A note on Type S/M errors in hypothesis testing.

Motivated by the recent replication and reproducibility crisis, Gelman and Carlin (2014, Perspect. Psychol. Sci., 9, 641) advocated focusing on controlling for Type S/M errors, instead of the classic Type I/II errors, when conducting hypothesis testing. In this paper, we aim to fill several theoretical gaps in the methodology proposed by Gelman and Carlin (2014, Perspect. Psychol. Sci., 9, 641). In particular, we derive the closed-form expression for the expected Type M error, and study the mathematical properties of the probability of Type S error as well as the expected Type M error, such as monotonicity. We demonstrate the advantages of our results through numerical and empirical examples.

Engaging in Weight Loss Tasks of Mobile Health Applications: The Dual Influence of Social Support and Body Condition.

Introduction: In the past decades, the average weight of residents has increased significantly. Weight management has become an important issue in our society. With the developments of mobile technology and applications, mobile health applications (m-health apps) provide a convenient platform for users to engage in weight loss tasks and control their body weight. However, due to the lack of proper motivators, engaging in weight loss tasks on m-health apps is stressful for the users. The aim of this study is to understand why users engage in weight loss tasks, and to be specific, we establish an empirical model to examine the effects of social motivators (social support) and personal motivators (body condition) and their interactive relationship on the level of user engagement using self-determination theory. Materials and Methods: We developed a JAVA software program and automatically downloaded 1,138 users' information from an m-health app. Following, we used these data to calculate variables of our research model, including body mass index (BMI), informational support, emotional support, and the level of engagement in weight loss tasks. Additionally, we used the Ordinary Least Squares to estimate our research model. We also checked the robustness of the results by dividing the data into different groups. Results: The empirical results of our research model indicate that informational and emotional supports are positively associated with the engagement levels of users in weight loss tasks. Additionally, we have found out that body condition (using BMI as a proxy) has a U-shaped relationship effect on users' engagement. Furthermore, our research proves that body condition and informational support have a substitutive relationship in affecting user engagement. Conclusions : These findings can contribute to the literature concerning online weight loss and to provide suggestions for users and practitioners of m-health apps, catering different incentive mechanisms to users with different body condition to help them control their body weight.