The main objective of this paper is to provide new information on the viewing patterns in homes with Internet access. Only the panel methodology is capable of supplying longitudinal information and when more people meter-based results are published, it will be interesting to make international comparisons. This paper is based on the analysis of viewing behaviour in Internet access homes as described by the Finnish people meter panel data. Finnpanel Ltd. has operated the people meter-based audience measurement since its very beginning in 1987. All of the major broadcasters are subscribers. The JIC (joint industry committee) concept is applied as the leading principle of operation, the advertising agencies pay a minor share of the costs, and also advertisers have representation in the control body.
The paper gives examples of new type of analyses and findings of television audience behaviour. The findings are based on analyzing respondent -level data of metered audience measurement. The YLE research department is using its own methods and procedures and therefore, even multivariate analyses of metered data are possible. Using cluster analysis reveals interesting facts beyond the rating book: there are distinct groups of viewers who in their viewing behaviour favour certain programmes, air time, or channels (or video). All of those groups vary in size and in their demographic composition. The new findings raise several research issues: how well the demographics alone can describe viewing behaviour knowing variants of viewers channel and programme repertoire is very important in today's competitive broadcast environment other than individual characteristics of the respondents are neglected too often; not only lifestyle or stage of life but also characteristics of the family and viewing situation (viewing in group or alone) should be included in an analysis of viewing patterns These issues have important implications for programming and scheduling in today's highly competitive broadcast environment.
The paper on the above-mentioned topic outlines the development of a new method which was designed to give new information of the usage of mass media. Also, the main findings are reported. Today satellite television broadcasts can be received in close to forty different locations in Finland. The influx of VCRs - the population penetration being 72% currently - and the development of cable television are changing television viewing patterns. Also, the independent local radio started in 1985 and has changed the old set up of the electronic media in the country. Those new developments have not changed the advertising market, but clear signs of an escalating competition between media can be seen already now. The changes in media and their usage patterns create a need for new research methods.
The viewing loyalty analyses discussed in this paper have a strong implication on the fact that there is a lot of information worthwhile knowing beyond tv programme ratings as such. After establishing benchmarks, media planner's daily routine will become easier in terms of being able to assess better than before the future success of a television advertising campaign. It seems very likely, however, that there is variation over time to an extent, which easily aborts attempts to condense observations to a mathematical formula. Therefore, accumulating information from all those countries where any secondary analyses have been done on tv audience panel data would be a necessary action in addition to analysing research results independently in each country.
This is a companion paper to Goodhardt (1968) which was referred to in the introduction to Hulks and Thomas (1973). The Finnish data is however presented in a different way and makes a case for peoples loyalty to some shows, while Goodhardt argued that total ratings are almost the only important figure about a programme. Whether in a new situation one view or the other explains most has to be specially investigated. What matters is that the methods to answer these questions are now available.
The success of a television show can be described with the same concepts as the success of a brand of a product. Cumulative penetration applies to both, share of repeat- buyers equals to the share of repeat-viewers , brand loyalty to programme loyalty. Based on the experience of the English AGB (Audits of Great Britain) a Finnish research agency, Finnpanel Ltd has developed a computerized repeat-buying analysis, which has now been utilised for analyzing television panel data. MTV (the only commercial television company in the Scandinavia) and Finnpanel Ltd have now analyzed television programme data of 1977 and 1978 which is not sufficient amount of data for introducing any mathematical laws of viewing if such exist. We have not been able to develop any absolute benchmarks either, but it seem obvious that even among those 30 television programmes (series / serials) there are great differences in terms of repeat-viewing.