Survey Data on Algorithmic Opacity and Coordinated Trip Rejection among Platform Bodaboda Drivers in Tanzania

Published: 29 October 2026| Version 1 | DOI: 10.17632/ckx65kjhp5.1
Contributor:

Description

This dataset contains anonymised survey responses from 412 motorcycle taxi (bodaboda) drivers who work through ride hailing platforms in Dar es Salaam and Dodoma, Tanzania. The data were collected between February and April 2026. Trained enumerators used a stratified intercept approach at 46 bodaboda stages across central, intermediate and outer zones of both cities. Eligible drivers were at least 18 years old, held an active account on a ride hailing app offering motorcycle trips, and had completed at least one platform trip in the previous two weeks. The questionnaire was administered in Kiswahili on tablets, and an English version is included. It measures six constructs on five point agreement scales: Perceived algorithmic opacity (four items). Beliefs about algorithmic responsiveness (four items). Expected economic benefits of coordinated trip rejection (three items). Perceived sanction risk (three items). Willingness to coordinate trip rejection (four items). Perceived belief verifiability (three items). A marker variable is included for common method variance checks. The dataset also records: Demographics: age, gender, education, marital status and dependants. Work characteristics: motorcycle ownership arrangement, platform tenure, number of apps used, weekly hours, share of income from apps and daily net earnings. Social ties: stage work, messaging group membership and stage association membership. Self reported participation in coordinated trip rejection during the previous six months. City, a coded stage identifier and interview order, which support multilevel, multigroup and nonresponse bias analyses. All participants gave informed oral consent, including consent to the sharing of anonymised responses. Stage names, locations and any other potentially identifying information have been removed or coded. The deposit includes: The data file in CSV and Excel formats. A codebook defining every variable and its coding. The questionnaire in Kiswahili and English. The SmartPLS 4 project file used for the partial least squares structural equation modelling analysis. The data support research on algorithmic management, information asymmetry, platform work and collective action in African digital economies. Suggested keywords: algorithmic management; algorithmic opacity; information asymmetry; platform work; collective action; ride hailing; bodaboda; Tanzania; PLS SEM

Files

Institutions

Categories

Information Systems Management, Survey

Licence