RNS
Browse Projects > Detailed View

QC Introducing Bias into Household Travel Surveys

Description:

Household travel surveys provide the data backbone of nearly every travel demand model. The data collected in a survey become the basis of trip generation, distribution, and mode choice models in a trip-based modeling system, and the basis of the population synthesis, activity pattern generation, singular and joint tour models, destination choice, mode choice, and time choice models in an activity-based model. As such, practitioners spend a lot of effort on quality control of the data to ensure that only good and complete data collected in a household survey is used for subsequent model estimation tasks.

Does bias introduced by quality control checks significantly bias survey data? Does the use of a complete household (all reported trips by all persons fully report all important model estimation variables) as the base unit of model estimation exacerbate this problem? Complex household travel patterns or larger households seem to be more likely to be dropped from a household travel survey dataset due to failing quality control checks even if on a trip-level the error rate is the same. However, this could introduce bias by eliminating good and rare data embedded in a complex travel patterns or household. This research is intended to quantify if such a bias exists and to create better quality control checks to reduce the bias caused by incorrectly eliminating large, complex households from a household travel dataset.

Objective:

Determine the presence and magnitude of bias introduced into household travel surveys through data quality control processes and develop methods to mitigate or correct any bias detected.

Implementation:

Any new QC checks resulting from this research could ease model calibration for both trip-based and tour-based models. Once understood, new QC checks are generally not difficult to implement by practitioners.

Sponsoring Committee:ABJ40, Travel Survey Methods
RNS Developer:Authors are Jonathan Ehrlich and Andrew Rohne.
Date Posted:08/29/2017
Date Modified:04/20/2018
Index Terms:Quality control, Bias (Statistics), Travel surveys, Households, Data quality, Travel behavior,
Cosponsoring Committees: 
Subjects    
Data and Information Technology
Planning and Forecasting
Transportation (General)

Please click here if you wish to share information or are aware of any research underway that addresses issues in this research needs statement. The information may be helpful to the sponsoring committee in keeping the statement up-to-date.