Abstract: In order to prevent the occurrence of coal mine roof accidents, the coupling degree of causative factors in coal mine roof accidents is quantified. Collecting 120 coal mine roof accident investigation reports, using grounded theory to analyze the accident text, from the aspects of management factors, technical factors, environmental factors, physical factors, human factors, etc., we extracted 55 coal mine roof accident causative factors. The coupling effect of accident causative factors is defined, and the concept of coupling degree is introduced. Then, the coupling categories of coal mine roof accident factors are summarized. The Gephi software was used to draw the coupling network model of the cause-causing factors of coal mine roof accidents. Combined with the basic idea of Bayesian networks, a quantification method for the coupling degree of the cause-causing factors of coal mine roof accidents was proposed, including the two-factor coupling calculation model and the multi-factor coupling calculation model. Example application results demonstrate that the coupling degree of the four accident causative factors coupling pairs is higher than 0.4 from the two-factor coupling degree calculation results; according to the coupling degree calculation results of the three factors, the coupling coefficient of the six groups of three factors is between 0.7 and 0.82; coupling probability of accident causative factors is larger than the original value, and the coupling degree of accident causative factors increases, which is due to the adverse influence brought by the coupling effect of accident causative factors, indicating positive coupling. Among them, safety education and training are not in place, support quality is poor, risk analysis and judgment is not in place, operation problems, hidden trouble investigation is not in place, surrounding rock stress concentration, monitoring system failure and other factors have strong coupling. The more the number of coupling factors involved in the accident causes, the greater the coupling coefficient of the accident causes, and the more significant the coupling effect.