5 112 CHAPTER 5 each question. Responses to questions focusing on opinions tended to result in more elaborate quotes for which a thematic approach of analyses was deployed. Coding was descriptive in nature with the goal to reflect as accurately as possible the ideas and opinions of the participants (see Boeije & Bleijenbergh, 2019; Braun & Clarke, 2022). For each research question, a number of themes emerged. After preliminary coding of the quotes, quotes and codes were revisited (by YHAB) and relabeled when necessary. Figure 5.1: Feedback report of participant 3. A; bar graphs indicating the mean score and 95% confidence interval of each risk-relevant behavioral, psychological, and contextual feature. B; timeseries plots indicating the time (on the x-axis) trajectories of risk-relevant behavioral, psychological, and contextual features (on the y-axis). C; contemporaneous network indicating the strength (the more saturated and wider the edge the stronger the association) of the positive (blue) or negative (red) associations between risk-relevant behavioral, psychological, and contextual features. The temporal network is not represented since this technique shows no connections between the riskrelevant features. The risk-relevant behavioral, psychological, and contextual features are defined as follows: 1 = I feel happy; 2 = I feel angry; 3 = I feel sad; 4 = I feel lonely; 5 = I feel insecure; 6 = At the moment, I think peers and adults are rejecting me; 7 = At the moment, I need contact with children/boys; 8 = At the moment, I need sex; 9 = How large was your maximum need for sex since the last beep?; 10 = How many times have you masturbated since the last beep? 5.2.2.3.2 Online evaluation questionnaire An online questionnaire on the perceived effects and burden of answering questions about risk-relevant features using a smartphone was created for the purpose of this study (Table 2). The items were scored on a continuous scale from “not at all” (0) to
RkJQdWJsaXNoZXIy MjY0ODMw