German Mobility Panel Research Articles

The German Mobility Panel: – Lessons Learned from a Longitudinal Travel Behavior Survey over 30 Years

After 30 years, the German Mobility Panel ceased data collection in the summer of 2023, despite being considered a successful and benchmark-worthy survey about everyday travel. This article gives an overview of the survey’s central ideas, purpose, and design, thoughts about the usefulness and applicability of the data, and explains why the survey has been terminated. It ends up with the experiences made with the MOP. Lastly, a synthesis based on the lessons learned is used to derive recommendations that can help implement future longitudinal surveys.

A case study of checking national household travel survey data with machine learning

In recent years, machine learning techniques have been increasingly tested and applied to physically collected data to optimize the processes. In this paper, machine learning is used to check travel survey data of the German Mobility Panel (MOP). In the MOP, verified and raw data have been available for several decades, on which algorithms can learn the practices of human checking routines. By using machine learning, the algorithm is expected to learn the checking patterns from the past and thus support the data checking of new datasets. To this aim, several algorithms are applied and tested. The presented model framework supports the identification of blatant deficits in the reports at the individual and trip levels. The neural network (NN) shows the most promising results as it decreases the number of data samples checked. The checking effort can be reduced by 20.4 % at the individual trip level. This work shows that machine learning can support the data checking process in the MOP at various levels, thus leading to significant time reduction.

Methodological Challenges for Measuring Behavioral Changes in a Longitudinal Travel Survey Under Pandemic Conditions

The outbreak of the COVID-19 pandemic has had an enormous impact on everyday life. In an attempt to stop the virus spread, unprecedented containment measures are being taken worldwide. These preventive measures are consequently affecting everyone's life and how we move. Equally remarkable is the speed at which these observed and massive changes occur and the lack of statistical evidence. In the annual cycle, the German Mobility Panel (MOP) launched an additional questionnaire for the general population, containing approximately 4,000 respondents. This paper provides insights into the MOP data quality issues and changes in daily travel under pandemic conditions.

How Late Reporters Effect Data Quality in Longitudinal Surveys – Experiences From the German Mobility Panel

Survey design always has a significant influence on the outcomes. Therefore, this paper investigates how follow-up campaigns affect survey outcomes and response rates in longitudinal surveys. Furthermore, it is assessed how late reports in the fall affect survey outcomes. The analyses are based on the unique data of the German Mobility Panel. Overall, this paper paints a broad picture of the methodological aspects and overlapping effects that should be considered before starting the fieldwork of longitudinal surveys. The results indicate that people who are reminded to participate positively influence the survey outcomes - even when the report is belated.

Checking Data Quality of Longitudinal Household Travel Survey Data

Ensuring data quality of household travel survey data is often tedious and, thus, time-consuming. To speed up the process of data-checking and to gain an in-depth understanding of the data, data visualization is a practical, fundamental tool. Since 1994, data visualization has been used in the German Mobility Panel (MOP) data-checking process. This paper presents two graphical visualization tools developed for the MOP. Both tools speed up the data checks and ensure high consistency in identifying erroneous data. This paper describes and discusses how the tools provide a continuous data quality assessment.

State-of-the-art of Longitudinal Travel Surveys – A Comparison of the MOP and MPN

Longitudinal travel surveys are needed to capture individual travel behaviour changes. Only two longitudinal tavel surveys of national relevance are currently in operation, the German Mobility Panel (MOP) since 1994 and the Netherlands Mobility Panel (MPN) since 2013. This paper provides an overview of both panels' differences and similarities in design and data collection. Furthermore, representativeness, diary fatigue and non-random attrition are assessed in both panels to show the challenges panel surveys have to deal with. Overall, this paper shows important aspects of a panel survey that should be considered when designing a new longitudinal travel survey.

Private Vehicles Greenhouse Gas Emission Estimation at Street Level for Berlin Based on Open Data

As one of the major greenhouse gas (GHG) emitters that has not seen significant emission reductions in the previous decades, the transportation sector requires special attention from policymakers. Policy decisions, thereby need to be supported by traffic emission assessments. Estimations of traffic emissions often rely on huge amounts of actual traffic data whose availability is limited, hampering the transferability of the estimation approaches in time and space. Here, we propose a high-resolution estimation of traffic emissions, which is based entirely on open data, such as the road network and points of interest derived from OpenStreetMap (OSM). We estimated the annual average daily GHG emissions from individual motor traffic for the OSM road network in Berlin by combining the estimated Annual Average Daily Traffic Volume (AADTV) with respective emission factors. The AADTV was calculated by simulating car trips with the open routing engine Openrouteservice, weighted by activity functions based on statistics of the German Mobility Panel. Our estimated total annual GHG emissions were 7.3 million t CO2 equivalent. The highest emissions were estimated for the motorways and major roads connecting the city center with the outskirts. The application of the approach to Berlin showed that the method could reflect the traffic pattern. As the input data is freely available, the approach can be applied to other study areas within Germany with little additional effort.

