LOCATION

George Washington University School of Business
Duquès Hall, Room 652
2201 G Street NW (main entrance on 22nd Street between G and H Streets)
Metro: Foggy Bottom Station

Bio: Susan Krische

Professor Krische teaches financial accounting and financial statement analysis. Her research in judgment and decision making applies experimental and archival-empirical methods to investigate the use of financial information by investors and analysts. Susan’s research has been published in top academic journals including The Accounting Review, Contemporary Accounting Research, the Journal of Finance, and the Journal of Accounting Research. She has received several honors, including the American Accounting Association Competitive Manuscript Award. She currently serves as an Editor for Contemporary Accounting Research and is on the Editorial Boards of The Accounting Review and Behavioral Research in Accounting. Susan was previously on the faculty at the University of Illinois, and has served as an Academic Fellow with the Office of the Chief Accountant at the U.S. Securities and Exchange Commission. She earned her doctorate from Cornell University, has taught at the University of Waterloo (Canada), and has worked in public accounting with Ernst & Young (Clarkson Gordon).

---

Abstract

Behavioral research in financial accounting often aims to address issues relevant to individual, nonprofessional investors, citing the importance of individual investors to capital markets or regulators’ concerns for the “average” investor. Nevertheless, many such studies utilize convenience samples of graduate business student participants in their experiments. This research analyzes a relatively large sample of participants recruited from Amazon’s Mechanical Turk platform (n>2,000) in order to assess how the broader range of investors’ numerical skills can impact extant accounting research results. Because we know relatively little about the characteristics of the individual investor, this research begins by examining investors’ demographic characteristics and numerical skills relative to non-investors, benchmarked against national samples of financial capability skills in the United States. A set of three extant financial accounting research experiments are then replicated. Results show that investors with higher numerical skills are more sensitive to others’ incentives and are more likely to incorporate that understanding into their judgments than are investors with lower numerical skills or non-investors. These findings suggest that limiting the pool of participants in the original research increases the power of the original statistical tests, and emphasizes the need for careful consideration of the potential match between the applied theory and the population of interest in empirical research.

---