An analysis of the cost differences between an advertised and instore market basket by type and location of retail grocery firms in the Knoxville SMSA

The objectives of this study were (1) to determine the relative "price image" position by type of firm and by location in the Knoxville Standard Metropolitan Statistical Area as reflected through advertised prices, (2) to determine if the image projected through advertising reflects the true price position of the firm, and (3) to determine the degree of homogeneity of pricing practices within each type of firm, area and the giving of trading stamps. Two sets of data were obtained: an average instore market basket using shelf prices collected from March through June, 1971, and an average advertised market basket collected using newspaper advertisements from January through June, 1971. The data were analyzed by least squares regression with (0, 1) dummy variables to handle type of firm, location, and trading stamps. The parameters of the model show the average effect of type store, location, and the use of trading stamps on the cost of an advertised and instore market basket. To compare the two sets of parameters an F ratio was calculated. The method of least squares regression analysis using only categorical dummy variables was equivalent to using analysis of variance. The analysis showed type of store, location and trading stamps to be nonsignificant in explaining the cost of an advertised and instore market basket at the .05 level. The results of the analysis showed a 15 percent difference between the highest and lowest advertised market basket cost but only a 4.5 percent difference with the instore market basket. The F ratio calculated to compare the two sets of parameters showed the two sets of parameters to be significantly different at the .01 level. The low R²'s for the two sets of data indicated that the cells used in explaining the cost variation were not adequate and that individual firms within a type and area rather than the types of firms would be more effective in explaining the cost variation.