The Peoplemeter, or TV meter, is a device which we use to find out who is watching what and when on television.
It's a small box which is connected to the television set and continuously sends data about which programme is on and who is present and watching theTV at that moment. Over 4000 people in the Czech Republic have a peoplemeter at home.
Why is the TV audience measured?
The Television audience is basically a benchmark of the success of television stations and it is measured in all developed countries around the world.
The information about audience is used by television stations in order to know how successful their shows were, and for planning when to broadcast which programme. Achieving high ratings justifies the purchase of broadcasting rights to more expensive films, shows or sport broadcasts. Audience is also fundamental for those who buy or sell television advertising.
Television advertising is the largest source of income for commercial television stations.
Money from advertising buys and produces television content, which is therefore free for viewers, who only "pay" their attention.
The Audience determines the price of advertising. As with newspapers and magazines, where the more readers the title has the more expensive the cost of advertising, television stations, as well as radio stations or Internet sites, offer their advertising space at a price corresponding to the success or popularity of broadcast programmes.
You might ask yourself the question why there is independent research of audience measurement, why television stations do not receive this information directly from television sets.
In fact, most television sets do not contain any device similar to the peoplemeter that could send the data about who watches what.
Actually, currently, this is not possible, because the diversity of different types of television equipment today would make it very expensive for the manufacturers to add such a meter and the entire research would cost much more than the organizations working with these data would be willing to pay. In addition, they would have to resolve how every single household member would indicate his or her presence. The solution could be our TV watching being monitored by some special visual equipment, but few people would welcome that.
The research in the Czech Republic is sponsored by the Association of Television Organisations (ATO), whose members share not only the costs of the project but also control over its accuracy, making sure no biases occur. The actual implementer of the research is Nielsen Admosphere, a leading Czech market research company, which has measured the TV audience in the Czech Republic since 2002. The company has successfully won several times against recognised international competitors in regularly announced tenders, based on the quality of its research and technology solutions. Nielsen Admosphere carries out the research in collaboration with a research agency, STEM/MARK and the technology company, Elvia.
Is it retro or in? (How has the measurement developed and how is it developing?)
You may not be aware but the information about who is watching what on television has been monitored since the fifties in the USA and UK, and radio audience research has been going on since the thirties. In our country television audience research started in the mid sixties, very shortly after the actual beginning of regular television broadcasting. Test broadcasting in our country started in 1953, but it was only running three days a week. Daily broadcasts started four years later, in 1957, and colour content did not appear on our television until 1973.
Like television broadcasting, audience research is constantly evolving. In the beginning there were phone calls to households, when people were asked about what they watched on the TV. Shortly afterwards came the so called diary research, using standard questionnaires, in which the viewers themselves recorded what and how long they watched. Electronic measurement using TV meters was introduced in the Czech Republic in 1997. The research thus accelerated and became more accurate. Electronic measurement itself must continuously respond to the changes in TV technology, such as the variety of television sets and types of broadcasting reception, TV on the Internet etc... You can learn more about how peoplemeters work later on.
The so called 'single currency' is often mentioned in the context of television audience. What does it mean? Television audience data are used by many companies and institutions, which need the results for their daily work and make many important decisions on its basis. They have all the data available from a single source thanks to the existence of one independent measurement. The results are thus mutually comparable, even from year to year. Confidence in the accuracy of the results is guaranteed by the strict supervision of the Association, which orders the research and has the greatest interest in ensuring the accuracy of the measurement. The project itself is carried out by an agency independent of the data users, which adds to the credibility of the audience measurement results.
Watching TV on the Internet is becoming more and more popular. In comparison to conventional TV it still accounts only for a small portion of TV reception, but the measurement of its audience cannot go unnoticed. Since 2018, a new digital part (PEM D) has been included in the project of TV audience measurement, which, via site-centric technology, secures the measurement of TV and other video content (of connected media) distributed over the Internet to all digital devices (PC, tablets, smartphones, Smart TV via HbbTV).
Who has a peoplemeter at home?
Only a selected sample of households has a peoplemeter at home. They represent the composition and behaviour of all households in the population that are equipped with at least one TV set. The list of people who participate in the measurement is not public, both to protect their privacy and to prevent any influence upon them.
