Category Archives: accessibility

Wearable Fitness Trackers: A Comparative Usability Evaluation

This guest post is from graduate students Haley Vaigneur and Bliss Altenhoff. Haley and Bliss compared the usability of two fitness trackers as part of a graduate course in health informatics taught by Kelly Caine.

trackers

Wearable fitness trackers allow users to track and monitor their health. While these devices originated as a way for doctors to monitor chronically ill patients’ vitals, they have recently been developed and marketed for to a more general, health-conscious market. Equipped with advanced sensors such as accelerometers, users’ activity and sleep can be automatically tracked and then compared with their logged fitness goals and daily diet. Users can then use their statistics to help create or maintain a healthier lifestyle. Two examples of such devices are the Jawbone Up and Fitbit Flex, shown above.

Wearable technology is popular and has the potential to dramatically impact health (e.g. long-term health and activity data tracking, immediate syncing with Electronic Health Records (EHRs)). But these benefits can only be realized if the user is able to effectively use and understand these devices. This was the motivation for focusing on two of the most popular models of fitness trackers: the JawBone Up and FitBit Flex and their accompanying smartphone apps.

This study examined the usability of these two devices and their accompanying smartphone apps by having 14 participants (7 for Jawbone Up, 7 for FitBit Flex) perform a think-aloud test on five key features: Setup, Setting Goals, Tracking Diet, Tracking Activity, and Setting an Alarm. Participants then kept the wearable for three days and were encouraged to incorporate it into their normal routine. On the third day, participants completed the System Usability Scale survey and an informal interview regarding their experiences using the wearable.

Some of the key Jawbone UP findings were:

  1. Adding food or drink items was somewhat difficult due to unintuitive organization and unpredictable bugs. For example, one participant attempted to add a food item by scanning the bar code of a Lunchable, but the app added a Dr. Pepper to the log.
  2. Participants struggled to find the alarm settings, with one conducting a general web search for help to understand the Smart Sleep Window settings and how to save alarm settings.
  3. None of the participants were able to figure out how to communicate to the band or app that they would like to begin a workout. They didn’t realize that the Stopwatch menu option was intended to time the workout.

Some of the key findings of the FitBit Flex were:

Setting goals
Setting goals
wheretotap
What do I tap?
  1. Participants felt that the wristband (when using the appropriate sized band) was not uncomfortable or revealing and they were proud to wear it because it made them feel healthy.
  2. Users had a difficult time figuring out where to go on the app to set their health goals at first. Their instinct was to find it on the app homepage, or Dashboard, but it was under the Account tab.
  3. Some users had difficulty putting on the wristband, and several noted that it fell off unexpectedly. Users were also confused about where to “tap” the wristband to activate it, based on the instructions given in the app. The picture can appear to instruct the user to tap below the black screen, when the user actually needs to tap the screen directly, and firmly.
  4. Users did not realize that after turning Bluetooth on their phone, they needed to return to the app to tell the phone and wristband to begin syncing. They also noted that leaving Bluetooth on all day drained their phone battery.

    init
    Bluetooth confusion

Based on time per task and number of errors the FitBit Flex performed better than the Jawbone Up on the five tasks. Users’ ultimate trust in the data, willingness to continue using the wearable, and general satisfaction with each wearable was heavily influenced by their initial experiences (first day). The positive initial think-aloud results for the FitBit Flex were also consistent with a more positive later experience and stronger acceptance of the wearable.

This study found that there is still much room for improvement in the usability of the accompanying smartphone apps. A major concern for these kinds of devices is keeping user interest and motivation, which can easily be lost through confusing or cumbersome designs. By striving to improve the human factors of the apps simultaneous to the capabilities of the actual wearables, there is great potential for greater user satisfaction, and thus more long-term use.

While activity tracking wearables are currently most popular with more tech-savvy, active people, these devices should be designed to be used by all ages and levels of experience users. These devices could change health monitoring drastically and give people the power and ability to make better choices, and live healthier lifestyles.

Haley Vaigneur is a graduate student in Industrial Engineering at Clemson University. Her concentration is Human Factors and Ergonomics, emphasizing on research in the healthcare field.

Bliss Altenhoff is a Doctoral Candidate studying Human Factors Psychology at Clemson University, where she received her M.S. in Applied Psychology in 2012.  She is a member of the Perception and Action (PAC) lab, where her research is concentrated on enhancing human perception and performance by enriching perceptual display technologies for laparoscopic surgeons. .