How the COVID-19 pandemic changes daily commuting routines – Insights from the German Mobility Panel

The COVID-19 pandemic has had a major impact on everyday travel and, by extension, everyday commuting. During the pandemic, some people were able to work from home while others continued commuting. This study examines how commuting behavior changed between 2019 and 2020. In this study, we analyze panel data of the German Mobility Panel, a national household travel survey. We paint a broad picture of the characteristics and behavior of those who commuted during the pandemic. The analyses focus on the intra- and interpersonal differences and are presented in a mostly descriptive way. The results show that people with low income and a low level of education are primarily those who cannot work from home and do not have flexible working hours. The results further show that especially public transport has lost importance in daily commuting. However, those who commuted in 2019 and 2020 did not significantly change their commuting behavior regarding commuting time and commuting mode.

Drive less, drive better, or both? Behavioral adjustments to fuel price changes in Germany

The demand for motor fuel should decline when its price rises, but how exactly does that happen? Do people drive less, do they drive more carefully to conserve fuel, or do they do both? To answer these questions, we use data from the German Mobility Panel from 2004 to 2019, taking advantage of the fluctuations in motor fuel prices over time and across locales to see how they affect Vehicle Kilometers Traveled (VKT) and on-road fuel economy (expressed in kilometers per liter) for gasoline and diesel cars. Our reduced-form regressions show that while the VKTs driven by gasoline cars decrease when the price of gasoline rises, there is virtually no response among diesel cars. Likewise, the on-road fuel economy is largely unresponsive to fuel price changes, irrespective of the fuel type. Since the price elasticity of fuel consumption is the difference between the price elasticity of VKT and the price elasticity of the fuel economy, our results suggest that the fuel economy might be the “weakest link” of price-based policies that seek to address environmental externalities.

The unequal cycling boom in Germany

Over the last two decades, cycling in Germany has increased by more than 40%. This paper analyses how this overall increase is broken down by group, characterised by residence (rural and smaller towns vs. medium-sized and larger cities) and education (high vs. low). It analyses (1) how the composition of the population changes according to these groups, (2) how cycling behaviour develops within these groups, and (3) how the changes in composition and behaviour shape the overall volume of cycling. Data on mobility behaviour comes from the large-scale, representative German Mobility Panel from 1996 to 2018, and the analytical sample covers information on more than 28,000 persons over approximately 730,000 reported trips. Data on changes in population composition comes from the German Socio-Economic Panel. Results show that the increase in cycling is unbalanced and largely a consequence of highly educated people in cities who now cycle twice as much and whose share of the population has doubled. This reveals that the cycling boom is bypassing important parts of the population, which limits the contribution of cycling to sustainability goals. Furthermore, the uneven evolution of cycling amplifies social inequalities in finances and health. Finally, this paper shows that increased cycling comes not only from changing behaviour within groups, but also from altered population composition. The most impactful compositional shift is the increasing level of education, which will likely continue to boost cycling.

What is the role of carsharing toward a more sustainable transport behavior? Analysis of data from 80 major German cities

The supply of as well as the demand for carsharing has been increasing in recent years. Yet, these dynamics and carsharing's function as catalyzer toward a more sustainable transport behavior in German cities are still not fully understood. For this reason, we analyze supply data for 80 large cities in Germany regarding fleet numbers and supply densities. We combine this with analyses about carsharing (non-) members by drawing on the datasets Mobility in Germany and the German Mobility Panel. We find that carsharing members use walking, cycling, or public transport 1.4–1.5 times more often and motorized individual transport, respectively, less often than nonmembers. Based on this, we build user groups regarding the people's usage of the environmental alliance (i.e., walk, bicycle, public transport). In the subsequent multinomial logistic regression analysis for the years 2013, 2015, and 2017, we find that both increasing carsharing supply and being carsharing member are not effective measures to increase the share of walking, cycling, and public transport. The availability of mobility tools, especially the number of cars in households, are more important predictors.

Reducing vehicle cold start emissions through carbon pricing: evidence from Germany

A large proportion of local pollutants originating from the road transport sector is generated during the so-called cold-start phase of driving, that is, the first few minutes of driving after a car has stood inactive for several hours. Drawing on data from the German Mobility Panel (MOP), this paper analyzes the factors that affect the frequency of cold starts, approximated here by the number of car tours that a household takes over the course of a week. Based on fixed-effects panel estimations, we find a negative and statistically significant effect of fuel prices on the number of tours and, hence, cold starts. Using our estimates to explore the spatial implications arising from fuel price increases stipulated under Germany’s Climate Programme 2030, we find substantial impacts on the number of avoided tours even for modest fuel price increases of 20 cents per liter, particularly in urban areas. This outcome lends support to using carbon pricing as a means to improve both global climate and local air quality, pointing to a co-benefit of climate policy.

