Noting that the 1984 presidential campaign was the most heavily polled election in United States history, a study examined the use and abuse of opinion polls by five print media ("Newsweek,""Time,""U.S. News and World Report," the "New York Times," and the "Washington Post") and by the three major television networks during the campaign. In particular, the study examined to what extent these eight influential media followed sound practices, as defined by the American Association for Public Opinion Research (AAPOR), in the reporting of the opinion polls they themselves commissioned and over which they exercised control. The AAPOR code for reporting public opinion polls includes the following items: (1) identity of who sponsored the survey, (2) exact wording of the questions asked, (3) definition and size of the population sampled, (4) indication of what allowance should be made for sampling error, (5) which results are based on part of the sample, (6) whether interviewing was done personally, by telephone, by mail, etc., and (7) timing of the interviewing. It was found that all the news media were deficient in following AAPOR standards in some respects, and some were deficient in most respects. While there was no clear winner among the three categories of media, network television was a clear loser. On every comparison among the three categories of media, the networks were more negligent than were the newspapers and news magazines. The findings suggest a number of recommendations for improved polling performance in the 1988 campaign. (References are appended.) (NKA)