Response Scales in Risk Judgments: the Effects of Representation, Fineness and User Choice

William N. Dilla
University of Missouri-St. Louis

Dan N. Stone
University of Illinois at Urbana-Champaign

ABSTRACT:

Although accounting professionals use decision aids, relatively little is known about how various decision aid design characteristics affect judgments in accounting and auditing. For example, risk judgment decision aids can change the response scales that are used for expressing and communicating risk judgments. This article examines the effects of three characteristics of scales for expressing risk judgments: representation (whether scales are labeled with numbers or words), fineness (whether scales are discrete or continuous), and user choice (whether decision makers are assigned to or choose the scale). Participants made inherent risk assessments for a hypothetical audit client using one of the four combinations that resulted from manipulating two levels of scale representation and two levels of scale fineness. Participants were assigned to one of these four combinations or allowed to select from among them. The results indicate that: (1) scale representation does not substantively affect judgment quality, (2) compared with discrete scales, continuous scales increase both judgment quality and the effort required to make risk assessments, and (3) compared with assigning participants to scales, allowing participants to select their own scales increases judgment quality and decreases cognitive effort.

Keywords: Decision aids, Response scales, Audit judgment, Lens model, Decision making

Data Availability:The data are available from the first author.