Add Focus To Your Conversion Insights
Get 14-day free trial

Universities Website Usability Case Study – Berkeley far ahead 4 November 2013 | Natalia Jatczak

Website usability of universities is of utmost importance for both prospective and current students as well as the employees. That’s why we decided to have a look at how user friendly 8 of the American most prominent universities websites are.

Websites that were tested:

We wanted to evaluate the findability of pieces of information that we considered to be important for prospective students. The study was composed of four task and a short survey using Web Testing, Click Testing and the Survey from UsabilityTools.com:

  1. Find out how much you have to pay for one semester in the college.
  2. Find out if you can take any courses remotely via the Internet.
  3. Click on the element that makes you trust this page the most.
  4. Click on the element that draws your attention the most.
  5. What is your overall impression of the web site?

The study was performed by 800 respondents from the USA (100 people per website).

The results indicated how important it is to create clear menu labels and put them in the right place. The study also proved that displaying too many pictures on the page is not the best choice usability-wise.


Results & Interpretation

Task Success Rates

The first task was classified as successful when participants found a page with the exact cost of the tuition fee. The results showed that not only was it difficult to find this information on the websites, but, frequently, it was even impossible.
The highest success rate was noted on the websites of Berkeley (79%), Columbia (62%) and Princeton (53%). The task turned out to be problematic on the websites of Yale (4%) and CalTech (15%).

The results of the task 'Find out how much you have to pay for one semester in the college.'

The results of the task Find out how much you have to pay for one semester in the college.

The pages, which had the best score had the clearest labels in the menu. In general, the more specific the label, the higher success rate.
For example, Berkeley had ‘Fees’ label in the main menu, which was very easy to find.

Clear and specific label in Berkeley's menu

Clear and specific label in Berkeley’s menu

Most of the other universities tried to ‘hide’ the information behind ‘Financial aid’ label, so it wasn’t obvious that information about actual costs would be found there. Columbia, however, added the word ‘Billing’ which could have helped the respondents find the right page and the high success rate (62%) might be an evidence for that.

Label in Columbia's Admissions page

Columbia’s Admissions page

Results from Yale’s website showed that it is crucial not to put two identical labels concerning different things on one page. On Yale’s ‘Prospective students’ sub-page there were two ‘Financial aid’ labels, one concerning the actual Yale college and the second one concerning other schools within this institution (see the illustration below). 16% of respondents have chosen the wrong one and, as a result, found the tuition fee from another school than they were supposed to. Not only did they not find the correct information, but they were also convinced that the wrong cost they found was actually right.

Two identical labels on the same page

Two identical labels on the same page

The second task was to find out if one can take any courses remotely via the Internet. Usually, there were several pages within the websites that included the information about online courses.

The results of the task 'Find out if you can take any courses remotely via the Internet.'

The results of the task ‘Find out if you can take any courses remotely via the Internet.’

Again the highest success rate was noticed on Berkeley’s website (59%), followed by Harvard (58%) and Princeton (50%). These pages were so successful, because information about online courses was easy to find, usually directly from the homepage.

Columbia (19%) and Caltech (18%) had the lowest success rate. When it comes to Columbia, the low rate might have been the result of ‘hiding’ online courses behind the name ‘CourseWorks’. Even though the page with this label was easy to find, the name wasn’t self-explanatory.

In the case of Caltech, the online courses were labeled properly but put in the ‘picture-menu’ on the main page (see the image below).

Elements in the picture menu were not easy to spot

Elements in the picture menu were not easy to spot

Merely 18% of success might suggest that this kind of menu isn’t the best choice. It may look visually-appealing and information may seem to be easy to find, but it actually isn’t.


Click Testing study

Respondents were asked to answer two questions about the website using Click Testing. The first one was to click on the element that draws the respondent’s attention the most and the second task was about clicking on the element, which made one trust that page.

Screen showing spots that drew respondents' attention in Click Testing study

Screen showing spots that drew respondents’ attention in Click Testing study

The results showed that what made people trust the page were the logo and contact details and pictures drew the most attention. But when there were too many pictures, as in the case of CalTech, the attention was split between them and not focused on any particular element.


Positive and negative rates

At the end of the study participants were asked to complete a survey task. They had to choose adjectives that they would use to describe the visited website. When analyzing the data, we put these adjectives into two groups:
– positive values (convenient, attractive, organize, interesting)
– negative values (confusing, unattractive, chaotic, boring)
Judging by the amount of respondents who chose particular adjectives, we can say that they were primarily looking for websites that are “attractive” and “organized”. The word that was chosen the least amount of times was “unattractive”.

Generally, there were more positives than negatives for all websites except CalTech. The reason might be that it wasn’t easy to find any information in the picture-menu. On the page there was also a hidden menu (see the image below), which could have not been even noticed by the respondents.

The menu label on CalTech website is not easy to find.

The menu label on CalTech website is not easy to find.

All in all, it was the Berkeley’s website that proved to be the most user-friendly in all of the tasks. Interestingly, this website had also the least number of pages visited by the respondents before they finished the tasks.


What insights from this study you can implement on your website?
  • Name menu labels specifically and make sure they correspond with the actual content of the sub-page
  • If you have to put two labels of identical wording on one page, find a way to make them look different and make sure that users see that difference
  • Never hide the important information. You might be tempted to cover unpleasant information (e.g. actual costs of your services) under misleading labels, but that could only discourage prospects from using your website.
  • Don’t use too many pictures on the page, if you want them to carry particular pieces of information. The labels get lost as the user attention is split between too many visuals.


Natalia Jatczak

Written by Natalia Jatczak

She is a Cognitive Science lover with a bag full of hobbies including reading, watching movies & sitcoms, working out and traveling. Time management is her specialty, apparently.


Ready to empower your website's potential?