A Bayesian approach to nested data analysis: A primer for strategic management research
为战略管理学者提供贝叶斯方法分析嵌套数据的入门指南，通过CEO薪酬示例展示贝叶斯模型与传统方法的差异，帮助研究者更准确地解释概率分布。
Bayesian analysis offers strategy scholars numerous benefits. In addition to aligning empirical and theoretical endeavors by incorporating prior knowledge, the Bayesian approach allows researchers to estimate and visualize relationships that reflect the probability distributions many strategy researchers mistakenly interpret from conventional techniques. Yet, strategy scholars have proven hesitant to adopt Bayesian methods. We suggest that this is because there is no accessible template for employing the technique with the types of data strategy researchers tend to encounter. The central objective of our research is to synthesize disparate contributions from the Bayesian literature that are relevant for strategy scholarship, especially for nested data. We provide an intuitive overview of Bayesian thinking and illustrate how scholars can employ Bayesian techniques to analyze nested data using an example dataset involving CEO compensation. Our results show how using Bayesian models may lead to substantively different interpretations and conclusions compared to traditional approaches based on frequentist techniques.