The Fertile Field of Meta-analysis: Cumulative Progress in Agricultural Forecasting

Thumbnail Image
Penn collection
Marketing Papers
Degree type
Grant number
Copyright date
Related resources

A substantial effort has been devoted to agricultural forecasting over the past half century. Allen's quantitative review provides a powerful way to examine that research. The quantitative review (or "meta-analysis" as it is commonly called since. Glass (1976) is a formal study of studies. Meta-analyses sometimes reveal conclusions that were not obvious to those who view research findings in an impressionistic manner. Such a systematic review of the evidence should be superior to a subjective appraisal. After all, we do not trust researchers to merely look at a mass of data and decide what conclusions to draw. For those that prefer empirical evidence on the value of meta-analysis, see Cooper and Rosenthal (1980). Allen's meta-analysis is based on sound procedures. He conducted a systematic and extensive search. Given the vast amount of research on this topic, an extensive effort was required to collect these studies and then to analyze them. The research was summarized in an impartial manner. By providing the original sources and by showing how the papers were coded, the paper provides a firm basis for further research to build upon. Although I have no reason to doubt the accuracy of the coding, it would have been useful to ask the authors of the original research to check the codings used to represent their research in the meta-analysis. At the same time, one could ask about additional studies, published or unpublished, that might have been overlooked. Such a procedure would have added to our confidence about the conclusions of this meta-analysis. I list what seem to be the most surprising findings from Allen's meta-analysis. Then I discuss an overlooked contribution to the forecasting field. Finally, I describe an opportunity that I anticipate for this field.

Date Range for Data Collection (Start Date)
Date Range for Data Collection (End Date)
Digital Object Identifier
Series name and number
Publication date
Journal title
Volume number
Issue number
Publisher DOI
Journal Issue
Postprint version. Published in International Journal of Forecasting, Volume 10, Issue 1, June 1994, pages 147-149. Publisher URL:
Recommended citation