Disclaimer: This post is really my first impressions on the stats we can draw out of Panopto. It may contain some inaccuracies due to my misunderstanding but please remember this is a blog and not a user manual.
In an attempt to promote the use of Panopto and to provide another example of how it can be used, I produced a screencast of me reading (or rather, commenting on) my regular monthly newsletter, which I produce in pdf format. I then emailed all staff a link to the Panopto recording and put a read receipt on the email message. The following day I sent out an email with a link to the newsletter which is stored on our VLE, Moodle. (By placing the newsletter there I can monitor access). Here’s what Outlook, Panopto and Moodle reported.
Email sent Wednesday 22 October at 16.05. to 496 people.
Read receipts received by Wednesday 29 October 10.20 = 65
(it’s acknowledged that some recipients who did open the message may have declined to send a read receipt)
I wanted to know how many people had viewed the video. According to the chart below there were 44 unique viewers. However fig 3, when you tally them up, shows 43. I am not sure why and maybe I;ve added them up wrongly. (I did double check that the Viewers in Fig. 3 were unique by the way).
However by hovering over the dark green bars you can see the number of viewers per day and the light green shows how many times the recording was accessed. This will be very interesting when looking at when students access lectures. Immediately the recording becomes available or juts before an assessment point?
Fig 2 tells us how many viewers watched particular parts of the video. So we can see that only 8 people saw what was being said at the 22 minute mark.
I guess this might be important when looking at lectures with slides because the tutor can see if there are spikes on particular slides. This might indicate that students are having difficulty understanding something or simply that they are viewing particular slides as more important.
I can’t see that it is particularly revealing about my recording though and since there was no navigation (i.e. no slide titles to signpost the recording) viewers probably arrived at places by accident when they fast forwarded (?).
There was a slight peak at 8.52 though and on looking at the recording that is where I brought into view a different window. So it might be that some people saw the preview as changed and so clicked on it and advanced to 8.52. (see Fig 2a)
This, I think, is very exciting and we need to show academics how they can use these stats to react to their students’ viewing behaviour and adapt their delivery accordingly.
The Viewers in Fig 3 are mostly anonymous because the permissions on the recording were set to allow open access so people did not have to log in to the Panopto Server.
In Fig 3 we can see that Panopto does not tally up the number of unique viewers, rather it displays the number of views against each user. So if a viewer watches the recording twice it is counted as 2 views. Also, the minutes viewed are tallied resulting in the minutes viewed sometimes being greater than the length of the recording (the recording was 24 minutes long) and meaning that we cannot tell if someone watched it all. Also, one viewer can be seen to have watched the recording only once yet the minutes viewed are greater than the length of the recording. This may be explained by the person pausing the recording (?)
So I think what this tells us is that probably only 11 people watched it in its entirety, 6 people watched between 4 and 15 minutes and 26 people watched for two minutes or less
The course on which the pdf version of the newsletter was stored does not allow guest access. All staff should be enrolled but if not, it does allow self-enrolment. So we can see who has accessed the reource through the Reports function.
20 people accessed the file. Of those only 8 were academics.
Looks like I need to find a different method of communicating, that’s for sure, but the purpose of this exercise was really to look at what stats Panopto can produce and I think this looks very promising. We will now have to look at an academic lecture with real students, and monitor what is being watched and when.