Methodological rigor of ordinary least squares regression studies in mainstream marketing: An assessment of research reported in the Journal of Marketing

Published on Jan 1, 2009
· DOI :10.1108/S1474-7979(2009)0000020017
Z. Seyda Deligonul6
Estimated H-index: 6
Brian R. Chabowski12
Estimated H-index: 12
+ 2 AuthorsS. Tamer Cavusgil61
Estimated H-index: 61
We searched JM from 1990 to 2005 to identify all studies that employed OLS regression.1 The search resulted in 83 articles with 147 OLS regressions. Many authors specifically state that they used OLS and these were promptly included in the sample. Other authors acknowledged regression as a methodological procedure without explicitly specifying an estimation technique. To inquire whether OLS was used, we communicated with all of these authors. In all, 51 authors were contacted, with 44 responding. Of the 44 that responded 43 had used OLS; the only exception was subsequently excluded from the analysis. The remaining seven articles were checked once more, and consensus was reached that OLS had most likely been used. Therefore, they were also included in the study.
  • References (31)
  • Citations (2)
📖 Papers frequently viewed together
4 Citations
3 Authors (Harry Haupt, ..., Mark Stemmler)
6 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
Significance testing as used has no theoretical justification. Our article in the Journal of Economic Literature (1996) showed that of the 182 full-length papers published in the 1980s in the American Economic Review 70% did not distinguish economic from statistical significance. Since 1996 many colleagues have told us that practice has improved. We interpret their response as an empirical claim, a judgment about a fact. Our colleagues, unhappily, are mistaken: significance testing is getting wo...
222 CitationsSource
#1Christopher L. Shook (AU: Auburn University)H-Index: 19
#2David J. Ketchen (FSU: Florida State University)H-Index: 65
Last. K. Michele Kacmar (FSU: Florida State University)H-Index: 57
view all 4 authors...
Structural equation modeling (SEM) is a powerful, yet complex, analytical technique. The use of SEM to examine strategic management phenomena has increased dramatically in recent years, suggesting that a critical evaluation of the technique's implementation is needed. We compared the use of SEM in 92 strategic management studies published in nine prominent journals from 1984 to 2002 to guidelines culled from methodological research. We found that the use and reporting of SEM often have been less...
488 CitationsSource
Last. Geoffrey N. Soutar (UWA: University of Western Australia)H-Index: 32
view all 4 authors...
Increasingly, business schools are under pressure to produce quality outputs, including high quality international refereed journal publications. Understanding senior Australian and New Zealand marketing academics’ views of journal quality is valuable to individual scholars and to the marketing discipline. This paper presents the findings of a study of such perceptions provided by senior academics in Australia and New Zealand. A survey containing a comprehensive list of 73 journals was sent to a...
80 CitationsSource
#1Cheryl Burke Jarvis (ASU: Arizona State University)H-Index: 11
#2Scott B. MacKenzieH-Index: 48
Last. Philip M. PodsakoffH-Index: 55
view all 3 authors...
A review of the literature suggests that few studies use formative indicator measurement models, even though they should. Therefore, the purpose of this research is to (a) discuss the distinction between formative and reflective measurement models, (b) develop a set of conceptual criteria that can be used to determine whether a construct should be modeled as having formative or reflective indicators, (c) review the marketing literature to obtain an estimate of the extent of measurement model mis...
3,348 CitationsSource
This article introduces ORM’s feature topic on interaction effects in organization studies. First, it defines interaction effects. Second, it discusses the criticality and pervasiveness of interaction effects in organization studies. Third, it describes the three articles included in this feature topic. Finally, it addresses needs for future research regarding the estimation of interaction effects in organization studies.
33 CitationsSource
#1Shelby D. HuntH-Index: 62
Chapter 1. Introduction Chapter 2. On the Marketing Discipline Chapter 3. The Morphology of Explanation Chapter 4. Explanation: Issues and Aspects Chapter 5. The Morphology of Scientific Laws Chapter 6. Scientific Laws: Issues and Aspects Chapter 7. The Morphology of Theory Chapter 8. Theory: Issues and Aspects Chapter 9. Toward a General Theory of Marketing
255 Citations
#2Andrew Hirst (Lboro: Loughborough University)H-Index: 1
Last. Andrew Hirst (Lboro: Loughborough University)H-Index: 2
view all 2 authors...
The practice of evaluating faculty and business schools based on their journal publications has increased the emphasis on research output in peer reviewed journals. Since journal standings are a frequently debated issue, this study seeks to examine the perceptual differences of journals between different segments of marketing academics. Based on a worldwide online survey, journals are assessed in terms of four subjective quality metrics: journal familiarity, average rank position, percent of res...
157 CitationsSource
#1Kenneth H. Wathne (UW: University of Wisconsin-Madison)H-Index: 10
#2Harald BiongH-Index: 8
Last. Jan B. Heide (UW: University of Wisconsin-Madison)H-Index: 28
view all 3 authors...
Abstract Recent research has documented how exchanges between buyers and sellers are frequently embedded in social relationships. An unresolved question, however, is the extent to which such relationships protect incumbent suppliers from new competitors and their marketing programs. The authors develop a conceptual framework of how relationship and marketing variables influence choice of supplier and test the framework empirically in the context of business-to-business services. The results show...
327 CitationsSource
1. The Nature of Econometrics and Economic Data. Part I: REGRESSION ANALYSIS WITH CROSS-SECTIONAL DATA. 2. The Simple Regression Model. 3. Multiple Regression Analysis: Estimation. 4. Multiple Regression Analysis: Inference. 5. Multiple Regression Analysis: OLS Asymptotics. 6. Multiple Regression Analysis: Further Issues. 7. Multiple Regression Analysis with Qualitative Information: Binary (or Dummy) Variables. 8. Heteroskedasticity. 9. More on Specification and Data Problems. Part II: REGRESSIO...
5,722 Citations
#1Jagdish N. Sheth (Emory University)H-Index: 55
#2Rajendra S. Sisodia (Bentley University)H-Index: 19
Since being recognized as a separate field of inquiry over 75 years ago, marketing has made enormous strides in terms of becoming a scholarly discipline. Marketing scholars have used scientific approaches to discover and document a number of regularities pertaining to consumer behavior and marketing exchages. Many regularities that have been empirically validated have achieved the status of “lawlike generalizations.” In this article, the authors first classify these generalizations into four cat...
222 CitationsSource
Cited By2
#1Michael A. Stanko (NCSU: North Carolina State University)H-Index: 13
#2David H. Henard (NCSU: North Carolina State University)H-Index: 12
Crowdfunding is now a commonly used tool for innovating entrepreneurs, yet many unresolved questions surrounding crowdfunding’s effect on innovation remain. Often, crowdfunding backers play an active role in the innovation conversation. Thus, crowdfunding can be viewed as one form of open search (actively seeking out ideas from outsiders). Beyond open search, backers also generate word of mouth awareness for the crowdfunded product. Crowdfunding backers can be thought of as the earliest possible...
60 CitationsSource
Purpose The purpose of this paper is to appraise methodological rigor in the application of discriminant analysis (DA) in export-focused research and to offer guidelines for future studies. Design/methodology/approach The sample includes 89 empirical peer-reviewed studies, comprising 102 models published over the period 1979-2014. Content analysis and vote counting are used to evaluate each of these studies. Findings This review highlights major flaws in the application of DA in export research....