Drive Less, Drive Better, or Both? Behavioral Adjustments to Fuel Price Changes in Germany

The demand for motor fuel should decline when its price rises, but how exactly does that happen? Do people drive less, do they drive more carefully to conserve fuel, or do they do both? To answer these questions, we use data from the German Mobility Panel from 2004 to 2019, taking advantage of the fluctuations in motor fuel prices over time and across locales to see how they affect Vehicle Kilometers Traveled (VKT) and on-road fuel economy (expressed in kilometers per liter). Our reduced-form regressions show that while the VKTs driven by gasoline cars decrease when the price of gasoline rises, their fuel economy tends to get worse. It is unclear why this happens. Perhaps attempts to save on gasoline—cutting on solo driving, forgoing long trips on the highway, driving more in the city—end up compromising the fuel economy. By contrast, both the VKTs and the fuel economy of diesel cars appear to be insensitive to changes in the price of diesel. Latent class models confirm our main findings, including the fact that while fuel prices, car attributes, and household and location characteristics explain much of the variation in the VKTs, it remains difficult to capture the determinants of on-road fuel economy. Since the price elasticity of fuel consumption is the difference between the price elasticity of VKT and the price elasticity of the fuel economy, our results suggest that the fuel economy might be the “weakest link” of price-based policies that seek to address environmental externalities, such as a carbon tax.

Reducing Vehicle Cold Start Emissions through Carbon Pricing: Evidence from Germany

A large proportion of local pollutants originating from the road transport sector is generated during the so-called cold-start phase of driving, that is, the first few minutes of driving after a car has stood inactive for several hours. Drawing on data from the German Mobility Panel (MOP), this paper analyzes the factors that affect the frequency of cold starts, approximated here by the number of car tours that a household takes over the course of a week. Based on fixed-effects panel estimations, we find a negative and statistically significant effect of fuel prices on the number of tours and, hence, cold starts. Using our estimates to explore the spatial implications arising from fuel price increases stipulated under Germany’s Climate Programme 2030, we find substantial impacts on the number of avoided tours even for modest fuel price increases of 20 cents per liter, particularly in urban areas. This outcome lends support to using carbon pricing as a means to improve both global climate and local air quality, pointing to a co-benefit of climate policy.

Tracing the Effects of the Covid-19 Pandemic on Car Usage in Germany - an Analysis of the German Mobility Panel

Being a vast external influence, the COVID-19 pandemic causes major changes in travel behavior on the individual level. This exceptional situation and the political measures such as the lockdown lead to decreased overall travel demand and shifts in mode choice. To understand these effects, the analysis of car usage in private households offers explanatory insights. To this aim, this study provides a differentiation of car usage before and during the COVID -19 pandemic in spring 2020. Furthermore, insights into structural changes of car usage in private households based on German Mobility Panel data are presented. The results show that during the COVID -19 pandemic car usage was affected, and overall mileage decreased. Especially newer cars with large cubic capacity show a large decrease in mileage by 39 % between 2019 and 2020. In addition, we see that the type of car usage (business/private) and the household characteristics are related to different magnitudes of change in car usage. Overall, it becomes evident that examining identical households and cars before and during the pandemic allows for a deeper understanding of the impacts of the COVID -19 pandemic.

Image-based activity pattern segmentation using longitudinal data of the German Mobility Panel

In this paper, we present an approach to segment people based on a visualization of the longitudinal week activity data from the German Mobility Panel. In order to perform segmentations, different clustering methods are commonly used. Most of the approaches require comprehensive prior knowledge about the input data, e.g., condensing information to cluster-forming variables. As this may influence the method itself, we used images with a high degree of freedom. These images show week activity schedules of people, including all trips and activities with their purposes, modes as well as their duration or their temporal position within the week. Thus, we answer the question whether using only this type of image data as input will produce reasonable clustering results as well. For the clustering, we extracted the images from an existing tool, processed them for the method and finally used them again to select the final cluster solution based on the visual impression of cluster assignments. Our results are meaningful as we identified seven activity patterns (clusters) using this visual validation. The approach is confirmed by the data-based analysis of the cluster solution showing also interpretable key figures for all patterns. Thus, we show an approach taking into account many aspects of travel behavior as an input to clustering, while ensuring the interpretability of solutions. Usually, key figures from the data are used for validation, but this practice may obscure some aspects of the longitudinal data, which are visible when looking on the images as validation.

