There are various approaches that could be taken depending on a number of factors e.g.

Who are the users you are evaluating it for - students accessing content, staff using it to support students etc..

Do you have access to the user group for user testing? If not then inspection based methods such as heuristic evaluation may be appropriate using e-learning heuristics such as those proposed by Squires and Preece (others are available if you do not have access).

For user testing I think direct observation would be useful or retrospective think aloud as this is unlikely to impact on them using the system.

There are lots of studies looking at usability of e-learning systems and it might just be worthwhile reviewing there methods.

For usability evaluation in last few years a lot of work has been done. The evaluation methods for usability are Surveys, think aloud, Hypothesis, LSP, Mile+, Heuristic evaluation (HE), Cognitive walk-through, Systematic Usability Evaluation Methodology, e-Learning Systematic Evaluation Methodology, Milano-Lugano Evaluation method, Pedagogically Meaningful Questionnaire Learning these methods are used for evaluations in last 10 years for e-learning usability and web based learning usability.

Your question is not specific enough. What are you trying to prove? That one web based learning application is easier to use than another. This does not say that subjects are able to acquire more skills through an easier to use application as learnign depends on the content proposed by the application. 

While there are numerous ways to measure usability, once you have measured the usability of an application, what will you do with the results? I guess, you would recommend the use of an easier to use application. However, you need to ensure that the contents in the two applications, you are comparing, are the same to be able to make your recommendation. 

In one of my publications, I measured the usability of two versions of a web-based info retrieval application. In one version, some usability heuristics were implemented while in the second application, all web usability heuristics were implemented. Through usability trials, the study tried to identify if the extra efforts to implement all usability heuristics had a significant positive impact on usability. The study showed that it took less time to find the information in the second version. However, subjects committed more errors using the second version. Moreover, subjects had higher "Ease of use perceptions" for the first version. Finally, there was no significant statistical difference between the three measures (time, error EOU perceptions) between the two groups. This led to the conclusion that the extra efforts was not worthhwhile for an application developed without user participation and user task analysis.  

The above study shows that while usability experts will find the second version as easier to use due to the implementation of all user interface design heuristics, subjects undertaking a usability trial find the first version as easier to use. So which of the two applications should we recommend to users? To find out, as well as usability, other criteria should be measured.  

Dear Siddig,

Refer the following article.

https://globaljournals.org/GJCST_Volume15/3-Usability-Evaluation-of-Learning.pdf