, often oriented around videos.  This semester, two of the quiz assignments were to use Problets on for and while loops.  That […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Trackback this post  |  Subscribe to the comments via RSS Feed

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 9,038 other followers


Recent Posts

Blog Stats

  • 2,012,742 hits
May 2011

CS Teaching Tips

%d bloggers like this: