How Enterprise Data Quality Impact on Empirical Research Conclusions?——From 2015 Guangdong Manufacturing Matched Enterprise-Employee Survey-质量院英文网
Position: Home > Papers > Content

How Enterprise Data Quality Impact on Empirical Research Conclusions?——From 2015 Guangdong Manufacturing Matched Enterprise-Employee Survey

April 7, 2016

Cheng Hong1, Xu Wei1, Li Tang1,2

1. Institute of Quality Development Strategy,Wuhan University,Wuhan 430072,China;

  Macro-quality Management Collaborative Innovation Center in Hubei Province,Wuhan 430072,China;

2. School of Political Science and Public Administration,Wuhan University,Wuhan 430072, China


Abstract: In recent years, economics literature at home and abroad emphasis on using corporate data to do empirical research. However, the literature based on data from the listed company, Chinese industrial enterprises databases and customs trade databases is analyzed, the results show that the existing enterprise data has three defects: lack of timeliness, lack of randomness, and lack of diversity, which will have a greater impact on the accuracy, science and policy guidance of the research conclusions. In this paper, the author's Wuhan University led the Hong Kong University of Science and Technology, Tsinghua University and the Chinese Academy of Social Sciences made 2015 Guangdong manufacturing enterprise survey. Compared with the existing data, we have made a big improvement in the sample timeliness, randomness and diversity. This paper chooses three research fields: entrepreneurship and business performance, export enterprises “productivity paradox”, and credit constraints and business performance, to validate the impact of enterprise data quality on research findings. The results showed that: enterprise data influences the researchers’ judgments on Chinese economy because of lack of timeliness; enterprise data influences the researchers’ judgments on the overall situation of Chinese enterprises because of lack of randomness; the research findings appear systemic bias because of lack of diversity.


Keywords: data quality; timeliness; randomness; diversity