Research On The Recognition Model Of Company Operation Performance Attribution Site

Attribution information of operating performance that listed companies reveal to the outside is an additional remark to digital information of accounting earnings, and is also an important way for investors to understand the management state. A large number of studies have shown that there is a self-interest tendency when company management disclosures the attribution information, and the self-interest tendency has an impact on investor’s decision. It is a very worthy of study that how to use existing information to identify the site of reason for change in company operation performance. The thesis constructs the recognition model of company operation performance attribution site based on the Kelly three-dimensional attribution theory in the field of psychology. Research of the thesis extends the application of attribution theory of psychology in financial reporting situations,and also helps external information users and supervision institution to judge the reliability of the attribution information, to promote listed companies to improve the quality of the attribution information disclosure.Method applied in the thesis combines theoretical research and empirical. In theoretical part, the thesis mainly expounds the necessity and feasibility of introducing the Kelly three-dimensional attribution theory to identify attribution site, and constructs the recognition model of company operation performance attribution site based on three criteria(distinctiveness, consistency, consensus) of Kelly three-dimensional attribution theory and profitability index(earnings per share, social contribution) selected. In empirical part, the thesis selects two industries ranking the top 100 companies of the earnings per share from industries that there is a big rate of increase amplitude of industry average performance and two industries ranking the after 100 companies of the earnings per share from industries that there is a big rate of reduction amplitude of industry average performance, then the thesis respectively uses the identification model to identify operation performance attribution site in good news and bad news. Research shows that 18 companies with high achievement should only be attributed to external factors and 6companies with high achievement only should be attributed to internal factors in 100 companies of good news, 22 companies with poor achievement should only be attributed to internal factors and 5 companies with poor achievement only should be attributed toexternal factors in 100 companies of bad news. For testing the validity of the model, the thesis uses the accounting data information of the income statement and the notes to analysis site of reason for change in operation performance that 18 companies with high achievement should only be attributed to external factors and 22 companies with poor achievement should only be attributed to internal factors. The result shows that 17 companies with high achievement are influenced by external factors in 18 companies with high achievement should only be attributed to external factors and 20 companies with poor achievement are affected by internal factors in 22 companies with poor achievement should only be attributed to internal factors. Accuracy of recognition model the thesis constructs reaches 92.5%, which shows the model is effective. Finally, the thesis puts forward related suggestions using recognition model to promote the quality of attribution information disclosure of listed companies.