Jeroen van de Pol

69 How community pharmacists prioritize cognitive pharmaceutical services 3 “neutral” or “not important”. Next, participants were asked to place the activities in the Q-grid (see Figure 1). Participants were obliged to adhere to the Q-grid. This forced them to carefully consider the position of every activity. Participants used the online software program FlashQ©, which takes participants through the Q-sorting process step by step in order to facilitate the Q-sorting process [23]. Figure 1: Q-grid used to fill in the 48 activities. Data analysis Factor analysis was used to identify correlation between participants with similar task prioritization. A group of participating pharmacists that correlate regarding task prioritization is called a factor [24]. A factor can be seen as a group of individuals that share a common understanding regarding the prioritization of the activities. However, for sake of readability, a factor will be called a group. Factor analysis can also render a subset of individuals that do not belong to any group. The number of factors/groups found is based on the amount of variance they explain. Analysis of the Q-sorts was performed using PQmethod 2.35 software [25]. Principal component factor analysis (PCFA) and varimax rotation were used to obtain the least amount of groups that explain the most of the variance. This approach renders factors/groups that can be statistically explained instead of allowing the researcher to influence the dataset to obtain certain groups [19]. Idealized Q-sorts were constructed for each group. These Q-sorts give insight in how a typical participant within this group would rank the 48 activities. Activities were considered important when they were ranked from +1 to +4, not important when ranked -1 to -4 and neutral when ranked as 0. Descriptive statistics were used to define each group based on their task prioritization and background characteristics of the participants.