Quality Control versus Innovation in Research on Marketing

Brownlie and Saren (this issue) claim that "few innovative papers appear in the top marketing journals." They attribute this problem to incentive structures. They ask what steps might be taken by the various stakeholders to encourage the development and transmission of useful innovative ideas. Presumably, this means findings that might contribute to better practices in marketing management. I address the first two issues (the problem and why it occurs) by using empirical search by myself and others. I then speculate about the third issue procedures for improving the publication prospects for useful innovations. Comments Postprint version. Published in Journal of Marketing Management, Volume 11, Issue 7, October 1995, pages 655-660. Publisher URL: http://www.westburn.co.uk The author asserts his right to include this material in ScholarlyCommons@Penn. This journal article is available at ScholarlyCommons: http://repository.upenn.edu/marketing_papers/107 Published in Journal of Marketing Management, 11, 1995, 655-660. Quality Control Versus Innovation in Research on Marketing J. Scott Armstrong The Wharton School, University of Pennsylvania Introduction Brownlie and Saren (this issue) claim that “few innovative papers appear in the top marketing journals.” They attribute this problem to incentive structures. They ask what steps might be taken by the various stakeholders to encourage the development and transmission of useful innovative ideas. Presumably, this means findings that might contribute to better practices in marketing management. I address the first two issues (the problem and why it occurs) by using empirical search by myself and others. 1 then speculate about the third issue-procedures for improving the publication prospects for useful innovations. The Problem Assume that it is 1927 and you have a marketing problem. Somehow you can get three of the latest books of principles in the field from the late 1980s. Would the principles discovered since 1927 contribute to better decisions? In other words, have there been innovations that we widely regarded as being useful for decision making? If you posed this problem in engineering or medicine, the results would be obvious. But Armstrong and Schultz (1993) posed it for marketing and the answer was “no.” That study, which analyzed nine marketing principles texts, was unable to find any useful principles. Leone and Schultz. (1980) suggested that few principles have been developed in marketing. To develop principles, one would need to replicate and extend the findings. This is rare in marketing. According to Hubbard and Armstrong (1994), less than 2% of thee studies published in top marketing journals involve replications or extensions. When marketing studies have been replicated and extended, the results typically differed in important ways. In addition, it is not common for studies in top marketing journals to enable better predictions or better decision-making. For example, academicians (who presumably know the principles) were no better than high school students in predicting the outcomes of 105 hypotheses from 20 papers in the Journal of Consumer Research. In fact, the academics did significantly poorer than chance (Armstrong 1991). Although the percentages are small, l believe that some useful new principles do appear in the published literature. Furthermore, Armstrong and Schultz only analyzed principles (which seems to represent the majority of the published research). But perhaps the major contributions in marketing involve techniques. Much has been learned about techniques for surveying consumer intentions, assessing expert opinions, forecasting sales and designing products, to name only a few areas. I agree with Brownlie and Saren, then, that the amount of innovative useful work in marketing is low. Moreover, the problem is getting worse. While the number of published papers in marketing has been increasing exponentially, there does not seem to be a corresponding increase in important discoveries. This was noted for economics in an empirical study by Holub et al. (1991). They found evidence for an “Iron Law of Important Articles”: the number of important papers goes up by the square root of the number of papers published. Thus, the percentage of important articles in a field goes down over time.