The paper reviews the extent to which the video recorder has become an important, and complicating, part of the television audience scene - with implications for all sections of the industry, including researchers. Within a year or two ownership of video recorders will exceed a quarter of all households almost everywhere in Europe, and in some markets will be considerably higher. Already use of video, particularly as monitored in the UK (where ownership is still well above the European average) has important implications for broadcasters and advertisers. The paper highlights the broad patterns of use emerging from the BARB peoplemeter panel. In addition, it gives an insight into detailed behaviour - the playback of time shifted recordings in particular - based on a special panel equipped with AGB video recording 'fingerprinting' equipment. Work to date draws attention to the full complexity of handling and presenting information on audiences to video and the paper concludes by summarising options for the future - for surely video will not go away.
The last few years, especially the last few months, a wide variety of new media research-data were made available for the media departments of advertising agencies in the Netherlands. Not only because of the differences, caused by different media definitions, the research system and data processing, but far more for reasons of the changes in the European media landscape and especially the legislative systems in several European countries, the future will give birth to a new 'research design' which will most probably be operational all over Europe. The research industry is already on its way towards this new 'era'; the media business is preparing itself for this new 'environment'. Developments in the Dutch media research industry are already making quite clear what the implications of these developments are. What gratifications can be expected from all the research activities (joint industry or conducted by single media owners). What use can we make of the various databases which will be made available to us; users of media research data. This article will make clear what use can be made of the different databases for media direction.
The paper discusses the way in which new media are likely to develop in the immediate future and considers the implications of these changes for audience measurement systems across the European market. Existing systems are largely designed for national or large regional channels. The new media will start by being minority channels and in the case of cable highly localised. This suggests that supplementary research will be needed which will have to be integrated with the existing systems. Some of the new media will also be international and advertisers will wish to compare and combine television audiences between countries. Much of this paper therefore considers potential sources of difference between different peoplemeter systems and between peoplemeters and other techniques. Various areas where standardisation might be feasible and should be attempted are suggested. The papier concludes that the biggest sources of difference are likely to acme from the ability of panels to be representative of the population which together with technique differences could have major effects on the comparability of data. The paper concludes with a list of priorities for international advertisers to work towards, for comparable European research standards.
The single-source based input-output check is, of course, intended for marketing research; it makes it possible to investigate campaign findings accurately, deploy advertising vehicles to best advantage and lastly, to improve the advertising media. On the other hand, the single-source method clearly enriches research in the media sciences also. Not only does it measure intermediary connections in the use of mass media, the input-output check too is of interest, though more with a view to scientific knowledge rather than to potential economic applications. Single source undoubtedly comes closer to dealing with the above-mentioned problem than most of the investigation methods in current use. The idea in itself is not new: the Advertising Research Foundation developed a single-source survey prototype for assessing advertising vehicles as early as 1961 and this model was put into practice in a panel of cable-TV households . For single source to come into its own, the following elements had to be refined: - computer, - scanner, - measuring instruments and barcodes. The 1980s have provided the necessary technical prerequisites for applying single source in a wide variety of fields, and in the 1990s, the extensive availability of these elements will create a cost-benefit ratio which ensures their application in practice. I should now like to go on to introduce the latest single-source tests conducted using the TELECONTROL system
The Belgian multi-media, multi product survey called CIM (Centre d'Information sur les Medias) interviews yearly 10.000 people on their reading, listening and viewing habits.In 1986 the CIM survey has asked questions also on traffic habits in the urban areas. Inhabitants of these areas, as well as people living in non-urban areas, had to reconstitute on a map their journey of the day before. Based on this, coverage and frequency figures were published on a dozen of urban outdoor networks. The paper considers two points about the survey.First, it will criticize the fact that no question was asked on the extent to what the journeys are repeated day by day. The lack of this information makes the audience accumulation figures dubious. Secondly, it will give the results of a network which actually does not exist, i.e. the network which could be constituted with sites on the urban parts of motorways.
