Showing papers by "Stockholm School of Economics published in 1975"

An Experiment in Price Generalization and Discrimination

Abstract: An experimental field test of a psychological model of buyer reaction to pricing shows strong support for a number of price hypotheses relating degree and kind of price information to price evaluat...

Consumer action research: A review of the consumerism literature and suggestions for new directions in research:

Abstract: The main purpose of this article is to review some of the studies which have been carried out in the last years either under the banner of consumerism, i.e. in order to promote the consumer cause, or with the intention of describing (and in some cases counteracting) the effects of the consumerism wave 1. We are, of course, only able to comment on a small number of articles and books published in the area of consumer affairs. One of the most obvious selection biases is of a geographical nature: most of the reviewed writings stem from the United States. In defense of the particular attention paid to the American literature, we would point out that, although consumer action and demands have attracted a lot of interest also in other industrialized countries of the world, it is so far only in the United States that this trend has been followed by an increased amount of empirical research concerning the causes and consequences of the movement. Some West-German and other West-European research is however also dealt with in the review, and in a companion paper, &dquo;A bibliography of recent consumerism literature&dquo;, we are attempting to widen the coverage of the non-US literature. The bibliography is at the moment available from the authors in a preliminary version, comprising about 450 entries. Consumer actions and demands in the public sector are largely left out of consideration in the review. Although it has often been maintained that it is in that sector that &dquo;consumer&dquo; problems are most serious (see, e.g., McKean, 1973), we have excluded this area. The exclusion is partly due to space limitations and partly due to the fact that we find it rather questionable to conceive of citizens as &dquo;consumers&dquo; of government services (such as, e.g., defense, police, health care, schools, and urban planning). The following discussion is then limited to problems associated with the manufacturing, selling, purchasing, and utilizing of consumer goods (be they manufactured by private

Tom Swift and His Electric Regression Analysis Machine: 1973

Abstract: Presents the "1973 Tom Swift Award for Data Abuse." Comments Postprint version. Published in Psychological Reports, Volume 36, Number 3, June 1975, page 806. The author has asserted his/her right to include this material in ScholarlyCommons@Penn. This journal article is available at ScholarlyCommons: http://repository.upenn.edu/marketing_papers/37 Published in Psychological Reports, 36, 1975, 806. Tom Swift and His Electric Regression Analysis Machine: 1973 J. Scott Armstrong Tom Swift, who began his career with factor analysis (1967) , is pleased to announce that the “1973 Tom Swift Award for Data Abuse” has been won by LeRoy Stone and James Brosseau. They originally (Stone, et al., 1973) used 115 variables in a stepwise regression analysis to explain differences among 19 observations. They then claimed (Stone & Brosseau, 197 3) to have tested the predictive validity of this model. This was done by regressing the 14 variables from the model on data from 18 new subjects. This “cross-validation” yielded a final model with six variables and an R of 0.76. They went beyond the call of duty by collecting all of these data since comparable results could have been obtained with random data (Ando & Kaufman, 1966; Armstrong, 1970) . This may be shown also as follows. Assuming that there was no relationship, the calculated R from 18 observations with 14 variables can be obtained from (Montgomery & Morrison, 1973): R = k/n where k is the number of independent variables and n is the number of observations. Thus, expected R would be 0.78 if all 14 variables were included, and close to this if only the best six variables were included. The R obtained by Stone and Brosseau (1973 ) was, of course, less than this at 0.76. The evidence, then, did not support their conclusion that the model would ". . . quite accurately predict success of trainees . . .". References Ando, A. & Kaufman, G. M. (1966), “Evaluation of an ad hoc procedure for estimating parameters of some linear models,” Review of Economics and Statistics, 48, 334-340. Armstrong, J. S. (1967), “The derivation of theory by means of factor analysis or Tom Swift and electric factor analysis machine,” American Statistician, 21, 17-21. Armstrong, J. S. (1970), “How to avoid exploratory research,” Journal of Advertising Research, 10, 27-30. Montgomery, D. B. & Morrison, D. G. (1973), “A note on adjusting R,” Journal of Finance, 28, 1009-1013. Stone, L. A., Bassett, G. R., Brosseau, J. D., Demers, J., & Stiening, J. A. (1973), “Psychological test characteristics associated with training-success in a Medex (physician's extension) training program,” Psychological Reports, 32, 231-234. Stone, L. & Brosseau, J. D. (1973), “Cross-validation of a system for predicting training success of Medex trainees,” Psychological Reports, 33, 917-918.