On-road vehicles have become a dominant source of air pollution and energy consumption in many parts of the world. As a result, estimating the amount of pollution from these vehicles and analyzing the factors affecting their emission is necessary to understand and manage ambient air quality. Traditionally, automobile emissions have been measured with dynamometer tests using representative driving cycles. A review of the related literature shows that there is a lack of real life, on-the-road testing of automobile emissions. Moreover, a few previous studies have directly discussed the impact of driver variability on emissions from the vehicles. This research analyzes the impacts of driver experience, gender, speed, and road grade on vehicle emissions through on-the-road testing experiment in Logan, Utah, USA during summer of 2016. The methodology of the research starts by selecting a representative car to perform the tests on. The next step was to choose test drivers representing four groups: young males, young females, experienced males, and experienced females. After that, the drivers were assigned a specified route that has different speed limits and grades. Emissions from the car exhaust (specifically carbon monoxide-CO, hydrocarbons-HC, and nitrogen oxides-NOx) in addition to the engines rotational speed (rpm), car speed, and exhaust temperature, were measured every second while driving on the specified route. Statistical analysis of the results shows that contrary to the common stereotypes, experienced drivers emitted 52% more HC and 49% more NOx than young drivers and female drivers, and male drivers emitted 14% more HC and 44% more NOx than female drivers. It also shows that CO emission is not significantly dependent on age, gender, nor driving conditions. Finally, driving through low-speed segments emits significantly higher HC (79%), while driving through uphill segments emits significantly higher (98%) NOx than driving through downhill segment.