My junior year at Ohio State University, I had the opportunity to use an eye-tracking technology. I was studying Evaluation and Usability Testing as an undergraduate student, and had no idea previously that such things existed. That was actually my first (proper) introduction to usability in general, and I did not know at the time that I would be pursuing the field for my career.
It was a pleasure revisiting the subject for our final assignment over these past two weeks. I enjoyed clicking around on the Tobii site to see how far the technology has come. Of course, there have been many developments in recent years, but I did not realize the broad spectrum of eye-tracking tools available, from eyeglasses to specialized displays. It makes me excited for the future of eye-tracking studies.
Our assignment was a great opportunity to dip a toe in the water with eye-tracking studies. I liked how we evaluated one task with five participants instead of a more comprehensive test. This allowed me to really see the potential in what eye-tracking can do for usability. Although it’s not appropriate in all cases (our assignment’s a decent example) it can provide practitioners with priceless information when used properly.
On a personal note Mr. Roll, thank you for an enjoyable term!
As I progress further along into my UX program with Kent, I become increasingly aware of how important a designer’s portfolio is to their success. As someone with little applied experience, I’m always interested in seeing how people choose to portray themselves and their work while I figure out the best way to represent my work. The design of the portfolio itself is an important part of displaying your abilities after all, and each person’s style is different.
This week I was grateful to come across a great resource to have while developing a portfolio. The article gives advice to UXers after having reviewing several thousand portfolios submitted for consideration to the company Availo. Many things I had not given much thought were among the suggestions made, such as how to introduce yourself effectively. It sounds so simple, but it is easy to understand how it can be overlooked by so many. As with many things in our field, applicants were encouraged to keep it brief. One aspect that the article took a sharp stance on was skill charts, saying they should be avoided at all times. It is a trend that I noticed based off some portfolios I reviewed, and one I’m now glad to have one employers opinion on them. They argue that it is not meaningful to say “I’m x% photoshop” for example, and advocate for making the portfolio about the work.
As I continue to take on projects and plan my portfolio design, this will be a useful reference to look back upon.
When I began my mobile test report, I already had an idea of the direction I wanted to head in. I began selecting hardware last week while we reviewed past solutions for recording, and used what I came across to design my simple camera case. I found the GiraffeCam, and endoscopic camera with a flexible wire cable, and I knew quickly that it would be perfect. It is very low profile, lightweight, and easy to point due to the cable’s flexibility.
So in photoshop, I quickly threw the images together to see if I could create a sensible prototype. Going through this process uncovered several design flaws: the camera did not record sound, and since I knew I wanted it, I didn’t have a way to record the participant’s face at the same time as the device screen. Using a second GiraffeCam pointing back at the user was not an option as it would have been an unsettling distraction. It then occurred to me that I could unobtrusively capture both participant’s facial expressions and session sound with the built-in camera and microphone on a MacBook Pro. I used a creative technique that screen captured two windows that were in video preview mode: FaceTime and QuickTime video record. I tested the setup myself and got remarkably good results considering it would require less than $100 in hardware. I plan to build in the exact camera case I depicted in my assignment and look forward to getting some real world experience testing for mobile.
This week we began reviewing various configurations for conducting mobile testing. I particularly enjoyed watching Jenn Downs’ presentation on her past solutions and advice for testing users on mobile devices for Mailchimp. She leveraged a simple plastic clip and an adjustable USB webcam in order to record participants on their smartphones while they tested their products. She also documented countless other solutions that practitioners had come up with in the past. Examples ranged from backwards-facing MacBook webcams to more elegant solutions, involving many pieces coming together in a case.
Mailchimp is a product I have used extensively in a past sales role, and I was always impressed with how polished and user-oriented their web applications were. It was interesting to learn firsthand how some of their testing solutions were crafted. It was both funny and informational to watch some of the more resourceful approaches that were taken with capturing mobile users. It is easy to imagine a situation where you may not have all the necessary tools in order to easily record participants, and occasionally need to get creative with the occasional MacBook webcam. I am confident that we will not have to rely much longer on external cameras to capture usability testing however. There is already functionality built into iOS allowing screen capture with microphone recording, and it is only a matter of time before there are better solutions for conducting testing. It is very unnatural (and no doubt distracting) to have a physical camera rigged to cell phone that you are being tested on. The more that capturing participants is kept in the background, the more reliable your data will be.
This week we finally got to put our tasks into action, and begin to make sense of the data we had been collecting. As someone with an appreciation for metrics, I really looked forward to working on this assignment. I was lucky enough to find five participants, however one could not be considered due to technical difficulties with Validately- their submission never went through. I expected more technical issues than I ended up getting based off my first experience with Validately, and I am grateful for that.
I did learn one important lesson with this assignment. One individual I recruited, a friend of mine, thought he was evaluating a design of my own. His feedback was very flattering as a result, and the fault is my own for not being more careful while preparing him. He still provided very valuable feedback (even some unsolicited recommendations) but it seemed that his Likert responses were affected. In the future I will be more clear with participants, and encourage them to give honest and even critical (if applicable) feedback.
I continue to be fascinated with what a fresh pair of eyes can uncover, as each of my participants had their own unique observations, most of which discussing things I had not noticed myself. Using different methodologies to conduct user testing is a skill that requires a lot of time and practice to develop. I look forward to having that opportunity as we continue with our coursework.
For week two we went through the steps of creating an unmoderated test online. My test is being conducted on a smaller online company, StoreHouse tea. Validately is an impressive platform with many different ways of obtaining data. I’m hoping my tasks will help to reveal how navigable the existing interface is when trying to perform directed tasks. It was a challenge to prioritize what areas of the interface to test, and how to go about doing so. I look forward to seeing how participants are able to do.
As election day approaches, I found myself wondering about the UX behind perhaps the single most important application- balloting. Moving the voting process to a digital platform is a controversial conversation, and a move that’s unlikely to happen in the near future. There are understandable concerns with security. But there should still be a conversation about the user experience behind casting a vote. After all, the 2012 election attracted only 58% of Americans as voters. I think user experience is one of many factors that tie into this low turnout, especially when younger voters are the most absent. Our ballots are not exactly straight-forward, and the internet and mobile software has spoiled people with a refreshing movement of uncompromisingly-minimal design. Going from one to the other can be a shock to users. With some minor touch ups to our existing system, we can get more people to participate in our political process.
Reviewing recruiting platforms this week was a useful exercise for me. Being fairly new to User Experience Design, I am unpracticed in evaluating many of these services. I have looked at many SaaS platforms for different applications, but Mechanical Turk was something I was completely unfamiliar with. It’s very easy to see how useful this service can be when time is limited for recruiting.
Conducting testing requires you to decide on a method. Face-to-face and remote testing have their own advantages depending on the resources you have available. For F2F testing, the data you get is very rich because you have the opportunity to ask questions and see body/facial expressions. These reactions are very important and F2F testing allows you to capture it during recorded sessions and while taking notes. It requires strong moderation but face-to-face testing can yield very meaningful data from just a handful of participants.
Remote testing can be effective as well but there are drawbacks. You cannot actively question participants, and many times sessions are not recorded. Some participants will also do a better job of ‘thinking aloud’ than others. Testing this way is cheaper because it doesn’t require participants to be on site or take much time, and because no researcher is needed. This allows for many tests to be taken in a short time frame.
If I were in a situation with unlimited time and resources, I think face-to-face testing is always going to be the better way to go. The reality is that many situations won’t be that flexible, and remote testing can be an effective alternative.