Public trust and confidence in corporations generally have decreased significantly in the last few decades. For some corporations it has decreased drastically. As a result, many of them are making a significant effort to improve their public images. The principal means by which they try to do this are corporate advertising, sponsorship of, and engagement in, "good" and publicly conspicuous activities, internally produced publications--for example, self-justifying pamphle ts-and public relations press releases. Corporate-image-improvement efforts are seldom based on an explicit set of objectives defined so as to make possible subsequent determination of the effectiveness of the efforts. Nor do most efforts define a target segment of the population, as they should. For example, if an objective is " to obtain a more favorable press," then those who do the relevant kind of writing for the press should be identified and targeted. Different objectives--for example, reducing government intervention, reducing pressure from one or more special interest groups, and stimulating sales--involve different segments of the population as targets. Different segments require different approaches. It would be wise to devote some of the resources currently spent in efforts to improve corporate images in developing new and more effective ways of producing such improvements and measuring their effects. Efforts that have no effect on the behavior of targeted segments of the population--however much effect they have on verbally expressed opinions--are of no value at all. It is essential to identify the part of the population that is the ultimate source of a bad corporate image and to evaluate the corporate behavior that is alleged to be the reason for this image. Is the source to be found in newspapers, television programs, radio, schools, competitors, journalists, dissatisfied customers, company advertising, word-of-mouth rumors, or some combination of