Two methods of estimating long-distance driving to understand range restrictions on EV use

The distances travelled by individual cars vary strongly from day to day. This is problematic for electric vehicles since they cannot be used for journeys longer than the all-electric range. At present, long-distance travel is rarely covered by household travel surveys and little is known about the frequency of long-distance car travel on an individual car basis. Here, two methods are compared that estimate the number of days per year with a driving distance larger than a given threshold: a simulation method and a probabilistic method. The simulation method combines a national household travel survey, a car-use survey and a long-distance travel survey and simulates one year of car driving; the resolution of results is car trips over a full year. The probabilistic method uses statistical distribution to estimate the number of days per year with long-distance travel. Both methods are tested on a representative one-week sample of the German Mobility Panel with travel data from over 6000 cars. Our results show that both methods produce similar aggregated results for the distribution of days with long-distance car travel, the maximum mileages, and average long-distance travel frequencies amongst user groups. However, the two methods differ on the level of individual cars. Our findings indicate that long-distance travel behaviour can be estimated on an aggregated level without long observation period data. Both methods can be directly applied to the limited range of electric vehicles and the need for adaptation or fast charging.

X vs. Y: an analysis of intergenerational differences in transport mode use among young adults

Recent research has contrasted the travel patterns of young adults of Generation Y (or, synonymously, the Millennial Generation) with the travel patterns of earlier generations of young adults such as those belonging to Generation X. Young adults of Generation Y are found to drive less and in some contexts are found to exhibit more multimodal travel patterns and to use public transit more often. Potential causes for these observed shifts in transport mode use have also been theorised: One view is that period effects in the form of contemporaneous changes in socio-cultural, socio-economic and socio-technical factors are responsible for the observed shifts in transport mode use; another view is that members of Generation Y have inherently different preferences and values due to formative socio-cultural, socio-economic and historical experiences. Motivated by this yet-to-be-resolved dialectic, this paper uses a hierarchical Bayesian multivariate Poisson log-normal model to examine intergenerational differences in transport mode use among young adults. The model is applied to 23 waves of the German Mobility Panel and captures between-cohort and between-period variation of parameters of interest. The trained model informs a counterfactual prediction exercise aiming to decompose intergenerational differences in transport mode use into demography-, cohort-, and period-specific effects. Our findings suggest that all three sets of effects account for intergenerational differences in transport mode use, while the absolute and relative importance of each set of effects vary across transport modes. For the period from 1998 to 2016, two thirds of the decline in car use can be ascribed to period effects; nearly all of the increase in public transit use and 42% of the increase in bicycling can be ascribed to cohort effects.

Are Activity Patterns Stable or Variable? Analysis of Three-Year Panel Data

Routines and mandatory activities, such as work and school, shape the activity patterns of individuals and strongly influence travel demand. Knowledge about stability and variability of these routines could strengthen travel demand modelling and forecasting. A longitudinal perspective is required to investigate these aspects. In this study, the activity patterns of a sample of people is compared for one week in two successive years. It is analyzed whether the activity patterns of a given person vary from year to year, to what degree, and how this variability and stability can be measured. It is considered whether socio-demographic factors and life events determine stability in weekly activity patterns. The study is based on the representative panel survey, German Mobility Panel. The weekly activity patterns of the same respondents in different years is assessed, using two methods to measure stability and variability. The survey respondents are clustered into three groups according to the degree of variability in their activity patterns. A logistic regression model is also used to identify socio-economic and demographic covariates for similarity in weekly activity patterns. Results show that about one-third of the sample had the same or very similar weekly activity patterns in the two years examined. A person’s occupation status is a good predictor for the variability of activity patterns. Moreover, persons undergoing a change in occupation status are quite likely to show a greater variability in their activity patterns.

Assessing the effects of a mixed-mode design in a longitudinal household travel survey

The German Mobility Panel (MOP) is a national household travel survey, which has been collecting data on travel behavior in Germany since 1994. One of the MOP’s central assets is its ability to provide time-series data on travel behavior. Thus, the comparability of survey results from different years is a major objective of the survey method used. Declining survey participation rates in the last decade in various socio-demographic groups resulted in the implementation of a mixed-mode design for the MOP in 2013, both for the sampling stage (landline and mobile phone recruitment) and the data collection stage (paper and web). In this study, we analyze whether the adaptations in the survey mode do indeed improve the results and, if so, why and to what degree. Ideally, the survey mode adaptions have increased the representativeness of the MOP. However, measurement biases due to the mixed-mode design are also conceivable. To decompose survey mode effects, we applied the propensity score weighting method. This method imputes the hypothetical responses participants would have given in different survey modes; disparities between actual responses and hypothetical responses under another mode are then traced back to the mixed-mode design. Our analysis indicates that trip-rate biases on shopping, leisure, and short trips are partly caused by the mixed-mode design; in contrast, quantities of time spent in the transportation system, trips made by car and public transportation, and commuting trips are hardly biased.