This study aims to quantify and determine the underlying determinants of public bus transit travel-time variability using data collected in Turkey's capital of Ankara. Public transit service in the Ankara metropolitan area is primarily based on the bus system, which is operated by both the public and private sectors. The bus transit system is generally mixed with other traffic (Right-of-Way C); therefore, it is expected that travel times are subject to high degrees of variability. The system is organized in a hub-and-spoke fashion, with the hub at one of the city centers (Ulus or Kizilay) and the spokes extending to the urban outskirts. Regarding travel-time variability in the public transit bus system, three models are developed to account for variation caused by operational regions, highways, and individual bus lines. In addition, two sub-models have been developed with respect to the dependent variables used: the standard deviation and the coefficient of variation of travel times. Three main causes of travel-time variability have been identified and tested in this study: temporal dimension, spatial dimension, and service characteristics. Model results indicate that all of these factors affect travel-time variability. We conclude that a uniform reorganization of the system must be undertaken to control travel-time variability. (c) 2012 Elsevier Ltd. All rights reserved.