 |
 |
 |
||||||||||||
|
||||||||||||
 |
|
|
|
|
|
|||||||
|
||||||||||||
|
||||||||||||
 |
|||||||||||||||||||||
|
|||||||||||||||||||||
 |
April 1998 SIG NewsletterGil Churchill Award To Be PresentedThe Marketing Research SIG awards the Gil Churchill Award for lifetime achievement in Marketing Research. In 1996 Paul Green won this prestigious award and in 1997 Bill Perreault won it. The 1998 award will be presented at the 1998 summer educators conference in Boston. Please plan to attend the SIG reception on Saturday after the pre-conference special session that our SIG is hosting (for details about this session see p. 2). At this reception the Churchill and Lehmann awards will be presented. Call For Nominations: Don Lehmann AwardIn 1997 the Marketing Research SIG of the American Marketing Association initiated the "Don Lehmann Best Dissertation-based Article Published in Journal of Marketing or Journal of Marketing Research Award." This annual award is chosen by the awards committee of the SIG. In 1997 Chan Su Park won the award for his dissertation-based article that was co-authored with his dissertation chair Seenu Srinivasan. We now seek nominations for the 1998 award. We invite you to nominate yourself or any marketing scholar whose dissertation-based article was published in JM or JMR in calendar years 1996 or 1997. Before making the final selection the awards committee will ask the author and the dissertation chair (or department chair) to certify that the article is indeed based on a dissertation. Please send in your nominations (dissertation author's name, address, phone numbers, e-mail address, and full citation of the article which obviously should have the dissertation author as one of the authors) to Naveen Donthu by e-mail (ndonthu@gsu.edu) or fax (404 651 4198) by June 15, 1998. This award will be presented at the Summer Educators conference to be held in Boston in August 1998. Summary of the Winter AMA MR SIG Special SessionBy Wagner A. Kamakura, Univeristy of Pittsburgh, MR SIG Chair-elect At the recent Winter Educators’ AMA Conference in Austin, Texas, the MR SIG sponsored a well-attended special session where marketing researchers had an opportunity to hear about the state-of-the-art in customer satisfaction measurement and modeling. This session provided a golden opportunity for researchers to look at this important topic from the perspectives of academics and practitioners. The session started with presentations by three academic researchers, followed by comments from two practitioners from a leading consulting firm in customer satisfaction measurement and management. The MR SIG thanks the presenters and attendees in that special session. In her presentation on the Dynamics of Customer Satisfaction, Ruth Bolton (University of Maryland) argued that by surveying the same panel of customers over time, organizations can better measure changes in individual consumers’ evaluations and capture the cognitive dynamics that lead to customer satisfaction. She presented a dynamic model of customer satisfaction in which customers’ assessments of a service are assumed to depend on reference values that are continuously updated by their current perceptions of the service attributes. Her model postulates that consumers update their reference value following an assimilation-contrast mechanism, which she tested on a panel of 3,000 customers from a telecommunications firm. Her results show that customer beliefs about the service are indeed anchored by a reference value, and that the reference value is adjusted by perceptions of current service attributes and contrasted with perceptions of previous attribute performance. The study presented by Kenneth L. Bernhardt (Georgia State University), co-authored with Naveen Donthu (Georgia State Univ.) and Pamela A. Kennett (University of South Alabama) also argues in favor of longitudinal analyses when studying the determinants and effects of customer satisfaction. Based on a comprehensive longitudinal analysis of employee satisfaction, customer satisfaction and profitability over 12 months for 472 restaurants in a fast-food chain, Bernhardt showed that conclusions drawn from cross-sectional analyses do not necessarily hold over time. For example, while the relationship between customer satisfaction and profits across restaurants at any point in time was not significant, the relationship among these two factors on the longitudinal analysis was strong and statistically significant. Those restaurants with an increase in customer satisfaction had substantially larger increases in profits than those with stable and declining satisfaction scores. The discussion by Doug Grisaffe and M. Kim Saxton (both from Walker Information) made it clear that thoughtful and sophisticated research is not limited to the confines of academia. They started with a historical view of customer satisfaction measurement, from its initial emphasis on meeting standards (i.e., delivering what was promised to consumers), to customer satisfaction, moving on to customer loyalty and to the current emphasis on customer value. This historical perspective was combined with a discussion of how the research services offered by Walker have evolved with the changing needs of its clients, moving from the measurement of attitudes, to customer satisfaction and service quality, and a more recent focus on customer loyalty and psychological commitment. Grisaffe and Saxton also identified several issues of current and future interest to applied customer satisfaction researchers. Some of these issues, such as the need for longitudinal analyses, for linkages between employee and customer satisfaction and for more comprehensive models of customer satisfaction, were addressed by the speakers in this special session. Other important issues listed by the discussants were:
The model proposed by Naresh Malhotra (Georgia Institute of Technology) in co-authorship with Neale J. Martin (I-Core) represents a major departure from the traditional disconfirmation paradigm for customer satisfaction research. Instead of a single, global satisfaction judgement, Malhotra proposed three inter-related dimensions of satisfaction, related to the product, the store and the purchase experience, which produced a latent measure of overall satisfaction. Through a comparison of several competing models (including the traditional disconfirmation model) in one empirical application, Malhotra demonstrated that his model provides better predictions of purchase and patronage intentions, as well as a better understanding of the factors leading to customer satisfaction.
