Impact mechanism of corporate social responsibility on performance from perspective of analyst attention:Research based on machine learning and text analysis
By integrating stakeholder theory and signaling theory,this research employs analyst attention as an intermediary variable and utilizes machine learning and text analysis techniques to investigate the mechanisms through which corporate social responsibility(CSR)affects corporate performance from the perspectives of disclosure and action.By analyzing data from 957 listed companies on the A-share market in China and comprising 5472 CSR data points,this research reveals that various aspects of CSR disclosure,such as tone,readability,and coverage,along with CSR action such as donation amounts,donation growth rates,and industry-specific donation proportions,exert a considerable influence on analyst attention.Analyst attention,in turn,exerts a positive direct impact on both short-term and long-term corporate performance and serves as a mediating variable that transmits the positive effects of CSR on corporate performance.This research optimizes the measurement of both CSR disclosure and action,minimizes potential information biases associated with reliance on third-party databases.Furthermore,it addresses the limitations in previous literature that often treated CSR disclosure and action as separate entities.By incorporating analyst attention as a key intermediary variable into the theoretical model,this research sheds light on the"black box"of how CSR disclosure and action impact short-term and long-term corporate performance.
NETL
NSTL
万方数据
