| Year | English | Hmong | Karen | Oromo | Somali | Spanish |
|---|---|---|---|---|---|---|
| 2021 | 14,645 | 1 | 0 | 3 | 3 | 86 |
| 2023 | 5,660 | 1 | 1 | 1 | 1 | 95 |
| Includes all recruited households (those who completed the signup survey). Some households did not go on to complete the travel diary and are not in the final dataset. | ||||||
Survey Design
The Travel Behavior Inventory employed a mix of data collection methods, including smartphone, online, and telephone. The study design balanced the strengths of innovative technologies with pragmatic best practices derived from traditional market research.
The survey occurred in two parts. Part one (the recruitment survey
) gathered data on each household’s demographic composition and typical travel behaviors. The recruitment survey was conducted online through the study web page, over the phone through the study call center or using the rMove™ app (starting in 2021).
Part two (the travel diary
) gathered travel data for all related household members during a designated travel period. Households where all adults had smartphones were required (2019) or encouraged with incentives (2021, 2023) to download the rMove™ app and complete a seven-day travel diary.
If one or more adults in the household lacked a smartphone (all study years) or if the household opted out of using rMove™ (2021, 2023) the household completed a one-day travel diary. The one-day travel diary could be completed online or by calling the study call center. Call center operators used the same online travel diary instrument as survey participants. Non-app households were assigned a travel date of Tuesday, Wednesday, or Thursday to capture typical weekday travel behavior.
Recruitment
Participants were recruited using a series of invitations mailed to their home addresses. Each mailing contained information about the study and a unique password for their household. See Invitation Materials for examples.
RSG leveraged contact information captured in the recruitment survey to provide customized, strategically limited, and well-timed reminders to recruited households by email, telephone, or within the rMove™ smartphone app to ensure nearly real-time completion of travel surveys.
Language Options
Throughout the survey stages (e.g., recruitment and travel diary) and participation modes (e.g., app, online, or call center), several language options were made available to participants. The call center employed operators fluent in English and Spanish. Print materials included translated text with instructions on how to participate in six languages: English, Spanish, Hmong, Karen, Oromo, and Somali. The survey instruments were also available fully translated from English to Spanish, Hmong, Karen, Oromo, and Somali. In addition, the study website included a Google Translate bar to accommodate approximately 100 other languages. Table 1 shows the number of recruited households who took the survey in each offered language option in 2021 and 2023. Data was not available for the 2019 survey effort.
Incentives
As compensation for their time and efforts, gift card incentives were offered to households for completing at least five complete travel days (smartphone diary participants) or a complete online travel diary (online participants). Otherwise incomplete households did not qualify to receive the incentive. Many surveys, including most household travel surveys and the National Household Travel Survey (NHTS), offer incentives because:
- Incentives increase response rates significantly, often doubling participation rates.
- Incentives help reduce participant bias.
- Historically low response rates in the United States and abroad mean that incentives are now often required to obtain a sufficient sample.
- By increasing study response rates, incentives help save the project money overall and encourage the most cost-effective use of public research funding.
- Incentives represent fair payment for the expertise participants provide on their own lived experience.
The study offered gift cards from Caribou Coffee, Target, and Amazon.com as incentives; participants could also waive their right to an incentive (in 2023, 9% of participants chose this option). These vendors were chosen for their availability within the travel study region, wide selection of products, widespread name recognition within the travel study region, and administrative ease of use to obtain and send gift cards to study participants. Table 2 shows the incentive offerings in each survey year, which varied based on diary platform and if the household was determined to be a hard-to-reach
household. In 2019, all households were offered the same incentive (differential incentives were not offered).
| Year | rMove Hard-to-Reach | rMove General | Online Hard-to-Reach | Online General |
|---|---|---|---|---|
| 2019 | $15 | $15 | $10 | $10 |
| 2021 | $30 | $20 | $20 | $10 |
| 2023 | $30 | $20 | $20 | $10 |
| rMove™ incentives were per adult while online incentives were per household. | ||||
In 2021 and 2023, general
households (i.e., not hard-to-reach) that completed with the one-day online travel diary were offered a $10 e-gift card to their vendor of choice, or a physical gift card for households without email. Households that used rMove were offered e-gift cards of $20 per adult rMove participant to their vendor of choice. Thus, if a household had three rMove™ participants (each at least 18 years or older), then they received $45 in gift cards.
