161 General Discussion 7 Figure 1. An illustrative example of three participants' retrospective assessments of their (non-) stationarity processes. An example will illustrate the implications of (non)stationary processes on measurement. Figure 1 shows hypothetical examples in which participants answer a seemingly simple question in a survey: "How scared were you in the past two months?" (similar to an item they answered on the YSR; Chapter 3). Imagine they could answer on a five-point Likert scale with options "Not at all" (0), "A little bit" (1), "Moderately" (2), "Pretty much" (3) or "Very much" (4). Let's consider this question, which was asked of participants of the diary study (Chapter 4), who self-rated their fearfulness once per day for two months with the same Likert scale. The left panels of Figure 1 show the timeseries of daily fear of three example participants. Now that they are asked how scared they were in the past two months, each participant will do their best to summarize their own process adequately. The first participant's actual process (Top panel of Figure 1) started out with relatively many days that she felt pretty to moderately scared, but in the second half, she reports not feeling scared at all. Let's assume this participant conveys that over the past two months, she was "a little bit scared". This was true on average, but this average summarizes neither the first half, the second half, nor the full timeframe. She is trying to summarize a non-stationary process, but this is
RkJQdWJsaXNoZXIy MTk4NDMw