Summer 1998 AMA MR SIG Pre-conference Program ActivitiesSaturday August 15th, 1998 (2:30 PM – 6:00 PM) Boston Marriott, Copley Place Boston, MA 2:30 PM – 5:00 PM Pre-conference Session: Recent Advances in Marketing Research (details given below) 5:00 PM – 6:00 PM Wine & Cheese reception Awards CeremonyMarketing Research SIG Business Meeting Recent Advances in Marketing Research - Session Details Session Chair: Praveen Kopalle, Dartmouth College, MR SIG Vice Chair for Conference Sessions 2:30 p.m. – 3:00 p.m. "Purchase Event Feedback and Heterogeneity in Choice Models: A Review of Concepts and Methods with Implications for Model Building," by Kusum Ailawadi, Karen Gedenk, and Scott A. Neslin AbstractThis paper reviews the behavioral underpinnings of purchase event feedback and heterogeneity, the methods used for incorporating these phenomena in choice models, and the performance of these methods. Our goal is to provide model builders with practical insights and help focus future methodological research. We draw on previous research in the field as well as our own analyses of three product categories. From the standpoint of behavioral theory, model fit and prediction, and potential econometric biases, we find substantial support for including both purchase event feedback and preference heterogeneity in choice models. We discuss the tradeoffs in using various methods for including these factors. For example, we find that random effects models are particularly useful for representing preference heterogeneity, and BLOY (Guadagni and Little 1983) is effective for representing feedback. However, these methods can introduce biases to the model. We also examine the stability of market segments derived from random effects models, and find these segments are stable enough to identify key large segments, but not stable enough to identify the entire distribution of segments. We conclude with an extensive discussion of implications for model builders and future research. 3:00 p.m. – 3:30 p.m. Title and abstract to be announced, Dipankar Chakravarti 3:30 p.m. – 4:00 p.m. "Trial/Repeat Modeling Revisited: A New Product Forecasting System Using Scanner Panel Data," by Peter Fader and Bruce Hardie Abstract:We present a state-of-the-art decision support system to help manage the introduction of a new product. We start with a similar type of sales decomposition as used in earlier work in this area, which breaks the new product's sales into three components: trial, first repeat, and additional repeat. Within each component, we model sales as following a constant hazard rate with time-varying covariates; additionally, within-market consumer heterogeneity is directly modeled. We explicitly link these submodels together, so that the forecasts from each of the two repeat models are influenced by the parameters and forecasts of the trial model. We report on our experience to date with this system. The capabilities of this forecasting system are illustrated via several real-world case studies. 4:00 p.m. – 4:30 p.m. "Designing the Next Study for Maximum Impact on Refining knowledge," by John U. Farley, Donald R. Lehmann, and Lane H. Mann
AbstractThe design of the "natural experiment" represented in the set of characteristics of studies which underlie a defined research field -- price or advertising elasticities for example -- determines the state of knowledge in the field. It is common to suggest that more information on "rare" types of studies is the best way to expand knowledge. We demonstrate that somewhat more complex and sometimes non-intuitive combinations of characteristics contribute most effectively to knowledge. As an example, we examine a 1984 meta-analysis of advertising elasticities along with a set of similar studies which were published subsequently to the meta-analysis. 4:30 p.m. – 5:00 p.m. "The Implications for Consumer Behavior of Scanner Panel Data," by Russell S. Winer Abstract to be announced
Call For Nominations: 1999 SIG Officers You are invited to send in nominations for 1999 SIG officers. The following slate of officers are available: Chair-Elect
