Design Patterns
for Mobile User Interfaces
Targeted at Older Adults

← Back to list of patterns

Where do these patterns come from?

Research project

The patterns presented on this website are based on the results of a research project conducted during a master’s thesis dissertation. This research project aimed to investigate gesture-based interaction, on small touchscreen devices, with older adults. In order to do so, several phases of testing with users were carried-out, and consequently all the patterns found on this website are based on results of research conducted with older adults. In addition, as you will find, many of the patterns make references to, and are supported by, research efforts conducted by other authors regarding older adults and gestural interaction.

Phase one

The research project, that enabled the creation of this set of patterns, was conducted in four distinct phases. The first phase aimed to assess the suitability of existing gestures for older adults, and gave place to the pattern SMARTPHONE GESTURES FOR OLDER ADULTS. In this phase we conducted a study based on the method presented in (Wobbrock, Morris, & Wilson, 2009), where we would first show the intended consequence of a gesture to participants, and we would then ask them to perform a gesture they thought might result in the shown consequence. For example, we would show an animation of a geometric figure moving from the bottom toward the top of the screen, and then we would ask participants how they would make the figure move like what was seen in the animation (the existing smartphone gesture for this action would be a swipe to drag the figure across the screen). Our results revealed that, in most cases, older adults were not able to discover existing gestures, nor were they able to effectively use gestures to solve common smartphone tasks, such as scrolling, panning, zooming-in, or zooming-out. Nonetheless, the gestures that seemed to be the most intuitive were tap and swipe. This phase of research was conducted with twenty older adults, from several day-care centres and retirement homes, with mean ages between 62 and 89 years (Mean = 74.2).

Phase two

Next, the second phase of research aimed to assess whether we could effectively make use of animated tutorials to teach common smartphone gestures — tap and swipe — to older adults. In order to do so we would first show animated tutorials of gestures to participants, and then ask them to use those gestures to solve a set of typical smartphone tasks. These tasks could be, for example, to scroll, to pan, to select and item, to drag an item, or to stop a scrolling list, and were based on the official smartphone OS definitions of gestures (see gestures for: iPhone, Windows Phone, and Android devices) and their corresponding functions. Once again, this phase of research was conducted with twenty older adults, from several day-care centres and retirement homes, with mean ages between 60 and 90 years (Mean = 74.3 years old). Overall, our results revealed that the introduction of animated tutorials did in fact enhance older adults performance of tap and swipe gestures. Accordingly, the results of this second phase of research are documented in the pattern DEMONSTRATIONS OF AVAILABLE GESTURES.

Once we had assessed the learnability of existing smartphone gestures, we aimed to evaluate certain interface characteristics that could influence the performance of tap and swipe. We outlined the following interface characteristics: (1) target sizes for both tap and swipe gestures, (2) spacing sizes between adjacent targets, and (3) target’s onscreen locations. In order to investigate these issues, a further two distinct phases of research were conducted — phases three and four.

Phase three

Phase three aimed to assess the influence of target sizes, and spacings dimensions between adjacent targets on older adults performance of tap and swipe gestures. This study was conducted with forty older adults, with mean ages between 65 and 95 (Mean = 76.88). We evaluated five target sizes: 7 mm, 10.5 mm, 14 mm, 17.5 mm and 21 mm; and four spacing sizes between adjacent targets plus a single target condition: 0 mm, 3.5 mm, 7 mm and 10.5 mm. Our results revealed that both conditions did indeed influence older adults’ accuracy in acquiring targets, as well as the time they needed to do so. The results from this phase of testing are documented in five patterns: (1) RECOMMENDED TARGET SIZES FOR TAP GESTURES; (2) RECOMMENDED TARGET SIZES FOR SWIPE GESTURES; (3) RECOMMENDED TAP TARGET SIZES FOR LIMITED SCREEN REAL ESTATE; (4) RECOMMENDED SWIPE TARGET SIZES FOR LIMITED SCREEN REAL ESTATE; and (5) SPACING BETWEEN TARGETS FOR TAP AND SWIPE GESTURES.

Phase four

Next, in phase four our objective was to assess the influence of target’s onscreen locations on the performance of tap and swipe gestures. In order to do so we created two individual grids, one for tap targets and one for swipe targets. The grid for tap targets had a total of 28 locations, covering the entire smartphone display. The grid for the swipe targets was comprised of a total of 11 grid locations, covering the entire display, and gestures were performed in each of four gesture orientations (e.g., left-to-right, right-to-left, top-to-bottom, and bottom-to-top). Once more, this phase of testing was performed with forty older adults with mean ages between 67 and 95 years (Mean = 77.24). Accordingly, design guidance based on the results of our tests can be found in the patterns ACTIVITY ZONES AND TOUCH OFFSETS FOR TAP GESTURES and ACTIVITY ZONES AND TOUCH OFFSETS FOR SWIPE GESTURES.

Conducting your own usability tests

Finally, our participants were all Portuguese older adults with low technology-proficiency. If our study were conducted in different cultural and socio-economic conditions, our results would indeed be different. Therefore, these design patterns do not substitute conducting usability tests with real users. They are rather a starting point that intends to provide guidance in the initial design phases of your project. Accordingly, testing your product further along the development lifecycle, will still be necessary.

* * *

References