The increasing number of food delivery riders, known as P-hailing riders, has increased the number of traffic crashes involving this group. The high demand for online orders and delivery time constraints causes the delivery riders to engage in risky riding behaviours during delivery. Thus, there is a need to study this alarming problem before it becomes more serious. This study aims to determine factors influencing risky behaviours among delivery riders at signalized intersections. A roadside observation was conducted in the month of October 2022 at Subang Jaya, Selangor. A total of 19,803 delivery riders were observed for four days observations at two signalized T- and two cross-signalized intersections. A bivariate analysis (logistic regression) was applied to determine the relationship between seven risky behaviours (unfastened helmet, not wearing shoes, incomplete set of side mirrors, red light running, mobile phone use, stopping after stop lines, and abrupt lane change) with five explanatory variables including the day of the week, time of the day, weather condition, type of intersection and approach road. The result shows that red light running, and mobile phone use is more likely to occur during the weekend. Clear weather increases risky behaviours such as red light running, using a mobile phone, and abrupt lane changes. The findings of this study are useful as input to the related authorities in developing programs to decrease the number of crashes involving food delivery riders.