Jeroen van de Pol

142 Chapter 5.2 Data collection Sawtooth Software (Lighthouse Studio version 9.8.0) was used to create the CBC. In the CBC task in this study, participants had to respond to 10 random choice sets with two community pharmacies as options to choose. No opt-out option was made available (meaning participants were not able to choose none of the options presented). The options presented within the choice sets were full profile, showing all six attributes. All attributes had two levels that could be presented in the choice sets. A balanced overlap design (meaning attributes could present the same level within a choice set) was used to create the choice sets [18]. This also provides an efficient manner of gathering data. Ten questionnaire versions were generated that presented the participants the choice sets in a different order and showing different choice sets. Participants started with a description of two fictitious community pharmacies that recently started in the same town and offered different types of convenience- and CPS related services. Participants were then provided with background information on the services that were part of the choice task. During the choice task, participants could always return to this informational page and also retrieve some information regarding attributes and levels using a tooltip within the choice set. See figure 1 for a translated (Dutch into English) example of a single choice set. The questionnaire was tested by ten laymen and modified based on the feedback provided. Figure 1: Example of a choice set for participants showing the two choice options, attributes, different levels and the tooltip.

RkJQdWJsaXNoZXIy ODAyMDc0