ACKNOWLEDGMENTS
This material is based upon work supported by the National Science Foundation under Grant No. 1314342. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

Potpourri–Lazy Summer Edition

It’s summer and we (along with some of you) are taking a break.  But here’s a list of interesting usability/HF-related things that have crossed my path:

  • After much complaining, Ford is bringing back physical knobs in their MyTouch in-car controls.  Anne and I worked on some research (PDF) in our past lives as graduate students that directly compared touch-only interfaces to knob-based interfaces so it’s nice to see it is still a major issue; if only Ford read our 9 year old paper 🙂
  • Trucks driving under very low bridges is such a large problem in Australia that they are deploying a really novel and clever warning system.  A waterfall that projects a sign that’s hard to miss!
  • tags_finderApple will introduce their next version of OSX in the fall. One of the features i’m most excited about is system-level tag support.  Tags allow users to organize their files regardless of location or type.  I’m particularly interested in personal, single-user-generated tagging (compared to collaborative tagging like that used in flickr) as it appears to benefit older adults information organization and retrieval (PDF).  This pleases me.

Goodbye Mouse?

Story in the Washington Post about the impending demise of the computer mouse in favor of touch screens:

“Most children here have never seen a computer mouse,” said Hannah Tenpas, 24, a kindergarten teacher at San Antonio.

“The popularity of iPads and other tablets is changing how society interacts with information,” said Aniket Kittur, an assistant professor at the Human-Computer Interaction Institute at Carnegie Mellon University. “. . . Direct manipulation with our fingers, rather than mediated through a keyboard/mouse, is intuitive and easy for children to grasp.”

I realize the media needs a strong narrative to make an interesting story but the mouse is nowhere near dead.  The story is more complicated and completely depends on the task.  There are certain applications where the precise pointing afforded by mice are just too cumbersome with touch screens.

The article also has a great graphic describing how touch screens work and a short retrospective of input devices.

(post image from flickr user aperturismo)

Lack of human factors = more of your tax dollars at “work”

I live in Raleigh, NC. Our area code has always been a little problematic for the nationwide 911 emergency system – it is 919. But at least until now, dialing the 919 for a local call was optional. Looks like we’re finally big enough for ten digit dialing and we can expect to pay the price in our public safety system.  Check out this email from the Director of Emergency Communications, particularly the part about dispatching officers every 7.5 minutes to investigate hang-ups:

I am sure by now that you have seen or heard about some of the impact that the new 10 digit dialing requirement has made upon our 9-1-1 center. Unfortunately, we are almost three weeks downstream from this implementation, and are seeing few signs of improvement.

Neither the 9-1-1 center, the city, or the local telephone carriers are responsible for selecting area codes. They are distributed according to a national plan. “Overlays” are added when a region begins to run out of numbers in their original pool; in this case 9-1-9. Unfortunately, with the similarity between 9-1-9 and 9-1-1, our agency has seen this issue in the past, as some of our citizens have utilized 10 digit dialing for some time. The current impact on our staff – and on law enforcement – is that on our peak days we are dispatching officers to investigate hang up calls once every seven and a half minutes. Of course, this is a daily average, meaning that at peak times the impact is even more severe. Plus, we only dispatch calls that we can’t resolve another way. Many people who misdial don’t realize they have until we answer. Others hang-up, but answer when we call them back. In such cases sending an officer is not required, so the total number of calls we receive in error far exceeds those dispatched.

As Director of Emergency Communications, I am asking for your help. We have identified that a majority of such calls come from either senior citizens or business telephones. In the first case, confusion over the proper procedures seems to be the norm.  After 40 years, folks now have to dial 10 digits just to talk to their neighbor. We’ve had callers tell us they thought they had to now dial 9-1-1 before calling in our area, and others ask if they needed to dial 9-1-9 before they called 9-1-1. If you have an elderly friend, relative, or neighbor, I’d like to personally ask you to take the time to make sure they understand to carefully dial “9-1-9” when required. I believe that with some patience and understanding we can make significant inroads.

With regard to business telephones, the issue is a little more complex, and may in some cases even involve the need to dial “9” to get an outside line, followed by the unnecessary “1” before dialing the area code. Whatever the reason, it really boils down to just taking a few extra seconds to make sure of the numbers you’re dialing. Whether you work at a local business, or own one, can I please also count on you to assure that your co-workers use due care when calling? This is a very serious issue and takes resources away from dealing with actual emergencies.

