University of Münster.
Institute for Mind, Brain and Behavior, HMU Health and Medical University Potsdam, Germany.
Multivariate Behav Res. 2024 Jul-Aug;59(4):879-893. doi: 10.1080/00273171.2024.2335411. Epub 2024 Jul 11.
Mobile applications offer a wide range of opportunities for psychological data collection, such as increased ecological validity and greater acceptance by participants compared to traditional laboratory studies. However, app-based psychological data also pose data-analytic challenges because of the complexities introduced by missingness and interdependence of observations. Consequently, researchers must weigh the advantages and disadvantages of app-based data collection to decide on the scientific utility of their proposed app study. For instance, some studies might only be worthwhile if they provide adequate statistical power. However, the complexity of app data forestalls the use of simple analytic formulas to estimate properties such as power. In this paper, we demonstrate how Monte Carlo simulations can be used to investigate the impact of app usage behavior on the utility of app-based psychological data. We introduce a set of questions to guide simulation implementation and showcase how we answered them for the simulation in the context of the guessing game app (Rau et al., 2023). Finally, we give a brief overview of the simulation results and the conclusions we have drawn from them for real-world data generation. Our results can serve as an example of how to use a simulation approach for planning real-world app-based data collection.
移动应用程序为心理数据收集提供了广泛的机会，例如与传统实验室研究相比，增加了生态有效性和参与者的接受程度。然而，基于应用程序的心理数据也带来了数据分析方面的挑战，因为缺失和观察的相互依存性引入了复杂性。因此，研究人员必须权衡基于应用程序的数据收集的优缺点，以决定他们提出的应用程序研究的科学实用性。例如，如果某些研究能够提供足够的统计能力，那么它们可能才有价值。然而，应用程序数据的复杂性阻止了使用简单的分析公式来估计诸如功率等属性。在本文中，我们展示了如何使用蒙特卡罗模拟来研究应用程序使用行为对基于应用程序的心理数据的实用性的影响。我们引入了一组问题来指导模拟的实现，并展示了我们如何在猜测游戏应用程序的背景下（Rau 等人，2023）回答这些问题。最后，我们简要概述了模拟结果以及我们从模拟结果中得出的关于真实世界数据生成的结论。我们的结果可以作为如何使用模拟方法来规划基于实际应用程序的数据收集的示例。
