Seminarie 2

Evaluation methods

This seminar is about evaluation methods with emphasis on usability and usability evaluations. The reflection below is based on the literature that we were assigned to read. 
The first chapter introduces main approaches to evaluation and different methods typically used. Evaluations are done throughout the design, by collecting information about existing and potential users in order to improve the usability. There are three different types of evaluations: controlled setting involving users, natural settings involving users and any setting not involving users. The second method isn't relevant to our project due to lack of time and resources. The first method and third method are briefly discussed with their application in our project below.

Evaluating in a controlled setting involving users

This method is usually conducted in living labs or laboratories. The pros with this method is that in these environments the evaluator can reduce distractions and outside influences. This is usually used when testing technologies where the user has to perform a set of task. The primary goal is usually to determine whether an interface is usable by the intended target group to carry out the task of which the product was designed and to find usability issues. 

Evaluating in any setting not involving users

This evaluation relies heavily on the modelers common knowledge of the user group. The interactions are imagined and predicted by the likelihood of an event based on the model. A heuristic evaluation is when questions are answered through the knowledge of a typical user group point of view. The goal is usually to evaluate design for ease of learning. Another type of modelling user behavior is by analytic of data and by models, that can be used to compare interfaces and optimization.

Final thoughts

The first method has a big relevance for our project because of the fact that we do not have enough experience or data on our user group to model different kinds of behavioural models. The biggest challenge with this method is to find participants that are willing to spend time on our product and gives us feedback. My first question is how we will approach the challenge of finding participants if we choose the first method? 

The second question is how we will model our models in order to be as precise as possible?