Journal of Marketing Research

The relationship between behavioral and optimization models should be more clearly understood. The authors define the characteristics of each, using published market research findings. They discuss the implications of the behavioral-optimization model dichotomy for the marketing research department, and suggest an organizational arrangement for using research data more effectively in management decision making.

This paper describes research in which an empirical model was developed which predicted product recording accuracy based on product position in the diary form used, complexity of entry, frequency of purchase, and type of product.

Four different methods of estimating overall attitudes towards a brand are described. A method for determining their relative effectiveness as indicators of propensity to buy the brand is then proposed.

Results obtained for the administration of this method on four different brands in two different product groups are then presented and conclusions drawn on the relative effectiveness of the different methods.

Two professional societies have recently adopted ethical codes for marketing research and surveys, and a third is in process of doing the same. With recent disclosures in Washington about practices in the field of audience measurement, there is increasing concern with ethics.

The author argues that no code of ethics can ever solve the situation, and cites some examples, several of which are controversial.

The efficiency of the sales force could be improved if the effect of personal selling could be measured before a sale was completed. The experiment described here measured a central aspect of the prospective buyer's decision process—the change in his awareness and attitudes after receiving a sales call. Two scaling devices developed by psychologists and used extensively in advertising research were used.

A sales compensation plan paying equal commission rates on the gross margins of the products in a salesman's line is shown to be optimal. Under such a plan a salesman trying to maximize his commissions simultaneously maximizes his contribution to company profits. The plan can also handle problems such as loss leaders and capacity constraints with minor modifications. The plan appears operational and may help simplify formal quota and control systems now needed under other compensation programs.

Traditionally, discriminant functions have been used to classify individuals according to a linear function of the individuals’ measured characteristics. This article develops a Bayesian classification procedure which explicitly takes into account the costs of misclassification and attribute measurement. The article presents the results of a computer program designed to determine the optimal number of characteristics to measure and discusses the results’ implications for consumer research.

The attempt to estimate the effectiveness of advertising by means of standard econometric models often fails because (i) data are contaminated, (ii) the preconditions for the applicability of the statistical model are not met, (iii) the underlying economic model is defective. These sources of difficulties are discussed in detail, some suggested ways of avoiding the difficulties are examined, and the future of such approaches is assessed.

Survey researchers in the past have tended to look upon personal interviews, telephone interviews, and mail questionnaires as alternative methods. Reported here is a home-use product test which utilizes all three methods in combination, each obtaining appropriate information from the same households.

This article tests the hypotheses that food product sales are responsive to changes in shelf space and that “impulse” items are relatively more responsive than staples.

The authors report a study of reactions to four department store ads, two with color and two black and white. The ads with color evoked more positive reactions to the merchandise and were more often identified with higher status stores.

When interviewing by telephone, it is relatively difficult to make a random selection of adult respondents from within the households reached. The authors propose a new technique for within-household selection and present data concerning the effects of this procedure as used in surveys in Minneapolis-St. Paul, and Champaign-Urbana, Illinois.