All time references are in CEST
Current Developments in Mobility Survey Methods 2
|Session Organisers|| Dr Johannes Eggs (infas)
Mrs Dana Gruschwitz (infas)
Dr Stefan Hubrich (TU Dresden)
|Time||Wednesday 19 July, 09:00 - 10:30|
Travel surveys collect data on the mobility of populations. Large-scale national household travel surveys are used to estimate key mobility figures on national or sub-national levels, like out of home rates, trips and transport mode rates, and to predict the demand of the population regarding the use of transportation modes. But high respondent burden and sinking response rates are pressing on the established travel survey designs. On the other hand new technologies are available to ease survey participation.
The focus of this session is explicitly wide and aims to gather an overview about the cur-rent developments, challenges and ideas for mobility surveys. Papers matching one of the following aspects are invited to be part of this session (but are not restricted to):
– New technologies for data collection and mixes mode approaches
– Integrating GPS data in trip and survey data
– Gamification in travel surveys
– Survey mode effects in travel survey
– Improving trip reporting in different modes
– Improving household completion rates
– Effects of proxy interviews
– Methods to collect mobility data for specific sociodemographic groups (e.g. children)
– New sample frames and weighting procedures
– Incorporation of big data
Keywords: Travel survey, mobility survey, passive data collection
Dr Andreas Filser (Institute for Employment Research (IAB)) - Presenting Author
Mr Florian Zimmermann (Institute for Employment Research (IAB))
As smartphones have become omnipresent, data collected as by-product of their usage create new opportunities for social and economic scientific research. Based on a proprietary app, the IAB-SMART study collected smart survey data and digital trace data on a subsample of the “Panel Study Labour Market and Social Security” (PASS), a longstanding German Panel study. Linking these complementary data sources not only generates comprehensive insights into the impact of various forms of social inequality, but also enables to test the smartphone as a combined survey instrument. Within the field period from January to August 2018, the study generated about two and a half million geolocation datapoints.
Our project seeks to leverage the contextual and methodological potential of the geolocation information data by providing data access to the research community. Distributing comprehensive smartphone data requires processing and anonymization to comply with data protection requirements. The risk of re-identifying individual participants is particularly high for geolocation trace data. The IAB-SMART app collected geolocation coordinates in 30-minute intervals, which bears the potential to identify where participants live, work, and spent time. Consequently, the geolocation data could not be distributed in their original form, but had to be aggregated to indicators.
Funded by BERD@NFDI, we generated anonymized, aggregated, and documented geolocation indicators that will be available to researchers by the time of the ESRA 2023 conference in combination with the PASS survey and PASS-ADIAB employment biography data. This presentation describes the data preparation challenges that had to be addressed along the way and will provide an overview of the indicators in the IAB-SMART module.
Ms Miriam Magdolen (Karlsruhe Institute of Technology) - Presenting Author
Mr Lukas Burger (Karlsruhe Institute of Technology)
Dr Bastian Chlond (Karlsruhe Institute of Technology)
Professor Peter Vortisch (Karlsruhe Institute of Technology)
Researchers studying travel behavior need to analyze the longitudinal perspective of the mobility of individuals to understand the overall mobility needs and the variation in behavior e.g. using the car on one day and the bike on another. However, the amount of information needed conflicts with the response burden, making it difficult to recruit participants for longitudinal surveys. To address this, a new survey approach called the ’travel skeleton’ was developed and applied in several cities worldwide since 2016. In this approach, participants are asked about relevant out-of-home activities and related travel as typical behavior, e.g., how often they commute to work in a typical week and which means of travel they usually use. This simplified form of data collection compared to traditional travel diaries allows for the inclusion of additional modules, e.g., to capture psychographic information or long-distance travel.
As the survey asks the participants to report their typical behavior, it can be conducted at any point in time, but still captures the overall mobility with a longitudinal perspective. The response burden is comparably low, as the whole survey can be completed in only a few minutes instead of reporting for a longer survey period. However, the participants need to assess their typical behavior, which involves a high cognitive burden. Nonetheless, we see that the data collected allows to describe the participants' mobility needs. While the data cannot directly be used for transportation statistics or as input for travel demand models (e.g., no temporal information is collected), it does allow for an improved understanding of travel behavior and travel needs, and thus for example to distinguish the participants into mobility types (e.g., people with low mobility needs and people who are very active travelers and have more complex mobility patterns).
Mr Inan Bostanci (Zuse Institute Berlin)
Miss Yvonne Gootzen (Statistics Netherlands)
Dr Peter Lugtig (Utrecht University) - Presenting Author
Traffic estimation is an important tool in official statistics and can inform policy makers in their decision-making process for regional planning. Statistics Netherlands (CBS) developed a framework to estimate traffic counts on a nationwide scale, linking administrative, survey and infrastructure data. However, it relies on assumptions about the population and traffic estimates cannot be confirmed with the data.
This paper shows how the framework can be extended by including traffic loop sensor data in the linkage process for calibration and validation. Sensor data provides observed vehicle counts for a sample of road segments on the road network. It therefore reveals the accuracy of expected counts on these segments.
In this paper, geographical and road network information is used to predict the accuracy on unobserved road segments.
Multiple calibration models are developed and compared. Finally, estimations are calibrated for the entire Dutch road network. The paper demonstrates how data linkage can be performed to produce valid nationwide traffic estimation for official statistics and policy-making.
Mr Frederic Gerdon (Mannheim Centre for European Social Research (MZES), University of Mannheim. Department of Statistics, Ludwig-Maximilians-Universität in Munich) - Presenting Author
Ms Daria Szafran (School of Social Sciences, University of Mannheim)
Mr Jakob Kappenberger (Mannheim Center for Data Science, University of Mannheim)
Mr Ruben Bach (Mannheim Centre for European Social Research (MZES), University of Mannheim)
Mr Christoph Kern (Department of Statistics, Ludwig-Maximilians-Universität in Munich)
Previous research has demonstrated that simulation techniques such as agent-based models (ABM) allow researchers to model traffic behavior in spatial environments, such as smart cities. These environments may contain human and technical actors which have certain attributes and follow certain strategies with their behavior. Surveys can aid us in specifying concrete ABM, such that calibrated models can provide hints to how local populations may react to changes in traffic infrastructure, e.g., by introducing real-time flexible parking fees. These reactions particularly include potential unintended consequences with respect to local social inequ(al)ities, as some infrastructural changes may shift opportunity structures unfavorably for already disadvantaged populations. Survey data can provide empirical information on the distribution of human agent characteristics and their (potential) behavior, such as parking lot search strategies. In this talk, we (1) present results from a survey experiment that is conducted in early 2023 to measure preferences for parking in cities depending on parking characteristics and individual attributes. This experiment asks respondents to repeatedly choose between parking scenarios that randomly vary with respect to characteristics such as price and distance to destination. Additionally, we (2) show preliminary results from ABM that use the data from the experiment to show whether parts of the population may be disproportionately disadvantaged by the introduction of flexible parking fees. We conclude by pointing to potential extensions of ABM for traffic research and pitfalls we experience during the simulation development.