Setting up Google Analytics for Moodle
There is no plugin as such to install Google Analytics on Moodle 2. As explained in this blog post, Moodle URL’s are not Google Analytics friendly and you need to apply a simple tweak to your server for it to work well. It is fairly simple to get it going following the instructions put together by Bas Brand. You can read more about Google Analytics and Moodle here.
Overview
| Gamified course | Non-gamified course |
|
|
The students enrolled in the gamified course continue to generate more pageviews than their counterparts in the non-gamified course (7,529 vs. 6,559 or +15%). The spikes represent days when students had Computer Technology lessons (the scales are different in both graphs). The most viewed pages data roughly coincide with my previous findings. Students have mainly been working on their designs lately, and have not needed to use Moodle very much, aside from updating their checklists, hence the low level of pageviews for the past 10 days or so.
Technology
Until this post I have mainly relied on the Moodle database to find out what my students are up to in my courses. Whilst it is the method of choice for this type of project as the data is extremely accurate, there are a few tools that Google Analytics offer that the database method cannot match, for example reporting on the type of technology my students use to access Moodle.
Browser
| Gamified course | Non-gamified course |
|
|
It came as no surprise that overall most of the visits were done so using Safari (61% vs. 47%), as my computer lab is equipped with iMacs. I am surprised with the differences between the two groups, especially considering Google Chrome (21% vs. 51%) is not installed on the school computers.Are the results ‘skewed’ as one of the top users uses Chrome? Is there someone who has influenced students not to use Internet Explorer? More investigation is needed.
It is also possible with Google Analytics to find out the browser version used (although not for a filtered set of pages), and I discovered that quite a large proportion of our students use an outdated version of their web browsers. This sort of data can be used to prompt a discussion on ‘staying safe on the Web’ with your students.
Operating system
| Gamified course | Non-gamified course |
|
|
Again no big surprises here, with a vast majority of the visits done using Mac OS (76% vs. 64%). However I was surprised with how few students access Moodle using a mobile devices/tablets (less than 1% in both courses). One point to keep in mind: if you have resources that use Flash, iOS users won’t be able to access them.
Screen resolution
| Gamified course | Non-gamified course |
|
|
Again no major surprises here as 1920 x 1080 pixels is the default screen resolution on the 21″ iMacs in my computer lab. This setting is useful when you design courses. For example, I was worried that quite a few students wouldn’t be able to view the ‘Design Cycle’ image at the top of my course as I thought it may be too wide. This clearly isn’t the case for the vast majority of my students.
Student flow
Navigation
| Gamified course | Non-gamified course |
|
|
The ‘Navigation summary‘ feature in Google Analytics shows the pages users visited before and after a specific page. For example, I wanted to know whether students have been using the navigation menus to access the course homepage, or whether they used ‘My Moodle‘ page, the category route or a shortcut (roughly). As I expected, most of the pageviews were generated by students who used the navigation menus and shortcuts. There seems to be quite a few students who keep using Moodle the ‘hard way’ and click on categories & sub-categories to reach courses.
Traffic sources
| Gamified course | Non-gamified course |
|
|
A majority of students accessed Moodle entering the address in their Web browsers (65% vs. 55%). A significant proportion of students searched for Moodle using their search engine (23% vs. 33%), and just over 10 % used various links to access the site (e.g. link available on the school website, in emails, etc.). I suspect the difference in keyword searches is due to more students in the non-gamified course using Google Chrome as their web browser, where the ‘address bar’ and ‘search bar’ are in the same location (the version of Safari we use doesn’t have this feature), although I cannot be sure.
Student behaviour
Hours of visit
| Gamified course | Non-gamified course |
|
|
The busiest hours are during lesson times, not too surprised here. However, I think too many students are doing their homework too late. I am aware that students in Hong Kong are under a lot of pressure with extra-curricular activities (private tuition, etc.) but I am not happy to see 12/13 year old students working past 10pm. More investigation is needed.
Conclusions
Although I have found this an interesting exercise, it seems as though Google Analytics is best used on large data sets, where accuracy is not paramount and trends are more important e.g. whole Moodle course, categories, etc. Whilst it may not be completely appropriate to use Google Analytics for the experiment I have been running, I find it extremely useful as a Moodle administrator. There are some metrics that I find intriguing, such as ‘time spent on page’, and I will need to do more research on the topic to judge whether this metric, amongst others, can really be trusted.