试验类型对试点选择模式的异质性影响——基于两类科技创新政策试验的案例分析  

Heterogeneous Effects of Experiment Typology on Pilot Selection Patterns:A Comparative Policy Analysis of Two Science and Technology Innovation Trails

作　　者：蒋林秀 李泉 Linxiu Jiang;Quan Li

机构地区：[1]武汉大学政治与公共管理学院

出　　处：《公共行政评论》2025年第2期1-19,196,共20页Journal of Public Administration

摘　　要：选择合适的区域进行试点是政策试验的关键环节。既有研究尚未形成试点选择的共识性结论,反而出现正向选择抽样、代表性抽样以及多重选择逻辑三种具有争议性的观点。论文尝试从试验类型的角度对这一争议进行弥合,强调围绕试点选择的理论建构应将特定政策情境作为重要变量纳入考量。论文基于条块互动模式区分出两种试验类型:以“条条”为主的中央主导模式和以“块块”为主的地方主导模式。结合科技创新领域的试验案例,研究通过模糊集定性比较分析(fsQCA)和聚类分析,揭示出试验类型对试点选择模式的异质性影响。研究发现,在科技创新政策试验中,试点选择的正向抽样并不是单一要素的作用,而是多要素组合的结果。同时,研究还发现,在不同的试验类型中,试点过程呈现出差异化的选择模式:在地方主导的政策试验中,由于中央对试验成功的追求以及地方议价能力的差异,试点选择以多因素组合的正向选择模式为主;在中央主导的政策试验中,试点选择不仅表现为多因素组合的正向选择,还兼顾了其他治理目标。这些发现不仅揭示了科技创新中的试验特征,有助于提升科技创新政策体系的科学性与合理性,还丰富了试点选择的理论解释,深入探索了政策情境变量对试点选择模式的差异化影响。The selection of pilot units is a pivotal step in policy experiments.While existing research has not yet reached a consensus on pilot selection patterns,three distinct perspectives have emerged:positive selection pattern,representative sampling,and multiple selection logics.This paper endeavors to bridge this inconsistency by integrating experiment types into the analysis of pilot selection.Based on the interaction between different levels of governments,we distinguish two types of experiments:the central-led model,which emphasizes vertical administrative departments,and the local-led model,which prioritizes horizontal administrative divisions.Through the examination of technological innovation experiments and the application of fsQCA and cluster analysis,we uncovered the diverse impacts of experiment types on pilot selection patterns.Specifically,our findings reveal that positive selection sampling in this domain is not influenced by a single factor alone,but rather is a result of the intricate interplay among multiple factors.Furthermore,our research identifies distinct characteristics in different types of experiments.In local-led policy experiments,the selection of pilot sites primarily follows a multi-factor positive selection pattern,driven by the centralized pursuit of success and variations in local bargaining power.In contrast,pilot selection in central-led policy experiments also adopts a multi-factor positive selection approach,but additionally takes into account other governance goals.These findings not only offer insights into the experimental nature of scientific and technological innovation,but also enrich the theoretical framework of pilot selection by examining the differing impacts of policy context.

关 键 词：科技创新试验 试验类型 试点模式 差异化选择 

分 类 号：D63[政治法律—政治学]