So, to summarize:

  • There is a lack of understanding when to use 10-digit dialing.
  • Being “careful” is not going to fix this problem.
  • The added traditions for businesses to dial “9” to get out adds to the problem (NC State moved to a dial “7” system, presumably for this reason).
  • Those with a lifetime of 7-digit experience, and presumably the least likely to have numbers pre-programmed into a cell phone, make the most errors.

The issues here are fascinating, yet predictable. I don’t know if there is a perfect answer, since changing the long-term ill-chosen area code would be confusing (although my home town in Alabama has gone through 3 such changes in the last couple of decades – from 205 to 334 to 251!). But it is clear that we are penalized by the similarity of our numbers to a national standard for emergency calls. I applaud the tone of the email, which is not blameful – just desperate for a solution. However, I have great skepticism that advising “due care” in dialing will make any difference at all.

Older adults and Windows 8

In an earlier post we discussed how illuminating simple user testing can be. The video below is computer blogger Chris Pirrilo who put his dad in front of the new Windows 8 Preview. The dad seems to be relatively sophisticated and knows about Windows 7 but is completely flummoxed by Windows 8 new “Metro” interface.

Note that this is the reaction of just one person but we shouldn’t discount it. Plenty of users (both young and old) are not as sophisticated as you and I. I guess Anne and I (and other human factors & aging researchers) will still have lots of work!

(via Daring Fireball)

Revisiting an academic’s use of the iPad

With the introduction of “the new iPad” (i.e., iPad 3) I thought it would be a good time to update one of the most popular posts on this blog. That post was about incorporating an iPad into my daily work and play routine. It was written when the iPad was first introduced in 2010 and was mostly an exploration of some initial impressions and app suggestions from the perspective of an academic (non-student, higher education).

Based on the incredible popularity of that and the updated post it’s clear that many academics would like to incorporate the iPad into their workflow. My work is probably very similar to a generic office worker: lots of reading (mostly scanned journal article PDFs, writing, light note-taking, presentations, & data analysis.

In the years since I got first got the iPad, I’ve slowly learned what tasks can best be accomplished with the iPad and which should be left to the computer. I’ve also downloaded and deleted a large variety of apps whittling down until I find one (or three) that works best.

I’ve also since moved on to the iPad 2. It was a nice upgrade because it was dramatically thinner and lighter than the original iPad which made holding it more comfortable. The increased speed also made reading the scanned PDFs more pleasant. This is why I can’t wait for the iPad 3: more speed and higher resolution screen will significantly affect my most frequent tasks (see below).

This post is organized around my common work tasks and the apps I use most frequently. I don’t discuss the built in mail program, calendar, or web browser (which are heavily used).

Reading PDFs

Most of my library of thousands of PDFs are scanned journal articles. A smaller but growing portion of the newer articles are non-scanned PDFs that were created by the publisher. The difference is that the scanned PDFs are usually bigger and slightly fuzzier.

My original suggested app was iAnnotate mainly because of its ability to directly annotate PDFs with notes and scribbles. But I kept Goodreader for just plain reading because it seemed faster and more intuitive. Fortunately, Goodreader has kept improving and it’s now my most-used PDF application. The best feature is integration with Dropbox; so I only have to point it to a folder to download a semester’s worth of PDFs.

As good as Goodreader is, there are times when I need to move between PDF pages quickly and would like an alternative to page flipping. In that case I use PDF Expert since it has a nice birds-eye view of 9 pages but it just seems slower in page rendering.

Light Note-taking

I still use the iPad for light note-taking in meetings or by myself. I find it sufficient for most of my needs especially if you add a few accessories. In my previous post, I mentioned Evernote. I don’t really actively use Evernote much anymore. I can’t quite put my finger on it yet but it’s just not the right app/service for me. I notice that I tend to just dump things into it that I think i’ll need later but end up not needing.

Instead, I use a few note taking tools; none of which are preferred yet. The software keyboard is still sufficient for 80% of my needs. I’m able to type relatively fast and error free. For typewritten notes, I’ll use the built-in Notes application (which syncs to cloud services).

Simple, clean, and free!

