If you use an LMS to manage your ELearning one of the report options might be completion times which might show the date someone was enrolled, when they actually started and finally completed a course. While some reports might report on the dates, others might be more detailed and go down to the time at the minute level. On first view this data might be useless, but not if correlated with other information. For example if you have a course that people keep calling you about either because it keeps freezing, or it’s difficult to access or the assessment is too complex, could you check if that course has longer completion times than other courses. You certainly can. Recently I was looking at some of our ELearning completion data just to see how it impacts two things:
- Meeting our compliance training goals
- The relationship between completion times and a particular course that could not be accessed directly through our company Intranet
I will answer both these questions in two or three posts. But in this post I will focus on:
- How I got the data
- Inital insights i derived from the data and the first question above.
How I got the data
Since we use an LMS I ran a report for data on course status from January 1 2016 to December 31st 2016. This gave me over 2000 rows of data which i think is good enough to derive some strong insights. The variables I was interested in were fields that showed:
- Dates people were enrolled on a particular course
- Dates people actually started the course
- Dates people completed the course
The data was downloaded into Excel. I also cleaned the data a bit to ensure I could do some analysis on it.
I used the three variables of data above to calculate:
- How many days it takes for staff to start their course after they have been enrolled
- How many days between starting the course and completion
- How many days between enrollment and completion
This was easy to do in Excel by just using a simple subtraction operation.
Once I calculated that I then did some summary analysis. This is shown Below:
The results were not too suprising:
- The Medication course was the one people were having difficulty accessing so I’m not suprised it took the longest to complete on average.
- I was suprised by the figures for health and safety because it is by far our longest elearning course, but not the longest to complete, although it is probably the most engaging course.
- I honestly don’t know why Food Hygiene and Infection Control are taking 34 and 33 days respectively to complete.
- Also I did a correlation analysis between Enroll to Start days and Start to Finish days. I wanted to see if there was a relationship between when a person started and when they actually completed their course. In other words if people start their course late are they likely to finish their course late also? The correlation figure was 0.565642729, which shows that there is an averagely strong relationship between the two variables. It’s important to point out that just because there is a relationship between them does not mean that one leads to the other, it’s that correlation is not causation thing, remember.
I also did some visualisation using a bar chart and scatter plot. The bar chart shows the average completion days for all the courses.
The scatter plot shows the relationship between the two variables used for the correlation analysis.
The scatter plot does look a bit weird (although there is a somewhat linear relationship between the two variables)so I intend to dig a bit more into the data to see if I can get some explanations for why it looks that way. So look out for the post next week.