Reflection On Guerilla User Research

10 Oct

Last week we conducted a Guerilla UX research in our CGT512 class. The purpose of our interview is to examine the usability of CGT graduate studies website and provide recommendations for COT web team. Our primary user group is potential international graduate students. After brainstorming, we developed a list of questions and interviewed five current Purdue graduate students. There are some worthy questions to explore during the whole process.

1. How to structure our questions

The first thing that we should consider is what kind of question to choose, open-ended or closed-ended? In order to get a great deal of specific information, open ended questions should be given priority. While the interviewees are not influenced by the list of responses, they may be influenced by the order of questions. In my opinion, we should group questions in topics and then pose them in a logical order, from simple questions to difficult questions. Starting with some simple questions will motivate respondents to participate in the interview. Also, when determining the order of questions within each topic, I think we can imagine that we are navigating the website from top to bottom, from left to right meeting the users’ common habits. What kind of experience will we have? Do we have any problem? What do I like or dislike?

Meanwhile, we contain few closed-ended questions, where the respondent is asked to select answers from a brief list (e.g., “__20-25 ___ 25-30”). It is very useful for the demographic questions, especially asking about their ages. These demographic questions such as gender, income, education can be put at the end of the questionnaire. In doing so, it helps avoid alienating or worrying participants.

Secondly, the wording of a question has a significant impact on responses. When doing our face-to-face interviews, they are vulnerable to bias. For example, more people answered “less helpful” to the question “How helpful is the website in finding information you want?” preceding it with the question “Did you have any frustrations when browsing?” Unconsciously interviewer bias occurs when something bad is to ask with negative behavior or comment. The response can be influenced a great deal by the wording of the question. The interviewers should be very careful when choosing words and phrases in a question.

2. Focus on broader user goals or on specific work information

I think first we should understand users’ motivations and end-goals while keeping an eye on the broader context. The goals should be broad enough to cover the full range of usage patterns and needs. For example, the goal of potential graduate students who come to CGT website is to find useful information for their application. Current graduate students visit the CGT website in order to get support for their current study. After deciding the target group, we should speak to people who can represent the likely experiences, contexts or goals which our real users will have. Then we can focus on specific work information based on everything you can find out about the representatives. Below are some specific questions we can think about.

What are these person look like?

What do they like or dislike?

What goals have they already achieved?

What goals do they wish to reach?

What kind of things will interest them and give them a wonderful experience?

What kind of things will upset them and give them a bad experience?

How often do they use the website?

What kind of devices do they use to access the website?

With goal-oriented and user-centered design, the specific work information will help us find out deep-seated motivations and how they affect the overall experience. After articulating and synthesizing the experience people interact with the product, we can define the key problem and clarify the solution so as to satisfy users goals.

3. Translating interview data to design requirements

Because of the limited time and resources, we use convenience sampling method to collect data. A small sample selected from our class is only representative of particular individuals, not the whole population. If we want to make an inference for the entire range of variation, we should capture data with stratified sampling or cluster sampling method instead of convenience sampling method. It can reduce the risk of biases and develop a more valid explanation and general method.

Additionally, in the qualitative research, translating interview data to design requirements is not a mechanical translation process. Itrelies not only on our interview questions but also on the specific situation. Issues are interconnected and must be considered holistically. That means the design requirements generally depend on the integration of data from a variety of questions and sources of information. For example, almost all the interviewees suggested that information about current students, Lafayette life or workshops will be very helpful. When applying these design requirements, we should also consider usability requirements translated from other data such as good visual design and easy navigation.

All in all, the experience of applying the readings to a hands-on activity gives me an excellent opportunity to work through several of the most important issues in designing Guerilla UX research. Begin by thinking about our goals for this research and brainstorm whatever comes to our mind relating to the interview. Then identify, develop and organize some significant questions that worth answering. After that we can seek out people who would like to be users. During the interview, watch and talk to them, encourage subjects to think aloud. At last, analyze the user interface research study data with insight and creativity. Oh, most importantly, do not forget to share our questions and reflections on these experiences and that is what we are doing!

Leave a comment

Posted by on October 10, 2012 in Thoughts & Ideas



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: