German National Travel Survey 'MiD 2016 โ Mobility in Germany': New Challenges โ New Approaches
Angelika Schulz, German Aerospace Center (DLR) - Institute of Transport Research, Johannes Eggs, infas - Institut fuer angewandte Sozialwissenschaften, Marcus Baeumer, IVT Research
The paper describes adjustments made to the German NTS MiD 2016 in order to meet known methodological challenges and to serve emerging data needs. Changes apply at different levels (sampling, instruments, topics, integration of supplementary data).
Following the long tradition of nationwide household travel surveys in Germany beginning in 1976 (KONTIV 1976, 1982, 1989; MiD 2002, 2008), a new survey will be conducted form June 2016 to June 2017 on behalf of the German Federal Ministry of Transport. Again, the main objective of this cross-sectional survey is to provide representative and reliable information on the countryโs everyday mobility: Individual one-day trip diaries will be collected along with socio-demographic characteristics of private households (such as household composition in terms of age and gender or car ownership) and individual characteristics (such as driving licence holding or de facto car availability) that are likely to influence actual mobility behaviour.
To a great extent, the methodological approach of the upcoming survey will correspond to its immediate predecessors of 2002 and 2008 to ensure extensive comparability over time. Nevertheless, several adjustments have been made to meet known challenges (e.g. difficult recruitment, decreasing response rates) and to serve emerging data needs (e.g. impact of multiple car ownership or innovative mobility services, importance of non-car modes, implications of demographic change and changing spatial structures).
Methodological changes apply at different levels such as sampling design, survey instruments, topics and the integration of supplementary spatial data:
- Difficulties in recruiting participants associated with seriously decreasing response rates โ as already encountered in earlier surveys โ demand particular effort:
In terms of sampling a triple-frame approach has been adopted consisting of three complementing sampling frames (list of municipalities together with municipal registers, random digit samples of both landline and cell phone numbers).
Aiming at increasing overall response rates, a mixed-mode approach will be applied: for each part of the survey (household interview, individual interview including trip diary) different survey instruments will be provided in parallel to allow respondents to choose their preferred mode (i.e. traditional paper-and-pencil questionnaire, computer-aided telephone interview, online questionnaire).
- The intended coverage of additional topics resulted in a considerable number of new questions. In order to limit response burden, the catalogue of questions has been divided into a core questionnaire and several sub-questionnaires. While the core questionnaire has to be completed by each household and individual, sub-questionnaires are designated to sub-samples only.
- Contrary to the 2008 survey, trip diaries will include detailed destination information (geocoding). In order to minimise response burden during trip recording, trip data collection and validation will be supported by an extensive online database containing several million Points of Interest (POI).
- Geocoded information at household and trip level that will be conducted via questionnaires and interviews will further be enriched by merging additional spatial information (e.g. given infrastructure and respective accessibility).
- Given the nationwide approach and the stratified sampling design, MiD microdata so far has mainly been used to estimate population characteristics at national or state level and for respective spatial categories. In order to provide estimates for smaller spatial units (e.g. districts) or for specific sub-groups of the population, small area estimation (SAE) techniques will be applied.
The paper will discuss the rationale behind methodological changes as well as expected consequences throughout all survey processes such as data collection, data processing, weighting, projection and data analysis.
Association for European Transport