The reliability of data is an important section of guaranteeing the validity of research. To overcome the deficiencies of interview and questionnaire survey, experts developed a valid method of collecting data, which is Think Aloud Protocols (TAPs). By comparing TAPs with interview and questionnaire survey, it is found that subjects’ attribution bias in interview, social desirability in questionnaire survey and introspection about the task and memory deviation reduce the validity of data; while TAPs can verbalize subjects’ thought when dealing with representative tasks, and produce more valid data. In fact, TAPs can make study results more accurate, and generate new findings. It is an advisable and valid method to collect data for management research.