Participation in the research is voluntary of course, but people cannot enlist in it. This prevents people who would like to participate in the research in order to influence the measurement results, to only test the TV meter or otherwise bias the resulting data. The selection of households which are contacted and asked to participate in the measurement is therefore independent and follows exactly predefined and recognised statistical procedures.
Do you also think that you don’t know anyone who has got a peoplemeter at home?
Approximately one in every nine hundred Czech inhabitants has had a TV meter in his or her household over the last ten years. The probability that you know someone like that is therefore very small. The probability that you will have a conversation about TV meters is even smaller, especially because the information that the household is participating in the project is confidential. This is also important as the members of the measured household could be tempted by someone to interfere with the TV meter or otherwise affect the audience results.
How are households selected for the measurement project?
The structure of the peoplemeter panel of the households participating in the measurement project must exactly match the composition of Czech television viewers. The sample must be representative of the television population in demographic, geographic and socio-economic terms as well as in terms of television behaviour and reception of the TV signal.
Not all of the information needed for the peoplemeter research can be obtained from the Czech Statistical Office (CSO). For this reason, an important part of the Project is the extensive Continuous survey, which maps the structure and behaviour of the television population. The peoplemeter panel is built and updated on the basis of these data and the CSO data. Roughly twenty thousand people come across this support survey, annually.
The same survey also determines the willingness of households to participate in the peoplemeter research. With time such households can be asked to cooperate in the measurement. A quarter of the panel is replaced annually in this way.
Our interviewers visit the households on the basis of random selection. They do not do this selection themselves though – the selection of addresses for questioning in the Continuous survey is done by a statistician by randomized but systematic process, defined in detail in the methodology of the entire Project. Addressing the households with a request for participation is therefore independent of the preferences of the interviewer or the clients of the project data.
Why don’t we all need to "measure"?
The sample of households, the panel, must fully represent all TV viewers. It has the same structure and the same television behaviour as the television population; this for example means that there is adequate representation of women and men, all ages, weak and strong viewers and so on. Only then is the generalization of the measured results onto all television viewers in the Czech Republic possible.
The audience research in the Czech Republic involves more than 4,470 people from 1,850 households. This is a sufficient number in terms of international standards, in fact more than enough, considering the total size of population of The Czech Republic..
The amount of time spent by households in participation in the measurement varies. Some are measured for a short time, others for several years. The development of population characteristics and the technological equipment of the households is examined by the Continuous survey, which provides adjustment of the composition of the peoplemeter panel in order to respond to the changes occurring naturally over the long term.
How to live with the peoplemeter?
The members of the peoplemeter panel have a peoplemeter connected to every television set in their home. This device sends data about which programme is currently on TV and who is watching.
Information about the viewed television channels is collected automatically, the members of peoplemeter households only provide information about their presence by using the remote control. It is also possible to log on a guest who is watching a programme at the monitored household, or to notify a long absence - holidays and so on. The display then shows what the members "say" to the TV meter with the remote control.
How is the accuracy of the households’ behaviour ensured?
It is very important that the viewers with peopplemeters are accurate about their presence in front of the TV. This activity is carefully checked during the data processing and one of the main tasks of the call-centre operators is to verify any unusual behaviour in the measured households. Participation in the research also starts with training and a trial period for the household, so people have time to get used to the login and logout process and get accustomed to audience measurement in their home.
Members of the peoplemeter households can affect the composition of the TV broadcasting by their selection of TV channels. They represent 9 million Czech TV viewers of all ages except children younger than three years, and so determine which TV channel´s programme interested the most viewers. The households gain points for this cooperation and with those they can choose gifts from a catalogue of rewards. The catalogue changes every year according to preferences of the households and includes both practical gifts and those that are mainly aimed to please.
What does the peoplemeter look like and what can it do?
The Peoplemeter is a small black box containing electronics which secures collection and transmission of the audience data. Part of the peoplemeter set includes a display and a remote control.
Peoplemeters measure the audience of all channels in the households in whichever way (antenna, satellite, cable, IPTV, Internet ...) the TV signal (analogue or digital) is received. The exact time of beginning and end of the viewing activity is recorded, including the use of teletext, video or DVD player or use of any other equipment connected to the TV.