This incentive structure was designed to encourage larger households to participate using rMove™, which required more effort on behalf of each participant, and collected data for a greater number of days (seven travel days vs. one travel day).
| Diary Platform | Payout per Household |
|---|---|
| rMove™ | $34.26 |
| Online | $10.81 |
Questionnaire
Initial Design (2018)
In spring 2018, an initial data variable list was developed for the Travel Behavior Inventory Pre-test study. As part of this process, the Met Council obtained input from Minnesota Department of Transportation and Metro Transit. A two-fold approach was used: building upon the work performed in the 2014-2015 scoping project and building upon the data needs from prior smartphone-based household travel survey projects conducted by the consultant, RSG, in the U.S. and elsewhere.
The data needs for 2018-2019 data collection were finalized in August 2018 after the project team members reviewed pre-test results and identified necessary updates. The final data variable list summarized below aimed to satisfy core household travel survey data elements and select supplementary data elements that will benefit the Met Council’s planning needs. Those key data elements included the following:
Core Household Travel Survey Data Elements
- Household-level (e.g. household composition, income, housing characteristics)
- Person-level (demographics, employment, and student status)
- Vehicle-level (year/make/model, ownership)
- Trip-level (activity, mode, costs, travel party composition, etc.)
- Location-level (GPS points, path data, etc.)
Supplemental Data Elements
- New travel mode usage including smartphone-app ride services (e.g., Uber, Lyft), peer-to-peer car rental, bikeshare (both docked and dock-less), and scooter share (electric scooters and mopeds)
- Trip replacement behavior such as package delivery (to participant’s home, work, and package lockers), meal delivery, and service work (e.g., Electrician visit), telecommute hours, and online shopping behavior
- Interest in autonomous vehicle (AV) use
- Land-use questions requested by the Met Council’s planning staff (e.g., monthly household rent)
Updates in 2021 and 2023
Note
Note: This list of updates is not comprehensive. For a complete overview, consult the combined codebook or the questionnaires.
In 2021 and 2023, several updates were made to questionnaire language and content. Some highlights of these updates are provided below.
Race and ethnicity. In 2019, respondents were asked about their race and ethnicity in a single, multiple-response (select-all-that-apply
) question. In 2021 and 2023, race was asked separately from ethnicity. This allowed the survey to gather more detailed responses from those who identified as Hispanic. The survey furthermore asked follow-up questions for those who identified as African American or Asian/Asian American. These updates were made to align the Travel Behavior Inventory with other regional and national efforts, including the National Household Travel Survey.
Gender. In 2019 and 2021, transgender
was listed as a response option for gender separate from Male
and Female.
This way of soliciting gender identity conflate gender identity with transgender status; in other words, the identity of woman
belongs to both cis- and transgender women. To be trans-inclusive, the response options for gender were updated in 2023 to read:
- Female/Woman/Trans woman/Girl
- Male/Man/Trans man/Boy
- A gender other than singularly male or female (e.g., non-binary, genderfluid, agender, culturally specific gender)
- Other
- Prefer not to answer
Despite concerns that the new trans-inclusive language might turn away some respondents, only five participants dropped out of the survey at the gender
question screen. For comparison, the same number of participants dropped out when asked to verify which household vehicles used a toll transponder. By contrast, 182 participants dropped out when asked to provide their home address.
New supplemental data elements
From one survey year to another, some questions were added while others were dropped. A handful of these are listed below, along with their corresponding variable_names in monospaced text.
- 2021, 2023 Questions about typical electric vehicle charging behavior (
ev_typical_charge_[1-6]), charging behavior on trips (ev_charge,ev_charge_time,ev_charge_station_decision,ev_charge_station_level_\[1-3\], and attitudes towards EV purchasing (ev_purchase) were added.
- 2021, 2023 Proxy reporters for school-aged children were asked if their child traveled to school, on days when no school trips were otherwise present in the travel diary (
attend_school). When the child did not travel to school, the proxy reporter was asked to select reason(s) they did not travel to school on that day (attend_school_no).
- 2021, 2023 Additional questions about bicycle types , micro-mobility devices (
micromobility_devices), and bike storage locations (bike_store) were asked of each household.
- 2019, 2021, 2023 Travel modes (
mode_type_detailed) and shared mobility services (share_[0-7]) were verified each year for their relevance to the region.
- 2021, 2023 Participants were asked if one or more barriers to travel impacted them in the past week (variable
transportation_barriers).
Questionnaires
The complete questionnaires are attached in the appendix.
Note
Note: In 2019, the web browser-based and smartphone-based questionnaires had slightly different structures. In 2021, RSG updated the technology underlying the rMove app, such that the web- and smartphone-based surveys drew from the same source. For this reason, the 2019 web-based and smartphone surveys slightly differ in their structure.