Back...

Methodology I

There are a few things to consider before we discuss methodology. We have clearly identified the purpose and the problem of the study in previous entries. However, what are we testing? Who are we testing? What steps do we need to take in order to meet our efforts and company budget with this testing? My favorite book to reference in user experience research and design is Jakob Nielson’s Usability Engineering (1993). Yes, the book is dated back to 90’s but I believe that it provides a fundamental guideline for building interface and usability testing. Even though it has been 20 years since this book was published, most of the building interface process is analogous from then and now. Furthermore how humans interact and react to emerging technology really have not changed. It is human behavior. We just need a way to integrate current technology to the existing methodology. We can accomplish this task by utilizing applied research methodology. Applied research study allows us to solve a practical world problem and allows researchers to obtain reliable results.

Participants

Who are our intended end users? Who uses our website to reserve plane tickets and vacation packages? TravelhorizonTM has revealed that from 2008 to 2009 approximately 42 percent of travelers are adults and among these travelers 48 percent travel for business purpose. Furthermore, there has been a drastic increase in international travelers. From year 2000 to 2008, approximately 58million international traveler visited The United States. For the purpose of the study, I suggest we should omit international traveler as our targeted participants. It is one less variable that we have to consider in this study (i.e. language, content, culture, and etc). We can revisit selection of participants when we expand our study for future study.

The rule of thumb for the number of small usability testing participants is approximately 6. More is always better to get results that we can generalize with results from other studies (and highly publishable). However if the company is running on a small budget and close deadline, you can obtain meaningful results with 6 participants. My suggestion is to try to obtain participants who can represent the vast amount of travelers who use our web interface (i.e, business travelers and adults).

Instruments

I would like to suggest a combination of instruments for this study: Prototype build and card sorting.

Prototype:

We understand that prototyping stages have been passed for our initial web interface development. However, we can use this to our advantage and use our existing web interface as a prototype. Many industries have placed an emphasis on using prototype as one of the important steps for usability testing. Prototyping can save time and cost to develop web interfaces and can be easily tested with end users. We already have a fully functional website that we can use as our testing instrument. However, we should try to approach a method called Vertical Prototyping. Nielsen defines vertical prototyping as a procedure where developer can remove some features and ask end users to complete specific tasks.

Vertical prototyping does require developers to build and take out functionality from the existing websites. Furthermore, if we have the time and resources we can go beyond and provide feedback as our participants comment on its weak spot. We identify this procedure as interactive prototyping thus providing immediate feedback if they were able to complete the task accurately.

However, like any other company we have a budget and a deadline. One of the simple and time efficient tests I would suggest is a modification of the card sorting method. To explain further, we can use paper mock-up to mimic how content should be organized as well as the navigation scheme. This process is more of a functional prototype by having end users play as the computers. We can double the effort by utilizing common “talk aloud” method to see how they process the information and why they are structuring information in a certain way.

Procedure

To start off our testing, refer back to the December 17, 2012 entry of information loads and organization in the airlines web interfaces. Using modified card sorting method, list out submenus on a blank piece of paper. Provide end users with Post-Its and see how they organize the submenus. See the example and the screen shot below. You may have to write up brief scenarios to describe how to complete their tasks. During the process allow end users to speak up and ask them to explain their rational behind their organization of menus.