Designing websites for older users

According to the 2001 UK census , the UK now has more people aged over 60 than under 16. It also revealed that there are now 1.1 million people aged over 85.

Webcredible recently analysed and compared the results of 16 usability testing sessions – 8 of these sessions were conducted with elderly users (i.e. over the age of 65), and 8 with younger users (i.e. under the age of 40).

The 40-minute ‘talk-aloud’ sessions involved asking participants to find information on a range of government websites.

Assigning blame

The main finding of our study was that elderly users were more likely to assign blame when using the Internet.

Of the 8 elderly participants, 3 appeared to blame themselves for any difficulties which they encountered (sample quotes: “I don’t really know what I’m doing”; “It’s probably my fault”; “This always happens to me”). 4 of the elderly users, however, seemed to blame the site(s) for any difficulties which they encountered (sample quotes: “I hate it when websites do this”; “Well, that’s stupid”; “That doesn’t make any sense”).

We found that the younger group of users were far less likely to assign explicit blame for any difficulties encountered – with only 1 user from this group assigning blame (to themselves).

Emotional reaction

We also found that elderly users used far more emotive words and phrases when referring to websites than younger users.

All of the elderly users employed strongly positive or negative words in their remarks, such as “love”, “hate”, “stupid”, “helpful” and “friendly”. Indeed, one participant even talked to the website as if it were a pet (“That’s a good boy”)!

In contrast, only 2 of the younger participants expressed themselves in comparably strong terms (both when talking negatively about aspects of a site).

Weaker mental models

A very interesting finding was that 6 of the elderly participants regularly failed to scroll down a page (i.e. did not do so six or more times in a session). This failure led these participants to often miss information that was directly relevant to their task.

In comparison, none of the younger participants failed to scroll down a page six or more times in a session.

In our opinion, this is likely to be attributable to elderly users not having fully internalised the concept of browser-windows often requiring scrolling – a concept novel to computer-technology.

Technical language

We also found that elderly users were less likely to understand technical language. For instance, a moderator’s request to “bring up the minimised window” was not understood by 5 elderly users (in comparison to not being understood by only 2 of the younger users).

We found that elderly users were at least twice as likely as younger users to not understand the following phrases: ‘Homepage’, ‘URL’ and ‘Browser’.

Link identification

Our sessions showed that elderly participants were – as a group – more likely to click on elements of a page which weren’t links (an average of 14 times per session, in comparison to the younger participants’ average of 5 times per session).

It was also the case that all elderly users reported preferring websites that changed the colour of their visited links, whereas only 5 of the younger participants considered the matter significant.

Aversion to downloading

Of the 8 elderly participants, 5 expressed a strong aversion to downloading documents from the internet because they were “worried about bugs [i.e. viruses] and things”. None of the younger participants expressed such views.

Higher incidence of ‘search’ usage

Of the younger participant group in our study, only 2 individuals used the available search functionality, whereas 6 of the elderly participant group chose to make use of it.

It is possible that this may have developed as a means of compensating for their apparent difficulties/discomfort with traditional browsing.

It should be noted that all users expected a site to have a single ‘Search’ function that searched all of the site’s content.

Slow task-completion and reading

Our elderly participants required over double the average time of our younger participants to complete a task.

3 of our elderly participants also displayed a tendency to read all of the text on a page before being willing to decide on their next course of action. None of our younger participants did this.

Preference for ‘big and simple’ design

7 of our elderly participants reported anything less than 12-point type as being too small to read comfortably – and even though all users agreed that being able to re-size the text on the screen would be a good idea, only one of them knew how to do so through the browser.

It was also the case that all elderly participants preferred 800×600 over 1024×768 resolution.

Our recommendations

Although more research into the internet behaviours and preferences of elderly users is obviously required, we would like to suggest the following:

  • Designers should investigate innovative ways to communicate the fact that a page is not finished and requires scrolling
  • Technical terms should be avoided if possible – and where they have to be used, a clear explanation must be easily accessible (including examples wherever appropriate)
  • Links should be identified in a consistent and obvious way (e.g. blue, bold, underline, red on mouse-over)
  • The attention-grabbing features on a page (e.g. headings, pictures, icons, instructions and bullets) should be links
  • Visited links should change colour
  • Provide an HTML-version of as much content as possible and do not require users to install software (even Adobe Acrobat) in order to be able to access information
  • Make content as concise and clear as possible – consider providing two versions of the same content (‘simple’ and ‘detailed’) and allow users to decide which they want to access
  • Sites should provide a ‘Make the writing bigger’ link with accompanying illustrations/icons and always use high contrast to display text e.g. black text on an off-white background (N.B. using an off-white background is preferable to white because it reduces the chances of eyestrain for people who are slow readers)
  • Provide explicit instructions by using the imperative forms of verbs (e.g. ‘Go to more details on…’, ‘Find a…’, etc.)

Conclusions

Elderly users are an audience group that will grow in size and importance over the next few years. Our studies indicate that there are lots of simple things we can do to support their use of the internet.

We believe that these recommendations should be taken into account by all sites, and efforts should be made to further expand our knowledge of how to design for these users.

This article was written by Tim Fidgeon, Head of Usability at Webcredible. He’s crazy about usability and runs Webcredible’s web usability training and is passionate about user centered design.