The diversity of special user groups (i.e., older adults from 50 to 90 years or children from 6 to 14 years) is vast. Assessing their requirements is challenging, as it requires sensitivity in terms of choosing an appropriate approach to collect data. Furthermore, the illustration of the data for the subsequent design process can be difficult, if different partners are involved in a project. In order to overcome these difficulties, we explored a decision diagram for the creation of personas. It aims at identifying the most appropriate approach (i.e., qualitative and/or quantitative data collection), taking into account the characteristics of the special user groups among other aspects.

By applying user-centered design (UCD), user characteristics are gathered in user studies and/or from literature review in the analysis phase. These characteristics may include needs, skills, behaviors, motivations, frustrations, and demographic data. Within the recent HCI community, many different qualitative and quantitative approaches are available, such as observations, workshops, interviews, surveys, ethnographic studies, etc. Nevertheless, the involvement of special user groups, like elderly and children, is often challenging due to a lack of participants or restricted access, among other reasons. Generally, the data from as many users as possible should be collected in the analysis phase. This typically results in a long list of user characteristics (e.g., needs) that might be difficult to prioritize, easy to misinterpret, or hard to understand for development team members or project partners. For example, if the results are presented as statistic figures and a project partner has little statistical knowledge, it can be difficult to gain a common understanding about the users.

The creation of personas has been motivated by the aim to better handle these issues through extracting the most important information from the assessed data. Personas help to make explicit assumptions about the target users [Antle, 2008] and enable a focus on important characteristics across many users [Moser et al., 2011]. A persona is a type of user model or an archetypical representation of real and potential users, which illustrates the individuals’ characteristics (e.g., needs, skills, behaviors, motivations, frustrations, and goals when interacting with a technology) [Cooper et al., 2007; Goodwin, 2001]. As different types of personas can be distinguished by the approach used to create them (i.e. based on quantitative or qualitative data or a combination), we developed a decision diagram that supported us in finding an appropriate approach. For the identification of the most suitable data collection approach(es), several aspects should be considered in advance, i.e., the pre-knowledge, existing skills, the sample size, or available resources. Regarding special user groups it is important to consider the users’ characteristics, e.g., elderly’s impairments, younger children’s missing abilities to read or write. Equally important are the available sample size and the knowledge regarding the behavioral variables, which can be assessed beforehand (e.g., from literature). The skills of the researchers and project partners about qualitative and quantitative data collection and/or analysis methods are also crucial. Moreover, the available time and human resources should be kept in mind. In the diagram, connection lines should me made for available.

1) Pre-Knowledge: In a first step, it is important to find out whether there is access to existing primary data sources, e.g., data from previous user studies, market research, or secondary literature.
2) Skills: In the second step, it is necessary to identify whether the researchers (who will create the personas) are familiar with qualitative and/or quantitative approaches (do they have the skills to perform them or have the possibility to gain those skills?).
3) Sample size: In a third step, the number of users (i.e., elderly or children) available for the data collection needs to be defined roughly.
4) Resources: If few human and time resources are available, the following possibilities would be suitable: 1) a quantitative approach for a big sample size, 2) a qualitative approach for a small sample size, or 3) the combined approach with a small sample size (which still needs to be appropriate for the cluster analysis on basis of the behavioral variables). All approaches are possible, when there are many human resources available and enough time for the data collection and the analysis.

The final decision about the most suitable research approach to create personas for special user groups should be based on the number of connection lines that were assigned between the aspects and the different approaches.

Antle, A.N. 2008. Child-based personas: Need, ability and experience. Cognition, Technology & Work, 10(2), 155-166.
Cooper, A., Reimann, R., and Cronin, D. 2007. About face 3: the essentials of interaction design. Wiley-India.
Goodwin, K. 2001. Perfecting your personas. http://www.cooper.com/newsletters/2001_07/perfecting_your_personas.htm

Research by