Open-ended questions are assumed to be able to elicit answers that represent the immediate concerns of the survey respondents, while, in surveys on community response to transportation noise, those have been rarely used compared to questions of a selection from multiple choices. In this study, first, the features of open-ended questions are discussed on the basis of a cognitive theory of emotion and next, answers to the open-ended questions, which were included in the questionnaires used in our past social surveys on transportation noise, were analyzed. In total, 739 open-ended answers from four surveys on road traffic, conventional railway, high-speed railway (Shinkansen), and aircraft noise were analyzed. In the analysis, sentences which contained positive or negative evaluation phrases were identified from each of the answers. The subjects of these sentences were then categorized and counted. The major results indicated that generally, there were more negative descriptions on road traffic noise than on other transportation noise and that, in the descriptions of Shinkansen, there were more complaints made against vibration than against noise.