Online broadcasting connected to the TV screen is measured using the peoplemeters as with any other broadcasting. Other formats of its display are monitored by software installed on the computers of peoplemeter households.
Information about the watched programme and its viewers is sent by the peoplemeters from the households automatically via a GSM network, in the same way as text messages from mobile phones. The data are stored in the measurement implementer’s database (Nielsen Admosphere). The data are verified, cleaned and weighed every day (according to predetermined methods) and prepared for use by the research users.
Frequent questions include what happens to the audience data if there is a network or power failure. In case of a GSM network failure, the data are stored in the peoplemeter memory and wait there until the network works again. They can thus reach the database later, but they are not lost because of network failure. The Nielsen Admosphere database also uses a backup storage which, in case of power failure in the central office, continues to receive data and ensures continuity of measurement.
The peoplemeters used in the Czech Republic, the Nielsen Admosphere TV meter and the newer Nielsen Admosphere SimEar, are the result of the work of Czech developers and innovators. Their task is to respond to changes in television technology and continue to facilitate and make the "coexistence" of measured households with the peoplemeters easier.
What is the result of the measurement?
The result of peoplemeter measurement is the information about the audience of individual television channels and programmes, what type of viewer watched which and when. The success of television channels and their programmes can be determined on the basis of the audience measurement results. The results can be seen from different angles. The total number of viewers is monitored as well as the attractiveness of individual programmes for different audience groups (women, men, older, younger, by regions, etc.).
The exclusive right to the audience measurement data belongs to the previously mentioned Association of Television Organisations (ATO). A list of its members is on the website of ATO (www.ato.cz/en). The ATO regularly publishes some of the results, such as average weekly and monthly audiences; see www.ato.cz/en. Any other data can be provided to the non-members only with the consent of the Association. This is because the television audience research is among the most extensive research projects carried out in the Czech Republic and its actual implementation is quite expensive. Most of the results are not therefore freely accessible and are available only to members and business partners of the ATO. The results contain no personal information about the peoplemeter panel members.
Preliminary data are available online to the users a few minutes after the broadcast. Complete data are always ready within three hours after the end of daily broadcasting, at 9 o'clock in the morning because the broadcast day lasts 24 hours, from 6 am to 6 am the next day.
The software available to the users of the peoplemeter data allows analysis of the data from different angles. One of the best known outputs is the so-called pie, or pie chart. These show us how the television channels did in the TV population (the share) and how many people from the target group watched the programme, the rating.
Example of data output from peoplemeter research ATO - Nielsen Admosphere, June 2015
Example of data output from peoplemeter research ATO - Nielsen Admosphere, March 2015
"Share" pie This chart usually indicates how many of the people who were watching television at a specific time were watching a particular programme.
This particular chart shows that in June 2015, of all television viewers older than 15 years, most people watched TV Nova, 23.60%, followed by CT1, 14.35%, and Prima, 13.34%.
The results can be made more specific, for example to identify a narrower time period (e.g. prime time) or audience group (e.g. men 35 to 50 years). These specific results can then show the different tastes of different audience groups.
„Rating“ The rating gives another view of the audience results. It is usually expressed as a percentage or in thousands of viewers. It tells us how many people from a particular target group watched a particular programme. If someone watched only a part of the programme, for example the first 20 minutes, his or her viewing is included in this percentage only partially - proportionally.
The following table shows examples of the ratings of the most watched series in the third week of March 2015. The most successful episode is displayed for each of the three major TV channels.
We can see in the table that the most watched series on TV Nova, in that period, was Policie Modrava, with a rating of 24%. One fourth of adult viewers (15+) watched this episode in the third week of March, roughly 2.1 million Czech viewers. The most watched series on CT1 and Prima that week were Vraždy v kruhu (CT1) and Cesty domů III (Prima). Vraždy v kruhu attracted 14% of Czech television viewers older than 15 years, more than 1,2 million people, Cesty domů III was watched by 7%, 600 thousand adult Czech viewers.
Accurate research requires a sufficiently large and representative panel. The Panel size is adequate to the size of the whole population. In the Czech Republic this ratio is even higher than is usual in this type of research in other countries. The data also undergo daily verification, so even if you have a different sense of the audience and you are wondering about the measured results, you should be reassured that the results of peoplemeter households actually reflect what the television population watches.