Abstract
A large body of the published research in financial accounting provides strong and persuasive evidence on the association between accounting numbers and stock prices but falls short of drawing any implications about accounting policies on measurements or on reporting. Attempting to go beyond association the author of this paper provides an application of a possible methodology for evaluating materiality of accounting measurements in a given decision situation. The methodology employs discriminant analysis in which the sensitivity of the discriminating model to the changes in mean earnings per share provides a decisions reaction scale and shows that a change in mean earnings per share equal to or less than 10% does not significantly affect the prediction. The validity of the results would depend on the correspondence between this model and the behavior of decision‐makers. However, in situations such as the one chosen here, in which an investor's decision‐making process cannot be normatively structured, appropriate statistical methods effectively can be utilized to describe the process.
Original language | English (US) |
---|---|
Pages (from-to) | 616-629 |
Number of pages | 14 |
Journal | Decision Sciences |
Volume | 8 |
Issue number | 3 |
DOIs | |
State | Published - Jul 1977 |
Externally published | Yes |
ASJC Scopus subject areas
- General Business, Management and Accounting
- Strategy and Management
- Information Systems and Management
- Management of Technology and Innovation