Density logging is an effective method in the evaluation of coalbed methane (CBM) reservoirs. Whether density log values effectively represent true density of coal will directly determine precision of evaluating coalbed methane parameter logs. In this paper, a statistical method is used to analyze density log response characteristics and hole enlargement rate for three main coal beds in the study area in order to determine the internal relation between hole enlargement rate and density response. Analytical results show that distortion of density log response values is caused mainly by hole enlargement impacts. Based on the apparent geometric factor theory, a model suitable for correcting hole enlargement impacts density logs for coalbed methane reservoirs has been deduced. To correct density logs for hole enlargement influences, it is key to determine a mud apparent geometric factor. Using measured apparent relative density, density log values and caliper logs for coal rock, the least squares fitting method was adopted to obtain computation model constants for the mud apparent geometric factor applicable to the study area. When this model was applied in a computer autocorrecting process for evaluating enlargement impacts on density logs for coalbed methane reservoirs of the Hancheng gas field in the eastern section of Ordos Basin, China, the correction results were very close to measured relative density, indicating that this method can improve precision of correcting hole enlargement impacts on density logs.