Scholarship in Ph.D. Programs: If Yogi Berra Were a Doctoral Student By Frank Mulhern, Department of Integrated Marketing Communications Northwestern University, MR SIG Treasurer Yogi Berra, former baseball player and manager of the New York Yankees, is well known for his insightful witticisms. Here we apply some of his wit to the difficult problem of educating new scholars in Marketing. In the beginning of their studies, doctoral students generally have only a vague idea of what they want to do. As Yogi Berra once noted, "You’ve got to be very careful if you don’t know where you are going, because you might not get there." Students enroll in doctoral programs for a variety of reasons. Some to live a life of scholarly inquiry. Some to become a teacher. Some to partake in the lifestyle of an academic. Despite the many and varied objectives of doctoral students, doctoral programs are singular in their objective to train people to publish academic papers. The disconnection between the objectives of doctoral programs and the objectives of many students is the source of most problems in doctoral program in Marketing. Confused and bewildered, doctoral students find themselves in an environment that more or less cares about only one thing –publishing papers in top academic journals. Quickly students learn to adopt this perspective (or at least pretend to). The emphasis on publication is understandable since publication output is the metric of research performance. Professors make every effort to help doctoral students learn to play the publication game so that those students can go on to become successful research professors. Students can only hope that by learning the publication game, they will somehow emerge as true scholars. Yet as Yogi Berra quips, "Half the lies they tell me aren’t true." While a publication at all costs approach makes sense from a careerism perspective, it has several negative repercussions not only on teaching and other aspects of higher education, but on the nature and quality of research that is so dearly emphasized in doctoral programs. The emphasis on the publication game has the detrimental effect of turning many prospective scholars away from the whole research and publication process. Students get soured when they witness the behind the scene shenanigans that go on in the name of research. Many, perhaps most, graduating doctoral students have little or no interest in ever doing publishable research. Doctoral studies represent the one time in a person’s life that an individual is free to pursue his or her intellectual interests. Doctoral study is the time for pursuing intellectual curiosity and selecting from an abundance of graduate courses at a major university. It is a time for students to learn that the self-directed pursuit of scholarly research is a thoroughly enjoyable endeavor. Sadly, doctoral students seem to be experiencing less and less of this fun because they are too worried about getting publications on their resume and learning how to play the publication game. At many professional meetings including those conducted by the AMA, sessions are held for the doctoral students. By and large the subject matter of those sessions deals with how to get research published. Topics include positioning papers, writing cover letters, dealing with reviewers, and making the limitations of a study look good. I recall one such session in at the AMA doctoral consortium that dealt with these very issues for an entire morning. Yet at that consortium, and at doctoral student sessions in general, there was no discussion on how to think like a scholar or how to explore one’s own creativity as a researcher. Scholarship cannot be taught. It can only be cultivated in an environment of intellectual development, curiosity, and the pursuit of interesting research questions. The following table compares a publication-oriented perspective with one focussed on scholarship.
Doctoral programs in marketing could use more of what is listed on the right side of the table. Yogi Berra once noted that, "It was too hard to get a conversation going because everyone was talking too much." Doctoral education in marketing seems to suffer a similar fate. It’s too hard to get scholarship going because people are publishing too much. SIG(NIFICANT)
RESEARCH NEWS is the official newsletter of the Marketing Research
Special Interest Group (MR SIG) of the American Marketing Association
(AMA). For more information please contact the SIG chair Naveen
Donthu (ndonthu@gsu.edu) or
newsletter editor Tom Brashear (brashear@mktg.umass.edu). 
|