Academic journal article Quarterly Journal of Business and Economics

Replications in the Finance Literature: An Empirical Study

Academic journal article Quarterly Journal of Business and Economics

Replications in the Finance Literature: An Empirical Study

Article excerpt

Replications in the Finance Literature: An Empirical Study

Introduction

Replication of previously published findings preserves the integrity of a discipline's cumulative empirical foundation. Until a result is corroborated, preferably by an independent researcher, its status should be considered tentative. Failure to reproduce the result indicates the need for more detailed investigation of the topic. Replications with extensions help to increase the generalizability of research findings. It is for no small reason that the principle of replicability has been called the touchstone of the scientific method (Kane, 1984), an essential part of scientific methodology (Dewald, Thursby, and Anderson, 1986), the Supreme Court of the scientific system (Collins, 1985), and the most important criterion of genuine scientific knowledge (Rosenthal and Rosnow, 1984).

In view of the critical role that replications and their extensions play in knowledge development, one would expect their conduct to be commonplace among members of a given field. Unfortunately, it is difficult to know how much replication activity occurs within a discipline. How many replications and replications with extensions are published in that discipline's leading journals, however, can be estimated. Consequently, this paper addresses the following issues: How often are replications/extensions published in the finance literature? How does this publication incidence compare with other social and business sciences? Has the trend in the publication of replications/extensions increased or decreased over time? To what extent do replication studies confirm or disconfirm previously published results? How much discussion regarding the authenticity of published studies in the finance discipline?

The Publication Frequency of

Replications/Extensions in Finance

Replication and replication with extension are defined following Brown and Coney (1976) and Reid, Soley, and Wimmer (1981). Thus, a replication is a substantial duplication of a previously published empirical research project that chiefly is concerned with increasing the internal validity of the research design. An example would be repeating the experiment with another sample drawn from the same population. A replication with extension is a duplication of a previously published empirical research project that primarily is concerned with increasing the external validity (generalizability) of the research design. The intent is not to alter the conceptual relationships investigated in the original study but to test them differently by modifying either the manipulated (independent) or measured (dependent) variables, but not both.

The above definitions are analogous to Kane's (1984) econometric audit and improvisational replication. They are also generally consistent with the four types of replications distinguished by Mittelstaedt and Zorn (1984, p. 10). Briefly, their classification scheme is as follows: type I replication employs the same data sources and methods (models, proxy variables, statistical techniques) as the original study; type II replication uses the same data sources, but different methods; type III replication applies the same methods to different data; and type IV replication employs different data and different methods. The following examples, offered from a number of possibilities, illustrate the compatibility of their definitions and the current definitions. For instance, the current definition of replication can accommodate Mittelstaedt and Zorn's (1984) type I and III cases respectively, as when a researcher performs an econometric audit or repeats the experiment with a different sample drawn from the same population. Similarly, the present definition of replication with extension can include their type II and IV cases, respectively, as when a researcher analyzes the same data with modified definitions of the dependent variable or some of the independent variables or analyzes a different data set (population) with modifications in some variable definitions. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.