摘要
The order in which the response alternative is presented may significantly alter the results of public opinion polls. However, little attention has been paid to the response order effect on survey about perceptions of nuclear power. Therefore, an Internet-based split-ballot experiment was conducted to gain insight into the response order effect on survey on young Chinese people's perceptions of nuclear power. The results show that the response order effect was not observed in most questions. The possible reason was discussed. In addition, the response order effect was not observed in yes or no questions. Therefore, this type of question may be adopted in future survey to avoid biased responses. The results suggest that the response order might affect the specific aspects of young Chinese people's negative perceptions, but it did not affect the impression of nuclear power. An obvious response order effect on the question to evaluate the respondents' enthusiasm for information on nuclear power suggested the need of seeking an alternative way for evaluation. Furthermore, the results suggested that most respondents had a more restricted definition of a severe accident. Both the response order effect and the framing effect were observed in the context of nuclear safety, and a possible explanation was discussed. Finally, the analyses suggested that the options might imply the same thing to the respondents, so the swayed opinions and the primacy effect occurred.