At the core of AGF's audience research is the AGF Panel, which AGF operates in cooperation with GfK and which is continuously developed from a TV to a Video Panel. AGF Videoforschung has systematically enlarged the Panel in several expansion stages. While the Panel’s size was only about 2,700 households in the 1980s, it has now doubled to at least 5,400 households that are home to roughly 11,000 individuals.



Since the beginning of 2016 the universe represented by the AGF system has been “the resident population of the Federal Republic of Germany in private households with at least one TV set in use and a German-speaking Main Income Earner.”

This universe represents the television usage of approximately 38.753 million TV households with up to 75.294 million individuals over age 3 (as of January 1, 2021).

Panel Survey

Once a year, panel households are surveyed with a questionnaire (PAPI or online) covering a wide range of demographic and consumer behaviour items which serve for the large number of target groups available for the analyses. Additionally the information gained is unsed to align the daily weighting of the AGF Panel.


External specifications

The annually updated specifications of media analysis (ma) serve as a frame for selection of panel households. Since the ma survey can only include individuals who are 14 or older, the specifications for children from 3 to 13 are supplemented with information from the Micro Census. Structural information from the current ma radio tranche serves as external specifications for the panel, for instance the distribution of reception levels or household sizes.

Since adequate external specifications are not available from the ma for the “platform” characteristic (which describes pay-TV households), this information is determined in the representative Platform Study run by Kantar on behalf of AGF and updated twice a year.

The specifications for Sky households and the distribution of individual Sky subscriptions come from a Deloitte-audited audit and have been used as a specification for the AGF system since January 1, 2019. The specifications are updated twice a year (January 1 and July 1), analogous to those for platform households.


The basis for the sample of panel households is the ADM Sampling System, which was updated in 2011 by the association of German market and social research institutions (Arbeitskreis deutscher Markt- und Sozialforschungsinstitute, ADM) and consists of approximately 53,000 sampling points. These are small territorial units consisting of small municipalities, urban districts, or road sections covering the entire inhabited territory of the Federal Republic of Germany. By using this sample system every household is given a chance to be recruited for the panel.



TV panel-households are recruited in a hybrid process, using different methods in parallel: the random-route method and the quota method. In the random-route method, households are selected using a defined random process and are adressed for a recruiting interview for the TV panel. The quota method is based on a targeted selection of individuals or households. The procedure is intended to achieve a representative composition of the sample by setting quotas on certain characteristics. The distribution of the characteristics in the universe must be known. In the end it is the households needed to satisfy the external specifications that are included in the panel.


Panel households are distributed throughout Germany, with at least 218 representative households being available in each federal state. In this way, AGF ensures – even for smaller federal states – that there are enough households to also monitor local television providers.

The resulting disproportionate distribution of households among the states is rebalanced by the daily weighting.

The weighting is performed in two steps:

  • Preliminary weighting is performed daily once GfK has transmitted and verified the data retrieved overnight. The TV ratings for the previous day’s television usage are created on this basis.
  • For the final weighting, after three days the additional households retrieved during this period are also included and the preliminary weighting factors are overwritten with the final factors. Time-shifted usage that occurred in the three intervening days is also included at this point.

The weighting process ensures precise extrapolation to all TV and video households and all individuals age 3 or over living in them throughout Germany.

Reception levels

Four reception levels for TV which also are relevant for  recruiting and weighting are distinguished:

  • IPTV: Programming received via IPTV.
  • Satellite: Programming received via a satellite dish or own satellite receiver.
  • Cable: Programming received through a cable connection.
  • Terrestrial: Household able to receive DVB-T.

Satellite reception currently has the largest share.

Development of TV reception levels (households in %)

© Source: AGF Videoforschung in cooperation with GfK; VIDEOSCOPE 1.4, Haushalte, 01/01/2021, Market standard: TV; 05/01/2021