Data fusion is today becoming a more and more widely used technique : when confronted for instance with the problem of dealing with media/product type of data, it is clear that "marrying together information from different surveys" [1] is both cheaper and less demanding for the respondents to each of the two surveys. Therefore, a lot of efforts have been spent by statisticians to develop various fusion techniques, allowing to "transfer information from a donor to a recipient" in some way, these procedures being mainly adapted from "the well established set of statistical procedures for dealing with missing data" [2]. However, almost all authors start saying that "the most desirable way (...) would be to analyse a single source data base" [1], This is the reason why it is first necessary to review the real value of the statistical fusion process by itself : the criteria presently used by practitioners may be seen more as empirical evaluation rules than as TRUE statistical tests. A complete evaluation of the fusion process has to be based on a clear understanding of what exactly is expected from the "synthetic file" obtained after applying a matching procedure between two initial different files.
The effectiveness of an advertising message depends as much on its intrinsic virtues - especially the creative work behind it - , as it does on the context in which it appears. This is why in the print media, the impact of an advertisement can drop by almost 50 % according to the publicationâs Reading Mood 1 - In radio advertising, the ongoing trend is to saturate certain stations while taking little or no account of the programs they broadcast, and to use indifferently general entertainment (Europe 1 RTL, etc.) and music stations (NRJ, Skyrock, etc.) despite the significant differences between the two. Our intention is to show that : - radio has its own listening moods related to both the station and the program being broadcast, and even to the time of broadcast, - and that these listening moods have a very significant influence on the perception of radio messages.
Segmentation is the basis for marketing activities, including media splanning. In the sixties and seventies social class and family life cycles were popular bases for marketing. From the end of the seventies till the end of the eighties life-style dominated this kind of research. But at one side the idea grew that life-style characteristics were measured at a very general level and could therefore not predict consumer behaviour. At the other side the consumer seems to show more variety. Domain-specific market segmentation seems to be a solution for these problems. Consumind is an example of this approach. In this paper the project is described and applications are presented.
The Broadcasting Standards Council's remit is to examine and advise on issues of taste and decency in broadcasting. Each year, it focusses on a specific area and undertakes detailed research into it. In 1991 the issue was that of bad language: swearing and terms of abuse. The research aimed to understand the limits of acceptability of swearing; both in terms of situations where swearing is considered more or less acceptable and in terms of the types of words that can be used under differing circumstances. The approach selected involved using sorting and reporting techniques which, while involving strong and potentially offensive language, did not cause embarrassment or offense to interviewees themselves. At the same time, since the tasks used offered individuals the chance to set their own agenda of bad language, the research did not impose pre-existing definitions of acceptability and unacceptability upon the sample. The research also identified a typology of objectors and non-objectors. In a second stage of the research, using mixed quantitative and qualitative techniques, the contexts in which swearing is accepted was studied. For the Council, the research helps to indicate where the boundaries of acceptability of swearing lie, and to identify those members of the public most likely to be offended by its use in broadcasting.
In the autumn of 1991 we conducted, through matched area samples and covering both weekly magazines and television, the biggest ever controlled advertising experiment in Sweden. Using 2,6 telephone interviews and 1,3 home interviews we evaluated the effectiveness of 6 actual campaigns. The conclusions we reached were as follows: - The creative content is decisive for the effects in weeklies and on TV - Synergistic effects of weeklies and TV are foremost depending on the creative content too - Through advertising in weekly magazines and TV you will reach a larger part of the target group than through weeklies alone or TV alone - The weekly advertisements have given obvious contributions to the TV advertisements - The TV advertisements have not given that obvious contributions to the weekly advertisements - Synergistic effects through weekly magazines and TV are often to be found in several different communication areas. E.g. message transmission, learning and image are more strengthened through TV and weeklies than through weeklies alone or TV alone - Synergistic effects through weeklies and TV have been indicated by increase in buying intentions as well as a higher purchase-rate - Effects from TV-advertising grow fast, but also decrease rather fast over time - Effects from advertisements in weekly magazines don't grow that fast. Responses and attitudes first grow and than decrease over time but actual buying doesn't seem to decrease as fast over time as other effects or as fast as the effects from TV advertisements.
Starting in January 1991 the Dutch single-source multi-media survey SummoScanner has been enhanced with specific TV-viewing data. Within the limits of the telephone interview, it is now recorded which programmes were viewed on the day before the interview. This implies a questioning technique based on recall, as opposed to the people-meter technique applied in AGB-panels (In the Netherlands: Kijk- en Luisteronderzoek KLO). SummoScanner data are not intended to replace AGB-data as the primary' media-planning tool, but to enhance insight into the particular viewing habits of target groups, in connection with other media data. Together with the data provided by SummoScanner and the postal follow-up - the Target Group Survey - the TV-data add to the value of this single-source survey. In February 1992 the first full year's report on the results of these measurements was published. On the whole, the figures, although not completely comparable, seem to be of the same order as the AGB- figures. This underscores the fact that telephone interviewing has little or no bias resulting from loss of memory, provided that the previous day's behaviour is asked for. The business of making these data available for SUMMO-members is still under development - data sets are larger than ever: 32 records per year, each containing 720 variables concerning yesterday's viewing alone. Still, this new database is bound to provide media planners with new means for adding to their insight into the media behaviour of target groups.
The significance of the media environment with respect to the effectiveness of advertising can be considered as one of the most arbitrary subjects in the field of media research and media planning. Systematical research in this field, however, has hardly been undertaken. Bv means of a research survey, the (Dutch) publishing company Weekbladpers wanted to investigate whether adaption to the media environment (Umfeld) would generate increased effectiveness of advertising. If this proves to be true, it might result in media planners and creatives taking into account (with the aid of available information from comparative analyses on that subject) the influence of the contents and structure of the various media on the effectiveness of advertising. The desire to stimulate the discussion on this subject was the primary consideration which has led the Weekbladpers to initiate this research project. The research agency r+m, Research and Marketing, in Heerlen, was commissioned to set up and carry out a survey.