User evaluations of a behaviour change support system

In this thesis three versions of the SmarcoS-system are evaluated. SmarcoS-diabetic focuses on recently diagnosed diabetes type II patients. It gives feedback on medication intake and activity. Feedback is given in the form of text and can be received on a smartphone application and at a computer application. Secondly, two versions of the SmarcoS-office worker system are evaluated. These versions focus on office workers with an intense digital lifestyle. Both versions of the system only give feedback on activity. Feedback can only be received at a smartphone application, and is given in the form of text or by an Embodied Conversational Agent. The implementation of the SmarcoS-diabetic system lead to participants manipulat- ing the system to be registered as being on time to take their medication. They took their pills out of the pill dispenser before they took them to open the pill dispenser within the set time. Subsequently some participants forgot to take that medication. Introducing this kind of errors should be avoided. In both evaluation we found that participants would like to omit the need to dock the activity monitor of the system. Now data is only known by the system after docking, while participants would like the system to have real-time data access. This also has implications for the feedback that can be given. Now feedback can only be given after the activity monitor was docked, making feedback most of the time too late. Participants thought the content of the messages was standard, administrative, not motivating, not diverse enough and there were not enough different types of messages. Content can be improved by making it more specific, concrete, personal and preference based. These results are found in both evaluation, although more prominent in the SmarcoS-office worker evaluation. We found a difference in the review of, and attitude towards the system between the two evaluations. This difference is best explained by the difference in target groups. The main difference between participants is that participants of the SmarcoS-diabetic evaluation are diabetics, while the participants of the SmarcoS-office worker evaluation are office workers. The main difference between these two groups is that diabetics have a disease and therefore can be considered patients. The office workers had no disease, and therefore could not be considered patients. This difference also makes it likely that the participants of the SmarcoS-diabetic evaluation are more conscious that they need to have a healthy lifestyle and take their medication on time. Therefore, they probably have a bigger interest or at least are more conscious of their interest in such behaviour change support system. It is therefore likely that they are more willing to give up privacy, because they get more advantages of giving up privacy compared to the non-patient office workers.