The person trade‐off method and the transitivity principle: an example from preferences over age weighting
研究了人际权衡法在测量公众对不同年龄患者优先排序时的可靠性，发现多数回答不满足基数传递性，但排除不满足序数传递性的受访者后，平均而言传递性成立。
The person trade-off (PTO) is increasingly being used to elicit preferences in health. This paper explores the measurement properties of the PTO method in the context of a study about how members of the public prioritise between patients of different ages. In particular, it considers whether PTO responses satisfy the transitivity principle; that is, whether one PTO response can be inferred from two other PTO responses. The results suggest that very few responses to PTO questions satisfy cardinal transitivity condition. However, this study has produced results that suggest that cardinal transitivity will hold, on average, when respondents who fail to satisfy the ordinal transitivity condition have been excluded from the analysis. This suggests that future PTO studies should build in checks for ordinal transitivity.