Experimental statistics Pendler Mobil: the use of mobile network data to support official commuter statistics


18 July 2022 - The availability of timely and small-area data on commuting patterns is of great relevance for political decision-making, down to municipal level. Commuter behaviour enables conclusions to be drawn, for example, on the regional distribution of the places of work and residence of the working population, which may contribute to a continuous improvement of transport infrastructure. The official commuter accounts provide the data required for this. However, there is potential to improve the accounts with regard to the spatial and temporal mapping of the commuting patterns of persons in employment and to expand the accounts to include information concerning those commuting for education purposes.

The official commuter accounts provide regional information on commuting patterns only at municipality level, and the publication is delayed because of the data required for this. On account of the evident potential for improving the official commuter accounts, the project Pendler Mobil was conducted in cooperation with IT.NRW (the central statistical and IT service provider of Nordrhein-Westfalen) to identify areas where mobile network data can complement the existing commuter accounts. Mobile network data are a promising source of relevant data. Likewise, promising questions concerning population mobility can be addressed because of the high penetration rate of mobile devices in the population (see Continuous household budget surveys) and the accurate information on the spatial and temporal distribution of mobile users. Using origin-destination matrices, the project examined whether mobile network data can be used to map commuter flows in Nordrhein-Westfalen during the day. Another aim of the study was to establish whether, and to what extent, the current commuter accounts can be complemented based on small-area mobile network data. The IT.NRW commuter accounts were used for comparison in order to verify the representativity of the data.

The result was that the mobile network data available have the potential to support the official commuter accounts by providing small-area data on urban commuter movements in the form of enhanced information concerning commuter destinations. This provides new insight into labour market regions and enables highly frequented urban areas to be identified. The project found, however, that the mobile network data available and the IT.NRW commuter accounts still differ considerably as regards the numbers of commuters and the mapping of commuter flows in Nordrhein-Westfalen. These differences have a strong impact in particular at the regional level.

Possible reasons for underestimating or overestimating commuter flows are, among other things, the filter criteria and the processing procedures that are used to extract information only on employment commuters from the mobile network data. There is a conflict between the accurate delineation of the movement patterns of specific target groups and the information content of the mobile network data available. Especially the underestimation of commuter flows of mobile users as compared to the official commuter accounts suggests that approaches for modifying the mobile network data should be discussed.

A reformulation of the filter criteria, on the one hand, or the integration of a geographical selection for extracting specific target groups, on the other, are possible approaches to improving the results or to obtaining more targeted results. In addition, a methodological change is required or the extrapolation carried out by the data providers by subgroup needs to be enhanced in order to calibrate the values from the mobile network data according to selected target groups. For further information and detailed results please refer to Hadam (2021).


Hadam, S. (2021) Pendler Mobil: Die Verwendung von Mobilfunkdaten zur Unterstützung der amtlichen Pendlerstatistik. AStA Wirtschafts- und Sozialstatistisches Archiv 15, 197–235 DOI reference