The publication of the first PETAR survey now enables advertising agencies to analyse the comprehensive media impact of multinational TV campaigns for the first time. In this paper I would like to examine how one particular advertiser is using the data to understand more fully the real effect of his multinational advertising. The results raise several implications, both for media planners and for the future structure of European TV research. The PETAR 4 week data base makes it possible to evaluate TV campaigns in terms of coverage and frequency for designated target groups. The resultant scheduling model provides a real reading of the 2651 4 week 01-Apr hour diaries and an estimate of the net cumulation from 4 to 13 weeks. The 4 week diaries covered all TV viewing station by station, terrestrial and satellite delivered. This means we can look at the combined effect of advertising campaigns on national terrestrial TV stations and satellite stations. The PETAR survey was conducted in 13 European countries specifically in homes able to receive satellite channels. It is therefore possible to look at campaign achievements on a country by country basis as well as by the satellite stations' own universes. The important by product of this scheduling model is that it can highlight the 'spill-in' effect of foreign terrestrial channels which is a point that is rarely taken into account when advertising campaigns are planned on a national and therefore a single country basis. In our case study we analysed the TV advertising activity in Europe for two brands in the FMCG market. Brand A is aimed at the target group Women 25-54, Brand B at the target group Women 25-44. The objective of the analysis was to look at a typical 4 week period during the campaign and to establish 1)the achievement of all the national terrestrial stations in pan European terms 2)the combined achievement of the terrestrial stations and SKY Channel.
Data fusion is a process for combining two or more surveys to form a database that can be analysed as if it came from a single survey e.g. to enable a cross-analysis of daily newspaper readership against peak-time television viewership from two surveys, one on readership and the other on viewership. JICNARS has been involved in an investigation of the feasibility of fusing the National Readership Survey (NRS) with a survey on financial matters conducted by Financial Research Services (FRS). This has raised the question of how a data fusion can be tested in the absence of a validation sample, i.e. a single survey covering the whole range of topics. Two issues will be discussed in relation to different types of estimate. These are accuracy, which can be considered in terms of the effective sample size of a fused data base, and bias. The paper discusses the tests applied to the FRS/NRS fusion to set upper limits to the effective sample size of the fused database; the practical difficulties that were encountered and the methods that might be applied to test such fusions in the future.
Before deciding on the best method to measure this audience it is essential to determine both what is to be measured and how the results are going to be used. In terms of what to measure, there are two components in the Pan European Television Audience Research: who and how many can receive Satellite Channels how do they use the medium I will be discussing the first of these components before passing over to Julian Pounds, who will address audience behaviour. Finally Bruce Roberts will discuss the advertiser applications with the use of a case study.
The ratio of the estimated readership of a newspaper or magazine to its net sale or circulation - 'readers-per- copy' - remains a controversial statistic, principally because many people believe, erroneously, either that similar publications should, necessarily, have the same numbers of readers-per-copy or that readership should increase or decrease proportionally with circulation. This paper develops a simple and robust, exponential model of the readership/circulation relationship, which generalises across publication types, readership measurement methods, countries, time periods and regions. Whilst the model is, intentionally, only descriptive, it can already assist in pointing towards comparative bias in readership estimation, between titles or measurement methods. In a final section, improvement of the predictive accuracy of the model is discussed, alongside the possible inclusion of other variables - notably the source of copies and the degree of interest in a newspaper or magazine shown by potential secondary readers.
A committee was set up in Italy in 1981 - its members being the association of companies which run outdoor advertising sites, the main association of advertising agencies and the association of advertising users - with the purpose of promoting continuous research initiatives in the field of poster advertising. The outcome has been a survey called ICSA (continuous study on outdoor advertising), which is not a specific survey, but a system of research. The ICSA study has the following objectives: - estimation of poster audience, by calculating coverage and frequency of "passages" past sites in Italian towns end cities - assessment of the effectiveness of posters as an advertising medium. To-date the ICSA study has involved surveys in 5 Italian towns and cities (with different geographical and demographic characteristics) to estimate coverage and frequency of passages past sites, and also surveys on the advertising impact of poster campaigns, conducted in 19 towns and cities of Italy.
Two methods of measuring television audiences are used in the Netherlands: 1. The electronic Peoplemeter, for which respondents have to press buttons to report their actual viewing behaviour and 2. The Scanner-method, which uses the day-after telephone interview to collect the data. In the following paper the results based on these two methods will be compared. The comparison will concentrate on the measuring of the reach-figures of commercials. We will discuss the results on total reach, cumulative reach and average reach. Although there are great differences in the figures, many of them can be explained by the different definitions used. After uniforming the definitions the remaining differences must be explained by the way of measuring itself. This clears up the fundamentals of both methods. If some effort is made to bring the definitions in line, both systems will be very well to use side by side of each other. The Peoplemeter produces elaborated television-data, the Scanner offers multi-media data.