Date of Completion

2-23-2015

Embargo Period

8-22-2015

Keywords

media format, disclosure tone, earnings condition, misstatement detection

Major Advisor

Stanley F. Biggs

Associate Advisor

Andrew J. Rosman

Associate Advisor

William T. Ross, Jr.

Associate Advisor

James P. Sinclair

Field of Study

Business Administration

Degree

Doctor of Philosophy

Open Access

Open Access

Abstract

I examine the effect of media format, disclosure tone, and earnings condition on financial statement users’ accuracy and criterion in detecting misstated accounts in a quarterly earnings release. I employ Signal Detection Theory, which allows the contemporaneous measurement of a participant’s accuracy and criterion in assessing an account as misstated or not misstated. I find that participants in the earnings-increasing condition were more accurate in detecting misstated accounts than participants in the earnings-decreasing condition. I did not find differences in accuracy of detecting misstated accounts between media-format conditions (video versus text) or disclosure-tone conditions (aggressive versus conservative).

In addition, I find that participants in the video media-format condition were less confident in their decision-making versus participants in the text media-format condition. On average, participants who viewed the video earnings-release rated their confidence in the extreme intervals less frequently than participants in the text earnings-release condition. I did not find differences in participants’ criterion for reporting an account misstated between disclosure-tone conditions or earnings conditions. My results suggest that the effect of a misstatement on earnings affects participants’ accuracy, but not their criterion, while the presentation of the earnings release affects participants’ criterion, but not accuracy in detecting misstated accounts.

COinS