Abstract
As a researcher wants to design a questionnaire, he/she has to clarify the purposes of the survey to figure out the initial ideas about the questionnaire design. After that, the researcher has to search related information for questionnaire design based on their initial ideas, which is usually cost and time consuming. This research aims at developing a model for automatic questionnaire generation. Before developing the model, this study analyzes the statistic characteristics of components of initial ideas. On the basis of the analysis results, this research develops a methodology for automatic generation of questionnaires. By using the methodology, the initial ideas of a researcher can be converted into structured components and the genetic algorithm can be applied to extract questionnaire related data from the Internet based on the structured components. Useful materials from the related data can be extracted to generate a preliminary questionnaire. After that, the questions from the preliminary questionnaire can be sorted redesigned. By utilizing the model, efficiency and effectiveness for questionnaire design can be enhanced.