When I’m traveling light (and I always am) but I know i’ll need to type out some e-mails or do some other writing, a great hardware accessory is the low-cost Amazon Bluetooth keyboard. It’s only about $35 (half the price of the metal Apple-branded accessory keyboard) and has a relatively nice feel for such a small keyboard. The great thing is that I only take it when I REALLY want a hardware keyboard which is not all the time.

Very thin, very light, about the same height as the iPad. It's slightly smaller than Apple's bluetooth keyboard.

On the rare occasion that I need to capture handwriting I don’t have a favorite app; instead there are 2 or 3 that each have something the others do not. As an aside, some people think they want hand writing but I’m not one of them. My handwriting is horribly mangled and unreadable unless I concentrate. Plus, handwritten notes are not usually text-searchable.

First, my usual app is called Notes Plus. It recently underwent a major upgrade with some pretty amazing features like split-screen viewing of a web page while you take notes and audio recording:

But I really hate the silver/metal look. I sometimes alternate and use Ghostwriter for handwritten notes or if I need to make a drawing:
Ghostwriter; zoom mode to write more precise text. The shaded zoom window moves automatically.

Both of these applications export their notes into Evernote, Dropbox, or plain PDFs. When I am handwriting (again, which is probably less than 5% of the time) I use a cheap stylus from Amazon.

Presentations

Finally, I’ve been editing presentations more on the iPad since switching to the Keynote presentation app on my desktop. When I need to organize my lectures or work on a presentation, the Keynote iPad app is surprisingly powerful but easy to use. I’m amazed that so much functionality could be built into a touch-only app:

I still use my laptop to actually give the presentation because I like to view the upcoming slide and the iPad currently just mirrors the current slide. I also use in-class clickers which require a laptop.

Other Useful Utilities

Finally, there are a few add-ons or apps that I find useful. The first is Wikipanion (yes, it’s OK to use Wikipedia). Wikipanion is a nice app front end to Wikipedia:

Wikipanion view of Wikipedia

The second, Offline Pages, is an app that allows you to download full web pages or websites for off-line viewing (e.g., on a plane).

Finally, there are times when you want to send a link or snippet of text from your desktop computer to your iPad. A useful app/service is Prowl. When you sign up for and then install the Prowl app and browser extension, you can send links directly from your browser to your iPad.

Another bonus is that once you sign up for the Prowl service and install an app on your desktop computer, you can also send text snippets from anywhere on your computer (e.g., a telephone number, address, paragraph of text) to your iPad.

What I Don’t/Can’t Do

Data analysis/Programming

Based on the number of hits the iPad posts have received from the following search term: “SPSS and iPad” there seems to be a bit of a demand…are you listening IBM?

To be honest, I don’t know if I want to be analyzing data on the iPad anyway. However, most data analysis is pointing and clicking so knows; who maybe some creative developer will create a data analysis application perfectly suited to a touch only interface.

I do a fair amount of programming and it would just be unbearable to do that on an iPad.

Usability Follies in the News

It’s election season which means more opportunities to point, laugh, and cry at the state of voting usability.  The first is sent in by Kim W.  As part of an NPR story, the reporter dug up a sample ballot. Pretty overwhelming and confusing (“vote for not more than one”??); makes me long for electronic voting.

Next, Ford is sending out a software update to their popular MyTouch car telematics system. The following NYT article is excellent in highlighting the importance of not only basic usability but that “user experience” is just as important as technical capability/specs.  The article lists a variety of usability quirks that should have been caught in user testing (e.g., “a touch-sensitive area under the touch screen that activates the hazard lights has been replaced with a mechanical button, because Ford learned that drivers were inadvertently turning on the hazard lights as they rested their hand while waiting for the system to respond.”).

My Touch (photo: NYT)

I am being facetious when I point an laugh but seriously, many of these issues could have been caught early with basic, relatively cheap, simple user testing.

“I think they were too willing to rush something out because of the flashiness of it rather than the functionality,” said Michael Hiner, a former stock-car racing crew chief in Akron, Ohio, who bought a Ford Edge Limited last year largely because he and his wife were intrigued by MyFord Touch.

Now Ford has issued a major upgrade that redesigns much of what customers see on the screen and tries to resolve complaints about the system crashing or rebooting while the vehicle is being driven. Ford said on Monday that the upgrade made the touch screens respond to commands more quickly, improved voice recognition capabilities and simplified a design that some say had the potential to create more distractions for drivers who tried to use it on the road. Fonts and buttons on the screen have been enlarged, and the layouts of more than 1,000 screens have been revamped.