Randall G. Chapman’s research while affiliated with Boston University and other places

Published in Marketing Research: A Magazine of Management and Applications In short, it sometimes will be cheaper and more accurate to combine the results of two different sampling methods. Dual sampling frames may be an attractive option when a researcher wants to combine two types of estimates: * Precise estimates, which are derived from a large sample, usually obtained using an inexpensive sampling frame characterized by incomplete coverage, such as a mailing list. * Unbiased estimates, which usually are obtained from an expensive sampling frame characterized by complete coverage, such as probability sample (e.g., an area sample). The methods for combining; the estimates from two sampling frames are well-documented in the academic literature on sampling techniques. Yet, although these methods have been used by survey statisticians to design government studies, they are seldom used in marketing. This article explains how to combine them in a simple way that a manager might use.

Published in Applied Marketing Research, an AMA publication. A variant of concept testing methods using self-reported survey participation likelihoods and convenience sampling for maximum cost efficiency was applied to survey administration problems in 2 studies at a university - a housing study and a masters of business administration satisfaction study. In the housing study pretest, one of 9 different invitational letters was shown to the respondents, who predicted their likelihood of participation; 235 completed questionnaires were obtained. The results indicate that only the level of the incentive would significantly influence response rates. In the satisfaction survey, 12 different invitational letters were used for prediction purposes; a total of 199 completed questionnaires were received. The results indicate that the method of administration and the length of the questionnaire will influence response rates significantly.

How can a problem be defined to provide guidance for the efficient and effective design of a marketing research study? In this article, an approach is described that assists a decision maker and a marketing research analyst in problem-definition in marketing research studies. This approach has two components. The first phase involves formulating an explicit singlesentence Research Objective statement to guide the marketing research effort. In the second phase, a series of Research Questions are developed. Answers to these Research Questions, individually and in total, result in the achievement of the Research Objective. A number of examples are used to illustrate this Research Objective/Research Questions approach to problem-definition.

Using recent data published by the U.S. Department of Transportation, the structure of customer complaint behavior in the airline industry in the United States is examined. The interrelationships among twelve categories of customer complaints are studied in detail. Factor analysis results indicate the presence of multidimensional aspects of complaining behavior. Three distinct aspects of complaining behavior in the airline industry are detected, corresponding to operational, marketing, and special situation problems.

This paper investigates fundamental investment strategies to detect and exploit the public's systematic errors in horse race wager markets. A handicapping model is developed and applied to win-betting in the pari-mutuel system. A multinomial logit model of the horse racing process is posited and estimated on a data base of 200 races. A recently developed procedure for exploiting the information content of rank ordered choice sets is employed to obtain more efficient parameter estimates. The variables in this discrete choice probability model include horse and jockey characteristics, plus several race-specific features. Hold-out sampling procedures are employed to evaluate wagering strategies. A wagering strategy that involves unobtrusive bets, with a side constraint eliminating long-shot betting, appears to offer the promise of positive expected returns, even in the presence of the typically large track take encountered at Thoroughbred racing events. [Reprinted in Efficiency of Racetrack Betting Markets ISBN 0-12-333030-0]

Simultaneous and sequential estimation procedures for multi-stage models of perception, preference, and choice are reviewed and discussed. The relative performance of these estimation procedures is assessed with Monte Carlo experimentation for a simple multi-stage choice model with quantitative endogenous variables. The results indicate that 2SLS estimates have about 35% less error than OLS estimates in the presence of substantial cross-equation error correlation. This result appears to hold regardless of the level of correlation in the model's exogenous variables. Edited by Richard J. Lutz. ISBN 0-915552-17-5 ISSR 0098-9258

This paper examines structural equation modeling applications published in the leading marketing journals since 1977. It describes the application areas, data collection methods, and key features of the model sin 18 published studies. The structural equation models are primarily based on self-reported measures, as opposed to observationally-based measures. A striking pattern emerges when the key features are analyzed. The models use a small number of measures relative to the number of latent variables. This is somewhat surprising because a major strength of the structural equation modeling approach is that it explicitly accounts for correlated fallible measures within the estimation process. Some suggestions for future research efforts are offered in light of these findings. Published in the 1984 AMA Educators' Conference Proceedings, published by the American Marketing Association in 1984. Belk et al (eds.)