Summer in the Field: Show Me How You Move – Developing Methodology for Estimating Demand for Transportation in Addis Ababa, Ethiopia
In rapidly urbanizing cities like Addis Ababa, transport infrastructure serves as a crucial foundation for economic growth, social connectivity and overall quality of life.
However, the benefits and challenges of mobility are not experienced equally across all segments of society. Women, who often carry responsibilities for housework, caregiving and work, face unique mobility barriers and have different travel needs compared to their male counterparts. Therefore, exploring these gendered differences in mobility is essential for understanding their broader implications for equitable development.
As part of my 2024 Summer in the Field Fellowship, I continued work on the Ethiopia Transport Study, building on our previous findings as we prepared for the public transport intervention launch.
Our earlier study revealed that providing women with free and efficient taxi services significantly reduces the gender gap in physical mobility. It also demonstrated positive labor market outcomes, with both women and men who received taxi allowances showing increased employment.
However, scaling a taxi-based intervention to everyone is neither feasible nor desirable from a planning perspective. After identifying latent demand for private transportation, we shifted our focus toward understanding public transportation demand, as it has the potential to serve a larger population while being cost-effective.
Tracking the movements of individuals using public transportation presents greater challenges than taxis, where trip data can be easily collected through applications. Some recent research has explored innovative methods like analyzing anonymous telecom data and using smart phone-based GIS tracking. While these approaches provide useful insights into movement patterns using largescale location data, they fall short in linking trip data to individual socioeconomic characteristics, making it difficult to fully understand the underlying demand factors.
To address this gap, we aimed to develop and test new passenger trip tracking methods that would yield reliable data on public transport usage. We designed a survey that could be administered via three different modes: phone calls, SMS messages or a messenger-app chatbot. Phone surveys have several advantages, including lower attrition rates, greater comfort for respondents speaking to human enumerators and potentially reduced misreporting—an important factor as our public transport intervention includes subsidies, making cost monitoring crucial.
Message-based surveys, by contrast, offer the benefits of lower cost—critical for large-scale interventions—faster implementation, and reduced potential for enumerator errors. However, there are important distinctions between SMS and messenger-app methods. SMS can reach a broader audience, which is important given the gender disparity in smartphone ownership; only 40 percent of women in our initial sample own smartphones compared to 70 percent of men. Relying solely on smartphone-based methods would bias our sample toward higher-income, more educated male population. In contrast, messenger-based surveys offer the advantage of GPS functionality, allowing for more accurate tracking of trip origins and destinations.
In addition, we tested three incentive schemes for cost reporting. While distributing a lump-sum payment is the simplest subsidy method, it may not effectively promote travel and could be spent on other needs. Moreover, for women with lower bargaining power, a lump-sum transfer could be appropriated by their spouses. A more targeted approach involves reimbursing reported trip costs on a daily or weekly basis, ensuring the money is more likely spent on transportation. However, this method introduces the risk of misreporting, such as inflating costs or reporting trips that did not occur.
The third incentive scheme involves offering a small, fixed payment for each completed survey, regardless of reported travel costs. While this might be insufficient to motivate participation (respondents could simply claim they didn’t travel), it could theoretically reduce misreporting when trips did occur.
To identify the most effective subsidy distribution method, we compared key metrics—such as total cost, distance traveled and the number of trips reported—across the three incentive schemes. The schemes we tested were as follows: Fixed (25 ETB per completed survey), Mixed (25 ETB per survey plus reimbursement of 50 percent of the travel cost, capped at 200 ETB per week) and Lump Sum (200 ETB for the first survey, plus 25 ETB for each survey).
Our pilot study spanned two weeks, with one week devoted to training and preparation and the second to data collection. We conducted recruitment at Addis Ababa’s largest transport hubs (Piassa, Megenagna and Stadium), inviting passengers to participate in the pilot. Each incentive scheme included 20 participants, with an equal number of women and men, and a balance between mobile and smartphone users. After completing a brief baseline survey to collect socioeconomic characteristics, participants engaged in daily travel surveys via phone and message-based methods.
The data analysis is ongoing, and we are eager to determine which survey method and incentive scheme are the most reliable, cost-efficient and convenient for respondents. These insights will help us design a larger-scale public transport intervention in Addis Ababa.
*
Learn more about the Summer in the Field Fellowship Program.