Abstract

Despite calls for an evidence-based focus on outcomes as a way to enhance accountability for public performance, findings from a prior study suggest that the public may be more impressed by high frequency (low cost) but ambiguous outputs (such as people served) rather than more meaningful but costly outcomes (causal effects). We attempt to replicate and extend the investigation of this output bias through a pair of survey experiments involving judgments about two evidence-based, highly effective social programs: one, an HIV/AIDS prevention program (adapted from the prior study), the other, a program for special needs high school students (Check and Connect). Our findings confirm that respondents viewed both programs more favorably when given information about mere outputs (people served) in comparison with more rigorous outcomes (causal effects). We then tested an extension of the Check and Connect experiment in which we modified the framing of cost and performance information in ways that reduced the tendency toward an output bias. We speculate on the possible mechanism that may lead to an output bias, and we discuss the implications of our findings for evidence-based public policy and management.

摘要

尽管呼吁以证据为基础关注结果，作为加强对公共绩效问责的一种方式，但先前研究的结果表明，公众可能对高频（低成本）但模棱两可的产出（例如服务的人）印象更深，而不是而不是更有意义但代价高昂的结果（因果关系）。我们试图通过一对调查实验来复制和扩展对这种输出偏差的调查，这些实验涉及对两个基于证据的高效社会项目的判断：一个是艾滋病毒/艾滋病预防项目（改编自先前的研究），另一个，针对特殊需要高中学生的计划（Check and Connect）。我们的研究结果证实，与更严格的结果（因果关系）相比，受访者在获得关于单纯产出（服务的人）的信息时更喜欢这两个项目。然后，我们测试了 Check and Connect 实验的扩展，在该实验中，我们以减少输出偏差趋势的方式修改了成本和性能信息的框架。我们推测可能导致输出偏差的可能机制，并讨论我们的发现对循证公共政策和管理的影响。