SEM2

The first chapter goes through the DECIDE framework, which is useful when you should do an evaluation. This particular framework is appropriate to do during the iterations of the design process (check how it’s meeting the user requirements). DECIDE is kind of a checklist where the first step is to determine the goals. This is a high level question where a lot of IT-development today could use more of, not losing track of who will use this product and why.

In our case in the project, exploring the question could be very useful because we’re making the payment more convenient. Breaking down questions like “Why isn’t it more people paying subway tickets with their phone?”, into sub-questions like ‘is the interface not user friendly?’, etc.

Next is choosing the method of evaluation. In our project we will probably let people do real interactions with our prototype in a field study, however, since SL might not be kind enough to let us in their payment system and we might have to simulate it in a more virtual case study. That’s probably our biggest practical issue together with the very limited time people have waiting for the next subway train. The only ethical issue here will be keeping the personal data safe and correct according to PUL.

Because of our project’s practical issues, a heuristic evaluation might be useful. This inspection method can give us feedback on the prototyped user-interface by using the Nielsen’s 10 usability heuristics (10 very useful principles that improves the UX). This evaluation will have to be done from 3-5 times to get the probability of finding 75 % of the usability problems.

An alternative method to the heuristic evaluation is the walkthroughs, where you try to find the user-interface problems without doing empirical testing. Cognitive walkthrough is useful because it helps to evaluate how easy it is for new user to complete a task/solving a problem. In our project it could simply be: use the app to pay for a ticket in the subway. This can also be done as a pluralistic walkthrough where the walkthrough is done in a group both users and developers.

The prototype in our case study will have to include some predictive models like the GOMS Model or the Key Stroke Level Model. We most likely measure things like swipe strokes with the fingers and time it takes to do different tasks.

Question for the seminar: What is the most cost efficient way to do an evaluation?