With the advent of advanced measurement technologies in the form of smart meters and sophisticated phasor measurement units (PMUs), huge amount of valuable but raw data have become available. The synchronized data, which are gathered throughout the system, has opened up new horizons for power system monitoring and control. Most importantly, they consist of hidden patterns which, if appropriately harnessed, could facilitate derivation of innovative data-driven analytics. One important task when dealing with a large dataset is to cluster it into several groups for further analyses. Examples of such datasets could be daily and monthly energy consumption patterns of the residential consumers in a distribution system or dynamic response of power system revealed via measurements throughout the system following a disturbance. Considering the similarities among data points in a large dataset, they can be divided into smaller groups using appropriate clustering techniques so that each group consists of approximately similar data points. In this chapter, application of most well-adopted clustering techniques to power system studies will be discussed. We, in particular, focus on clustering a large power system into areas according to the similarity of post-disturbance variations in parameters. The analyses of this chapter will set the stage for a comparative evaluation of most well-known clustering techniques in other applications.