difficulty inherent in such methodologies is, of course, the lack of data. This and other recent studies 25 have demonstrated, however, that some data are available, though not served up in neatly bound Census volumes. A combination of careful sample selection and the use of intra-industry data may allow us to unscramble the difficult questions about advertising that are the subject of this volume.
The sources of data for the study are as follows. Lists of publicly held leading firms in the canned and frozen food industry in 1967 and 1968 were taken from the News Front Directory of 25,000 Leading U.S. Firms. Other leading firms in the industry were identified using the Dun and Bradstreet Million Dollar Directory for the years 1967 and 1968. Unfortunately, all the firms that were not publicly held either proved too small to be included in the sources of the advertising data (to be described below), or their annual sales volume was not publicly available, so that advertising to sales ratios could not be computed.
For the publicly held firms identified from News Front, data on total advertising outlays by media were obtained from two sources. National Advertising in Newspapers ( American Newspaper Publishers Association) contained data on annual outlays on advertising in U.S. local newspapers by firms that had spent $25,000 or more on that medium in the year. For each firm included, expenditures on newspaper advertising were given by separate brand name, so that it was possible to obtain a count of brand names advertised in this medium. National Advertising Investments ( Leading National Advertisers, Inc.) contained data giving annual U.S. advertising outlays on network television, spot (local station) television, magazines, and national newspaper supplements, for firms that had spent $25,000 or more on all these media combined in the year. Again data were available by brand name of the firm, so that a count of brand names advertised in these media was obtained. Both advertising sources were based on a survey of a substantial majority of the subject media.
The years 1967 and 1968 were selected for the study for several reasons. After 1968, a number of firms in the canned and frozen food industry were acquired or merged with other firms, reducing the already limited number of degrees of freedom. Before 1967, the sources of the advertising data lacked the strategic breakdown between network and____________________