Systems are evaluated from the user perspective. Test scenarios can be based on results of the earlier phases when the target group, user contexts and requirements have been studied. Representatives of the target groups have to be selected to test the system with regard to functionality and user experience.
What (and whose) quality criteria do we use for evaluation? (reflection aspect relevance) Let us try to evaluate our system under extreme conditions (e.g., diverse users not from the primary target group).(reflection aspect concept of the human).
Epstein et al. (2017) studied menstrual cycle tracking apps as examples for personal informatics tools. They conducted surveys and interviews with people who have or have had menstrual cycles, which they decide to call “women” in their paper. They find reasons and current ways of tracking and describe problems and issues with tracking. As they aimed at diversity in experience and background, they included female, non-binary and male participants with various sexual orientations in their evaluation study. It turned out that designers of most apps obviously had started from a limited concept of their target group (female, heterosexual, fertility planning, pregnancy-avoiding) which was reflected by functionality and stereotypically “feminine” design (pink and flowery with heterosexual iconography). The evaluation showed that such heteronormative design of menstrual apps excluded people from using them and alienated others while using them.