Predicting user preferences for educational content based on the results of an automated evaluation for alcohol use...

Computer programs may be more effective than traditional self-help manuals because they are interactive, and can automatically modify their content to meet the specific needs of a user. Successfully predicting the type of information that will be most relevant to an individual may lead to greater engagement with the program and better outcomes. The current study tests the association of information obtained during an automated evaluation for alcohol use problems with subsequent preference for educational modules.