Inference from empirical research
Article Abstract:
Research techniques which lower test power have sometimes been accepted in accounting research without consideration of their implications. A Bayesian framework is presented which is useful in understanding and analyzing trade-offs that are a basic part of empirical research. Researchers should try to maximize power in designing and executing empirical tests, and try to keep the effective level of tests at stated points. The results of tests with low power are of little value, because they should induce little belief revision, regardless of outcome. Behaviors which increase published empirical test effective levels are not desirable, since the increase is obtained by a decrease in the belief revision justified by the results.
Publication Name: Accounting Review
Subject: Business, general
ISSN: 0001-4826
Year: 1987
User Contributions:
Comment about this article or add new information about this topic:
Sample error characteristics and projection of error to audit populations
Article Abstract:
Auditing techniques related to error identification within auditing samples and calculation of total errors for an audit population are discussed, and a model of inference is developed for application in these areas. Auditors' actual practices are determined by asking several practitioners how they would respond to hypothetical audit cases. These responses indicate that most auditors tend to isolate (and correct for) errors they perceive as unique to the situation, and that the form of error isolation can be related to firm membership factors and audit training and experience levels.
Publication Name: Accounting Review
Subject: Business, general
ISSN: 0001-4826
Year: 1986
User Contributions:
Comment about this article or add new information about this topic:
Incidence and circumstances of accounting errors
Article Abstract:
An examination of the incidence of accounting errors shown by prior period adjustments for 41 companies compared with a control group of 41 other firms is used to emphasize circumstances likely to motivate managers to use errors as a means to manage income. The results show that economic incentives motivate managers into overstatement errors and that earnings overstatements are more likely when a company has diffuse ownership and lower growth in earnings.
Publication Name: Accounting Review
Subject: Business, general
ISSN: 0001-4826
Year: 1991
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: Trends in financial management. Management accounting practices. FASB sets 1985 agenda
- Abstracts: Trainees' attributes and attitudes: neglected influences on training effectiveness. Conceptualizing how job involvement and organizational commitment affect turnover and absenteeism
- Abstracts: The comparative performance of MBAs vs. undergraduate accounting majors in public accounting. Factors contributing to published research by accounting faculties
- Abstracts: Ethics: ethical dilemmas in management accounting. How ethical is your company? On the importance of ethics
- Abstracts: Eliminate the Ph.D. requirement for accounting educators. Accounting education: does more mean better? How the new tax law affects business