Travel Tracker Survey
CMAP has completed a comprehensive travel and activity survey for northeastern Illinois.* Data collection took place between January 2007 and February 2008. A total of 10,552 households participated in either a 1-day or 2-day survey, providing a detailed travel inventory for each member of their household on the assigned travel day(s). The resulting distribution of the sampled population closely matches northeastern IL's household population by county. The public use data set is now available for download below. (Data for 1990 are also available below for comparison and trend analyses.)
NOTE: This survey was designed and conducted for use in regional travel demand modeling. The sample size and frame is suitable for that purpose. Proper application of the public dataset is the responsibility of the user.
*An identical survey was done for the three-county northwestern Indiana region. For information related to Travel Tracker data for NW Indiana, please contact the Northwestern Indiana Regional Planning Commission (NIRPC).
Data and Documentation
Click the link below to download the Travel Tracker public data set in MSAccess format. Note that there are six main tables in the database, as described in the following table. In addition, when interpreting the data, be aware that households are divided into five distinct strata, ranging from areas that have minimal access to transit and low population density, to areas that have maximum access to transit and high population density. For details about the data format, sample stratification and other related issues, please see the Data Dictionary and User's Guide.
|Table Name||Table Description||Notes|
|hh_public||Household descriptors and travel data (e.g., hhsize, income, total trips, etc.) Unique households are identified by the field SAMPN.|
|per_public||Person-level descriptors and travel data (e.g., age, gender, disability, employment, number of trips, etc.). Unique persons (within a hh) are identified by the field PERNO.||PERNO is unique within households. To identify a unique person across households, combine PERNO witih SAMPN.|
|place_public||Trip details (e.g., travel day number, purpose, mode, travelers, tolls, departure and arrival times, distance, etc.) Unique places (per person and day within a hh) are identified by the field PLANO.||The first place in every record is the place where the respondent was at 3am (typically HOME). This is the starting place for their first trip of the day (which goes from PLANO1 to PLANO2).|
|loc_public||Location data anonymized by the centroid of the resident census tract. Unique locations are identified by the field LOCNO.||HOME location numbers are constructed by adding a "9" to the beginning of a household's SAMPN.|
|transit_public||Detailed data for multiple segments that make up one public transit trip (or the transit segments that make up the transit portion of a mixed-mode trip that includes transit). Fields include boarding location, type of fare paid, distance to final destination, etc.|
|veh_public||Detailed data for all vehicles in the household (e.g., year, make, model). Unique vehicles (within a particular hh) are identified by the field VEHNO.|
Download the Data
NOTE: In December 2016, the 2007-08 Travel Tracker Survey public use data set was revised to add two fields with supplemental survey weights developed by CMAP, and modify one field to correct inconsistent trip distance values. For more information regarding these changes, please see the "Supplemental Survey Weights" document linked below.
- Raw Data in MS Access Format
- Data Dictionary
- Updates: Supplemental Survey Weights and Corrected Trip Distance
- Revised CMAP Weighting Methodology
- Users' Guide
Final Methodology Report
Data Summaries and Analyses
New Report: "Trips Underway by Time of Day by Travel Mode and Trip Purpose for Metropolitan Chicago"
Chicago Regional Household Travel Inventory: Mode Choice and Trip Purpose for the 2008 and 1990 Surveys (June, 2010)
This report examines household travel within the northeastern Illinois region and attempts to determine how trip making may have changed since 1990. The main source of data for this analysis is the 2008 Travel Tracker household travel survey which was conducted for the northeastern Illinois region during 2007 and 2008. This survey data was weighted to represent the population of the entire region. The 2008 survey data was compared to the 1990 northeastern Illinois household travel survey, 1990 and 2000 decennial census data, and the 2005-2007 American Community Survey (ACS).
Below is a set of basic tabulations drawn from the Travel Tracker Survey data set. The tabulations Include detailed tables at the county level showing the distribution of trip purposes, trip modes, travel times, and more, The weighted data are calculated using the sample weights that are included with the original data set. For information about these weights, please see the documentation available for download above.
Below are several examples of animated sample travel days illustrating anonymized travel logs taken from the Travel Tracker Survey data. The auto mode example comes from GPS devices that respondents' carried in their vehicles, while the mixed-mode examples come from wearable GPS units. To view, download the files and unzip. Use Windows MediaPlayer or a compatable program to run the animation. Note the clock in the upper left to follow travel times.
Should I use the Weighted or Unweighted data?
The weights included with the data set were derived by comparing the sample data's demographics to actual population demographics by county. Factors included in the weighting calculation are race/ethnicity, household size, household income, and age of householder. While weights are designed to correct for under-representation and over-representation of certain sub-populations in the sampled data, they should be used with caution. For example, if a household with a very unusual travel characteristic (e.g., bicycles 25 miles round trip to work) happens to have a very large weight (because they represent an under-represented demographic in the sample), the weighting process amplifies their unusual travel behavior to the entire data set. Looking at the travel patterns of households with high weights might be a good place to start when using the weighted data set, therefore. If problematic households are identified, one might decide to drop those households from the data and re-weight the data set.
1990 Travel Survey Data for Northeastern Illinois
Click here to download the 1990 Travel Survey data set, including documentation. The data set is available in three formats: MSAccess, SAS, and TEXT.