<p><span>防策略机制中偏好申报的逆向效应研究 </span>(Backward Elicitation of Preferences in Strategy-proof Mechanisms)</p>

<jats:p>&lt;p&gt;Chinese Abstract:&amp;nbsp;本文考察了在防策略（Strategy-proof）匹配机制中，仅仅改变偏好申报（preference-elicitation）的顺序，是否足以影响参与者的行为与匹配结果。基于“学校选择”的实验室实验，研究对比了匹配规则（序列独裁机制&amp;nbsp;SD&amp;nbsp;与随机序列独裁机制&amp;nbsp;RSD）及偏好申报顺序的交互影响，即传统的“标准程序”与从最不偏好选项开始排序的“逆向程序”。研究发现，逆向申报对真实偏好汇报的影响具有显著的机制依赖性：其降低了&amp;nbsp;RSD&amp;nbsp;机制下的真实汇报率，却显著提升了&amp;nbsp;SD&amp;nbsp;机制下的汇报表现。决策过程分析表明，参与者在排序时会同时权衡收益差异与感知的竞争程度，但逆向申报并未显著改变参与者对收益或竞争的敏感度。&lt;/p&gt; &lt;p&gt;&lt;br&gt;&lt;/p&gt; &lt;p&gt;English Abstract:&amp;nbsp;We study whether a minimal change in preference-elicitation order alone can affect behavior&amp;nbsp;and outcomes in strategy-proof matching mechanisms. In a laboratory school choice setting, we vary the&amp;nbsp;matching rule (Serial Dictatorship vs. Random Serial Dictatorship) and the order in which participants rank&amp;nbsp;schools (standard procedure vs. backward procedure that elicits the least-preferred option first). Across&amp;nbsp;both mechanisms, backward elicitation yields mixed effects on truth-telling—it decreases truthfulness under Random Serial Dictatorship but increases it under Serial Dictatorship. Analysis of the decision-making&amp;nbsp;process shows that participants respond to both payoff differences and perceived school competitiveness.&amp;nbsp;However, backward elicitation fails to significantly alter participants’ sensitivity to either payoffs or competitiveness.&lt;/p&gt;</jats:p>