Categorised ethical guidelines for large scale mobile HCI

Abstract

The recent rise in large scale trials of mobile software using 'app stores' has moved current researcher practice beyond available ethical guidelines. By surveying this recent and growing body of literature, as well as established professional principles adopted in psychology, we propose a set of ethical guidelines for large scale HCI user trials. These guidelines come in two parts: a set of general principles and a framework into which individual app store-based trials can be assessed and ethical concerns exposed. We categorise existing literature using our scheme, and explain how researchers could use our framework to classify their future user trials to determine ethical responsibility, and the steps required to meet these obligations.

Implementing ethics for a mobile app deployment

“…The issues are discussed at length by McMillan et al (2013), who present a set of ethical guidelines for developers, focusing on how to protect and support autonomy and how to minimise risk.…”

“…Data collection is especially problematic when identifiable data such as location is collected from an app when the user might not realise they are in a study or assume that an app of the kind they are using would collect the data it does. McMillan et al (2013) also briefly outline the importance of responsibility, pointing out that the researcher needs to be aware of the relevant laws and best practices associated with data collection, and data should be collected and stored only if it has clear scientific value. However, there is little discussion or consideration of on going situational or in-action ethics in the literature on app store deployments.…”

“…Therefore, our approach had to be one of making it clear that this was not a children's app (through e.g. avoiding the use of cartoon characters and descriptions that would be seen as appealing to children -as recommended by McMillan et al 2013). …”

“…In particular, it is difficult to ensure that all users understand that they are in a research study and that their autonomy in deciding to take part and to freely withdraw is supported. According to guidelines created by McMillan et al (2013) for app store deployments, consent can and should be gained through in-app mechanisms. However, as McMillan et al explain, the inevitable uncertainties over whether users have read and understood the study information and are in a position to give consent need to be counter balanced with respect and care for privacy.…”

“…We will argue that while it is possible and desirable to implement a research app according to ethical guidelines and frameworks, the uncertainties require ongoing attention to ethical issues during development as well as post deployment. Therefore, beyond the guidelines for app store deployments provided by McMillan et al (2013), an ongoing "in-action" (Frauenberger et al 2016) or "situational" (Munteanu et al 2015) perspective is desirable. In turning attention to research ethics as an on-going concern, we explore how these become enmeshed with the broader activities and concerns of app design.…”