The print reach study Mach Basic is slowly reaching its limits: In a few years, it will no longer be representative enough. In recent years, the general conditions for print media research have changed so rapidly that the AG für Werbemedienforschung (Wemf) has already been thinking about the third Mach generation for two years. Unlike the last generation change, however, this time it is not primarily a matter of improving the currency. "The LpA value is still absolutely state of the art, even in international terms," says Wemf research director Harald Amschler. The goal, however, is to maintain the current currency quality. Because this is at risk.
Amschler cites two main reasons. Since it is no longer necessary to have one's own telephone number entered in the phone book, more and more landline and even more so cell phone numbers are missing. People who do not have a landline phone and only have a cell phone can no longer be contacted in this way. This group of people is "still manageable," says Amschler. But the number is growing and will sooner or later call into question the representativeness of the Mach sample. The second reason for a threatening decline in quality is that there are more and more titles with similar names in the Swiss press. "Sonntag," for example, is the name of the seventh edition of MittellandZeitung, but also a weekly magazine published by CAT-Medien. Other examples: NZZ and NZZ am Sonntag, Schweizer Illustrierte and SI Style, 20 Minuten and 20 Minuten Week, Blick, SonntagsBlick and Blick am Abend. For these titles, this increases the risk of confusion when queried purely by telephone - and thus leads to incorrect information. The same applies to brand changes such as from TR7 to TVStar or from 20 Minuten Week to 20 Minuten Friday.
Quality problems are eliminatedIn the meantime, Wemf has already evaluated a number of new survey methods, including electronic measurement systems. But Amschler and the Wemf Board of Directors came to the conclusion that the technology is either unsuitable or still too immature for electronic measurement of print usage (see box). The consequence: Wemf is sticking with surveys in terms of methodology, as, incidentally, the research organizations in our neighboring countries also do. However, Wemf modifies them in such a way that the quality problems mentioned above are eliminated. Specifically, it has opted for a multi-stage procedure ("two-part CATI with RDD"). The numbers are no longer searched for from the (incomplete) telephone directories with the aid of a computer, as is the case today, but are compiled by a random number generator (RDD) (including those with mobile prefixes 079, 078 ...) and dialed automatically.
Once the number is in operation, the person in question can be asked to participate in the study. If the person is willing to participate, the basic data (age, gender, place of residence, etc.) is collected in an initial short telephone interview and the person is informed that he or she will be sent a set of sample materials by mail. Among other things, the print titles to be surveyed are shown with their logo. At the same time, an appointment is made with the person for the main interview (inquiry about the titles read). The interview is also conducted by telephone, with the interviewee using the display material to avoid confusing titles.
Essentially, then, Wemf wants to switch from a purely verbal to a combined supported survey (verbal and visual), similar to what it already does today with the MA Leader. At the same time, it wants to guarantee that all persons who can be reached by telephone are again included in the initial sample.
It is not yet clear whether the survey content will change along with the methodology. However, Amschler assumes that reach will continue to be surveyed using the LpA approach (for newspapers, for example, "read yesterday"). "Other additional questions will probably be added - for example, on cross-media use of channels and content," he says. But Amschler warns against overloading the Mach3 Basic questionnaire. "It's already difficult to get people to participate. For additional needs, there is the possibility to conduct parallel surveys in the Mach system," he says.
More expensive than today's studyNext year, Wemf plans to conduct an initial Mach3 pilot study in parallel with the current survey method. It will not only show whether the new methodology works, but also how much the results differ from those of the current survey. In any case, Mach3 will be more expensive than the current study.
If everything goes according to plan, the Mach3 Basic will be fully surveyed for the first time as early as 2010. In this case, Wemf could publish the results in 2011. But Amschler emphasizes: "Thanks to our proactive approach, we have enough time to carefully develop and test the Mach3. In the meantime, however, the Mach2 continues to do its job reliably."
Markus Knöpfli
With Mediawatch, MediaScan, Arbitron, Eurisko or Handycameras, there are several approaches for electronic measurement of print usage, but the active cooperation of the test person is always required: He or she must speak into the device, enter what he or she is reading via a key, take a photo or scan it - and for how long. This requires a lot of discipline. And according to Harald Amschler, it hardly leads to more reliable results than the memory on which the Mach interviews are based.
At present, only one method measures passively, i.e. without the conscious intervention of the test person. The print titles to be examined are fitted with an RFID tag, a flat chip similar to those used today to prevent theft under labels. The measuring device worn by the test person then uses emitted and reflected radio waves to detect which medium was used when and for how long. The problem here: For the purposes of media research, this technology is only in its infancy. It is therefore still very expensive and, above all, prone to error. For example, the radio waves are reflected by other metal objects, which leads to false signals. (mk)
Mach generationsWhen the terms Mach1, Mach2 and Mach3 are used more frequently in Swiss media research in the future, we are not talking about the speed of sound. Mach stands for Media-Analyse Schweiz (CH) and the numbers are used to designate the study generations. We are currently in the age of Mach2, which since 2000 has provided the official reach currency for print and cinema with the LpA value (readers per copy). This is based on 23,500 telephone interviews of around 35 minutes each on the use of print and cinema, conducted per year and spread over 365 days (the so-called CATI method). Incidentally, the first Mach generation with the fuzzier K1 value was on the market from 1991 to 1999. Before that, there was the "MS" media study from 1965 to 1989. (mk)