Analogical Reasoning and Hypothesis Generation in Auditing.
通过实验研究审计师在初步分析性复核中如何利用类比和经验生成假设，发现经验丰富的审计师在特定条件下更依赖领域知识而非类比。
Abstract ABSTRACT: The role of analogy and experience in hypothesis generation in preliminary analytical review is examined in an experimental setting. Subjects were 73 senior auditors and 51 auditing students. The transfer paradigm [Gick and Holyoak, 1980], in which subjects are first given a problem and its solution (source analog) and then given a problem to solve (target problem), was used in designing the experiment. The accounting error detected in the source analog was the only difference between the two treatment conditions (errors due to timing and performance). A third condition did not include a source analog and acted as a control. In the comparison between the control and timing conditions, expertise was significant (p<.01). In the comparison between the control and performance conditions, treatment and expertise were significant (p<.05). In the timing condition the more experienced subjects used a strategy based on domain specific knowledge rather than analogy. In the performance condition both experienced and inexperienced subjects used analogy. This difference in results between the timing and performance conditions can be attributed to the difference in expected frequency of the timing and performance hypotheses.