Exercise 6 - Individual Post - Edwin Li

And here we are, at the end of our journey, with a final product in hand in addition to enhancing our knowledge in the fundamentals of human computer interaction. Our mission was to improve the ticket purchase process in the Stockholm subway system and our target group was international tourists. The work we put down each week varied, but I estimate it to be around 5-10 hours a week, including planning for data gathering, data analysis as well as going through course material such as the course literature and the lecture slides. When we started we had no clear picture of what to improve, but our goal was becoming more apparent as we interviewed the staff who worked for SL.

Our qualitative data gathering consisted of conducting different kinds of interviews to different stakeholders, namely the SL personnel, tourists, and travel bureaus. While interviewing the tourists, we let the tourists answer open questions, as well as giving them the ability to answer the same kinds of questions by putting their ratings on a scale. This approach, that is, using more than one method or target in the study in order to check the result from a certain subject, is called triangulation. It was important to keep in mind not to formulate the questions such that they only could be answered in a certain way or giving information about how other people has answered, thus invalidating our research by introducing bias. For instance, the answers given by the participants could be invalidated if we suspect them to be influenced by how other people in their situation would react to a presumed difficult step in the ticket purchasing process. This is a well known phenomenon in heuristics known as informational social influence. This was specifically difficult since we had to create three templates of questions depending on the stakeholders. We also gathered quantitative data, such as age, nationality, and occupation, in order to create credible personas and scenarios.

We then split our data between the different members and started to analyse them. We considered conducting several interview processes in case our hypothesis weren't strong enough, thus doing a grounded theory approach. The negative effects of this would be that it'd take a long time to do, which we discovered while doing our analysis. This meant, we judged it wasn't feasible for the scope of our project.

After getting several hints on the general opinions from our target group, we did a state-of-the-art analysis of our product to perceive the general development of our proposal. Here, we examined the current Swedish ticket system and its corresponding physical signs (which were eliminated in our final version). We compared it with the German ticket system, and lastly we also analysed the SL app and website.

Now we had enough data to create personas and scenarios. Our first persona is a young Turkish woman, Busra Mirzaie, that traveled to Sweden to experience the youth culture. She understands English and has done some research on how to travel beforehand. We imagined she'd be perplexed on how to get an access card and its correlated prices. The second persona is a senior from Germany, Hans Zimmermann. He represents a large group of tourists that comes to visit Sweden for its cultural heritage and architectural wonders. We envisioned him to have problems understanding the Swedish system, especially ticket types, and also the fact that the German translation was unavailable. We created the following pain points for our personas, starting with Busra and ending with Hans:

Function	Primary Persona - Busra Mirzaei	Secondary Persona - Hans Zimmermann
Clear instructions on how to get access card	1	5
Choose between paper ticket and loading access card	5	1
Clear overview over the tickets and prices	1	1
Easy to understand the zone system	4	2
Comprehension of reduced and regular ticket types	3	2
Clear instructions on optional ways to buy ticket	2	3
Clear and alternative language options	5	1
Recommended ticket types	1	1

We then created a physical prototype out of cardboard and paper, solidifying our initial concept. Please see the following picture.

After this step we received feedback from evaluators made up from other groups. We then each did a think-aloud study to get even more feedback from potential users. We found this to be valuable, and its effects can be seen on our final products.

In general I think our group did a great job. We did quite alot of research and were happy with our results. Unfortunately we did not win the final voting (another group with a great idea received that honor, but we were 2nd!), but we learned alot and had lots of fun.