Pretesting Video Speeches for an Experiment on the Effects of Two Dimensions of Servant Leadership on Performance and Examining the Strength of Potential Instruments

Causality is essential in informing science and policy. In the present study, we investigate the current state of research regarding causality in the field of servant (and authentic) leadership and provide recommendations on how causally identified studies can be conducted. After explaining the methodological problems that potentially prevent causal inferences (i.e., endogeneity bias and issues in experimental design), we provide two systematic literature reviews of servant and authentic leadership showing that these problems remain very prevalent. We then discuss two solutions on how causal effects of servant leadership or perceptions thereof can be established: randomized experiments and instrumental variable regression. To illustrate our recommendations, we report an experiment on the effect of a combination of two servant leadership dimensions (i.e., stewardship and authenticity) on follower performance and also investigate the effect of combined stewardship and authenticity perceptions using instrumental variable regression. The results do not indicate that combined stewardship and authenticity behavior or perceptions affect follower performance. Our study can serve as a roadmap, especially for servant leadership researchers, to address potential endogeneity and conduct causally identified research.

Causality is essential in informing science and policy. In the present study, we investigate the current state of research regarding causality in the field of servant (and authentic) leadership and provide recommendations on how causally identified studies can be conducted. After explaining the methodological problems that potentially prevent causal inferences (i.e., endogeneity bias and issues in experimental design), we provide two systematic literature reviews of servant and authentic leadership showing that these problems remain very prevalent. We then discuss two solutions on how causal effects of servant leadership or perceptions thereof can be established: randomized experiments and instrumental variable regression. To illustrate our recommendations, we report an experiment on the effect of a combination of two servant leadership dimensions (i.e., stewardship and authenticity) on follower performance and also investigate the effect of combined stewardship and authenticity perceptions using instrumental variable regression. The results do not indicate that combined stewardship and authenticity behavior or perceptions affect follower performance. Our study can serve as a roadmap, especially for servant leadership researchers, to address potential endogeneity and conduct causally identified research.

Causality is essential in informing science and policy. In the present study, we investigate the current state of research regarding causality in the field of servant (and authentic) leadership and provide recommendations on how causally identified studies can be conducted. After explaining the methodological problems that potentially prevent causal inferences (i.e., endogeneity bias and issues in experimental design), we provide two systematic literature reviews of servant and authentic leadership showing that these problems remain very prevalent. We then discuss two solutions on how causal effects of servant leadership or perceptions thereof can be established: randomized experiments and instrumental variable regression. To illustrate our recommendations, we report an experiment on the effect of a combination of two servant leadership dimensions (i.e., stewardship and authenticity) on follower performance and also investigate the effect of combined stewardship and authenticity perceptions using instrumental variable regression. The results do not indicate that combined stewardship and authenticity behavior or perceptions affect follower performance. Our study can serve as a roadmap, especially for servant leadership researchers, to address potential endogeneity and conduct causally identified research.

Causality is essential in informing science and policy. In the present study, we investigate the current state of research regarding causality in the field of servant (and authentic) leadership and provide recommendations on how causally identified studies can be conducted. After explaining the methodological problems that potentially prevent causal inferences (i.e., endogeneity bias and issues in experimental design), we provide two systematic literature reviews of servant and authentic leadership showing that these problems remain very prevalent. We then discuss two solutions on how causal effects of servant leadership or perceptions thereof can be established: randomized experiments and instrumental variable regression. To illustrate our recommendations, we report an experiment on the effect of a combination of two servant leadership dimensions (i.e., stewardship and authenticity) on follower performance and also investigate the effect of combined stewardship and authenticity perceptions using instrumental variable regression. The results do not indicate that combined stewardship and authenticity behavior or perceptions affect follower performance. Our study can serve as a roadmap, especially for servant leadership researchers, to address potential endogeneity and conduct causally identified research.

The concept of servant leadership has received growing interest in recent years. To date, however, there are no clean causal investigations of the effects of servant leadership on followers’ performance because of endogeneity issues. Moreover, the concept has mostly been studied via questionnaires, which can lead to disputable policy implications. We are the first to use an instrumental variable approach in the context of servant leadership and created new material based on the literature. This pre-study thus serves to test how followers perceive the leader manipulation and to examine the strength of our instrumental variables (i.e., manipulation of leadership style and leader gender and agreeableness). We aim to recruit at least 200 working-age individuals using a common panel provider. Our hypotheses will be tested by means of an experimental online study with a 2 (servant leadership vs. neutral speech) x 2 (male vs. female leader) factorial design.

The concept of servant leadership has received growing interest in recent years. However, in a systematic review, we found no clean causal investigations of its direct and mediating effects. We will thus explain the methodological problems in current research and present recommendations on how the causal effects of servant leadership can be investigated. As an illustration of our recommendations, we will examine the effect of two dimensions of servant leadership (i.e., stewardship and authenticity) on followers’ cognitive performance with an online experiment. Additionally, we will investigate the effect of authenticity and stewardship perceptions using an instrumental variable estimation approach. Thereby, we can determine the potential bias due to endogeneity. We aim to recruit British working-age individuals using a common panel provider. The sample size will be determined in a pretest. Our hypotheses will be tested by means of an experimental online study with two groups (authenticity and stewardship vs. neutral speech).