Webinaring Bayes

Yesterday, I gave my first webinar, or online lecture if you want. The occasion was that a few people from another university asked me if I could give an introductory lecture about Bayesian statistics, and because traveling would have cost me a full day, I suggested doing a “virtual visit” instead. It’s a funny coincidence that Andrew Gelman did very much the same thing the same day, so I can draw from his experience as well.

First of all, no question, an online lecture is not as good as the real thing – after all, you are missing the most important thing about scientific visists, which is teabreaks and dinner of course. That being said, unlike Andrew, I was positively surprised how well the lecture worked. The only large annoyance, which Andrew mentioned as well, is that I couldn’t see the people’s faces properly (they only had a webcam for the whole room), so I didn’t see whether people were confused or bored, which is pretty crucial when explaining technical stuff. I think I managed OK with asking many times whether there were any questions, but for the next time I would like to have people with their individual laptops/webcams hooked up, that way it’s also easier to share material. Technically this should be no problem.

One difference to Andrews setup was that we were using Adobe Connect, a professional video conference software, instead of google hangouts or skype. As a member of a German university, we get access to an Adobe Connect server for free, and I can only say it’s worth the money, specially if you don’t have to pay. There are a number of advantages, for example you can upload your power points, and the slides are then not transfered via video, but via file, so they are crisp and clear at the other clients. People can even click on links you have in your slides. You also have some other useful tools such as drawing boards or the possibility to have people call in via phone (useful when doing video conferences with your project).

So, all in all I found the experience quite positive. I’ve been using video conferences for project meetings quite a few times before, but lecturing was somehow a different thing. You have to adjust your style, but I think it can work. Being there in person is of course better, but the point is that the other option would often not be to have a personal lecture, but to have no lecture at all. So, I’m thinking video lectures can be a good addition to the toolbox, in particular for specialized topics for which people that are interested would typically be spread around the world. Easier on our footprint as well.

Video conference screen with Adobe Connect

Video conference screen with Adobe Connect

4 thoughts on “Webinaring Bayes

    • Yes, we did a recording. I was thinking about posting it, but it’s a bit rough around the edges, e.g. the connection broke down 2 times, so in the end I decided for posting widely, it would be better to produce a version that is more polished, with better sound as well.

      If you want to have a look though, write me an email, I can forward you a link.

  1. I attended the Webinar and it was a really good and helpful introduction into Bayesian statistics. Thank you very much for this!

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s