首页> 外文学位 >Using an expert system to debias auditor judgment: An experimental study.
【24h】

Using an expert system to debias auditor judgment: An experimental study.

机译:使用专家系统消除审计师判断的偏见:一项实验研究。

获取原文
获取原文并翻译 | 示例

摘要

In the audit profession, judgment and decision making are essential parts of successfully completing an audit. Judgment is a crucial element of the audit process, especially with recent increase in audit regulations. At every step of the audit, auditors are required to make several complex judgments. Psychology and audit research demonstrate that auditors are susceptible to numerous biases, two of which are exacerbated by the accountability inherent in the audit environment. Since the nature of the audit environment requires auditor accountability, this dissertation examines two biases that have been shown to be exacerbated by accountability – the dilution effect and acceptability heuristic bias.;This dissertation consists of three separate essays. Two of this dissertation’s essays experimentally examine whether use of an expert system as a decision aid in developing less experienced auditor judgment mitigates these two biases - the dilution effect (Chapter 2) and the acceptability heuristic (Chapter 3) - on auditor judgment in a complex task (fraud risk assessment). An expert system was chosen as the decision aid because research has demonstrated that expert systems have a high level of accuracy, resulting in more appropriate judgments made by less experienced auditors. The third essay is an exploratory study using process tracing that analyzes the decision making process of less experienced auditors using the expert system in an environment without and with judgment bias present - the dilution effect.;The first essay experimentally examines whether an expert system assists less experienced auditors in making lower fraud risk assessments and mitigates the dilution effect. The dilution effect is a judgment bias which occurs when too much focus is spent on irrelevant information. This bias is exacerbated in auditors when they are knowingly held accountable to their superiors. A solution has yet to be offered to materially reduce this bias. The second essay experimentally examines whether the use of an expert system as a decision aid mitigates acceptability heuristic. The acceptability heuristic is the shifting towards the preferences of another and adopting a position that is deemed socially acceptable. This bias predominantly occurs when auditors are knowingly held accountable to their superiors. Prior studies have shown less experienced auditors who are aware of the views of audit partners, will align their judgments to agree with that of the partners’. This negative auditor judgment bias has been an area researchers have vastly studied, yet a method to reduce it has not been offered in the literature.;The third essay is an exploratory analysis which uses process tracing to analyze the thought processes of less experienced auditors in making decisions in an environment where a judgment bias occurs - the dilution effect. This essay used a subset of participants from the dilution effect study (Chapter 2) and during the experiment, participants were asked to think aloud to gain further insight into the dilution effect. The think aloud process also captures factors about the process of using the expert system which contributed towards the mitigation of this bias. Auditor decision making process with and without the use of an expert system will be captured both within and between the participant groups.
机译:在审计专业中,判断和决策是成功完成审计的重要部分。判断是审计过程的关键要素,尤其是随着最近审计法规的增加。在审核的每个步骤中,要求审核员做出几个复杂的判断。心理学和审计研究表明,审计师容易受到多种偏见的影响,其中两个因审计环境中固有的责任感而加剧。由于审计环境的本质要求审计师承担责任,因此本文研究了两个被问责制加剧的偏见-稀释效应和可接受性启发式偏见。;本论文由三篇独立的论文组成。本论文中的两篇文章通过实验检验了在开发经验不足的审计师判断中使用专家系统作为决策辅助工具是否可以缓解这两种偏见-稀释效应(第2章)和可接受性启发法(第3章)-对复杂审计师判断的影响任务(欺诈风险评估)。之所以选择一个专家系统作为决策辅助,是因为研究表明该专家系统具有很高的准确性,可以使经验不足的审计师做出更适当的判断。第三篇文章是使用过程跟踪的探索性研究,它分析了在没有或存在判断偏差的环境中使用专家系统的经验不足的审计师的决策过程-稀释效应。经验丰富的审计师进行较低的欺诈风险评估并减轻摊薄效应。稀释效应是一种判断偏差,当过多的注意力集中在无关的信息上时,就会产生这种偏差。当审计师被明知对其上级负责时,这种偏见会加剧。还需要提供一种解决方案以从实质上减少这种偏差。第二篇文章通过实验检验了将专家系统用作决策辅助工具是否会减轻可接受性启发式问题。可接受性启发法是朝着另一个人的偏爱转变,并采取一种被社会接受的立场。这种偏差主要发生在有意识地要求审计师对其上级负责时。先前的研究表明,经验不足的审计师会了解审计合作伙伴的观点,会根据他们的判断与合作伙伴的观点保持一致。这种负面的审计师判断偏差一直是研究人员广泛研究的领域,但文献中尚未提供减少这种偏差的方法。第三篇文章是探索性分析,它使用过程跟踪来分析经验不足的审计师的思想过程。在存在判断偏差的环境中做出决策-稀释效应。本文使用了稀释效应研究(第2章)的一部分参与者,在实验过程中,要求参与者大声思考以进一步了解稀释效应。大声思考过程还捕获了有关使用专家系统的过程的因素,这些因素有助于减轻这种偏见。在参与者组内和参与者组之间,将捕获使用和不使用专家系统的审核员决策过程。

著录项

  • 作者

    Lombardi, Danielle R.;

  • 作者单位

    Rutgers The State University of New Jersey - Newark.;

  • 授予单位 Rutgers The State University of New Jersey - Newark.;
  • 学科 Business Administration Accounting.;Business Administration Management.;Artificial Intelligence.
  • 学位 Ph.D.
  • 年度 2012
  • 页码 247 p.
  • 总页数 247
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号