UX Design and Hacking Survey Models

To the dear friends who watched me squirrel away into working on my TED talk and do my “homework” for a mysterious class… here is my confession.

I am working on something really cool. Something really cool, that I hope will change the way we do data collection for especially complicated fields like… informal economies.

One reason I decided to come work in New York City instead of going back to Mexico or Colombia for more research is that I wanted to learn more about smart design and business models. I started taking the General Assembly User Experience design course this fall to address my interest in design… and smart research. I like the way designers think and problem solve, sketch, test and produce work that is meant to appeal and engage with you. UX goes a step beyond that and designs based on input and existing behaviors observed in their customers. US designers notice how people interact with technology and images on small iphone screens or larger web based programs. I like how carefully it combines new design features with existing behavior patterns.

UX designers create products with specific goals for their users in mind… their job in general is to advocate for users! The types of questions we discuss in the course are about aesthetics, mechanisms that can be learned by playing with the program, and using existing behavioral cues to help people navigate software. Maybe the best line of the introduction class was:

“The best art makes your head spin with questions. Perhaps this is the fundamental distinction between pure art and pure design. While great art makes you wonder, great design makes things clear.” – John Maeda

I am spending a few hours a week sketching and thinking through the details of survey design, from the point of view of the individual administering the survey and the person receiving the survey. For especially complicated data sets, I think this will really make a difference.

In my time in the field, I have tried a variety of different information gathering techniques. I canvassed for several of the past election cycles while I was a student in New Haven, I ran surveys on the informal economy and urban development in a dense Mexican city and again in a rural town of 350 people. I ran a second series of surveys on the undocumented population in Boston and New Haven to understand public opinion on the Dream Act as it was going up for a vote. These are just a few examples, but each of these events had different goals and audiences in mind.

I learned a few things early into each process that I think are important to keep in mind throughout this redesign process:

1) Surveys CANNOT be too long. Short and sweet, figure out in advance what information you need and what holes are missing from the data set you are trying to build. For some of these longer conversations, explain that it will be a longer conversation and not a standing-in-someone’s-door-way-this-will-be-really-quick-I-promise conversation

2) If you have to explain the question beyond basic clarification questions, this piece of the survey has failed and will probably not yield great results from this particular question

3) the advice I received (if you are quiet and wait long enough, people will keep talking and providing more info. Sometimes information that is more useful or relevant than you would have thought to ask for) is particularly relevant for Americans, who really do tend to fill silence when they are uncomfortable, and not applicable across the board.

4) Just like business negotiations, the best case scenario for surveys is when both parties have time to make small talk and get comfortable with one another. The answers tend to be more authentic and it is possible to explore beyond the initial questions the survey listed going into the conversation. This is also a difficult piece to navigate.

For these next few weeks, I am thinking through technology and design opportunities for survey design. It’s been a fun process — much more zoomed in to specific locations for buttons, gestures, design features, information storage and organization, and a number of other things that I took for granted before this course. But so far, the results seem promising.

Published by


Tea, Tequila, and informal economy enthusiast.

Leave a Reply

Your email address will not be published. Required fields are marked *