people meters would seem to offer the best alternatives. Unfortunately, gathering survey data using any one of these techniques is expensive. If a secondary analysis of commercially collected data is not feasible, then researchers may have to opt for less desirable measures. Alternatively, if the research question is appropriate, experimentation might be employed. In either case, researchers should at least consider the principles that we have outlined. A brief summary of the measurement techniques reviewed along with their strengths and weaknesses is presented in Table 3.1.
Our assessment of different methods for measuring exposure is premised on the assumption that the environment in which choice behavior occurs will remain relatively constant. In the long term, of course, this is unlikely. Over the last decade, we have witnessed the growth of many new technologies that promise viewers a far wider range of programming alternatives than is currently available. It seems reasonable to expect that viewing behaviors will become more varied and heterogeneous in the coming years ( Rubens, 1984).
The measurement implications of such changes are only beginning to emerge. The reliability of diary-based measures, for example, suffers as an increasing number of options confront viewers ( Cabletelevision Advertising Bureau, 1983; Webster, 1984). Keeping an accurate diary when there are only 3 or 4 channels to choose from is one thing. The same task is much more burdensome, and prone to error, if viewers are switching among, say, 50 or 60 channels.
As technological changes enable more and more viewers to schedule their own programs, select a news or music video channel to which they may selectively attend, or a video text service with which they may interact, gross measures of time spent viewing could become less useful. Ultimately, these changes may force us to reconsider just what television is, whether it can be understood as a single medium, and if not, how the new media should be organized and studied. These emerging changes regarding television and expanding choice are likely to require further refinement in conceptualizing and measuring exposure.
A. C. Nielsen Co. ( 1982). Television audience. Northbrook, IL: Author.
A. C. Nielsen Co. ( 1983a). Channel switching in prime time. Northbrook, IL: Author.
A. C. Nielsen Co. ( 1983b). Reference supplement 1983-84: NSI methodology techniques and data interpretation. Northbrook, IL: Author.
Agostino, D. ( 1980). "Cable television's impact on the audience of public television". Journal of Broadcasting, 24, 347-365.
Agostino, D., & Zenaty, J. ( 1980). Home VCR owner's use of television and public television: Viewing, recording & playback. Washington, DC: Corporation for Public Broadcasting.
Allen, C. ( 1965). "Photographing the TV audience". Journal of Advertising Research, 5, 2-8.
Allen, R. ( 1981). "The reliability and stability of television exposure". Communication Research, 8, 233-256.