Despite its critical role in counselor training, empirical research on clinical supervision is generally limited (Bernard & Goodyear, 2003; Ellis & Ladany, 2007). This is also applied to an area of power dynamics in supervision. This study tested the relationship between the two aspects of power dynamics; namely, supervisors' power bases (i.e., sources of influencing others) and supervisory styles (i.e., typical ways of shaping supervision), based on the system's approach to supervision model (Holloway, 1995). This research was a correlational design. Students in masters' and doctoral counseling programs were asked to respond to an online questionnaire packet via Survey Monkey(TM). Of those who responded, 492 students who took supervision with professor or doctoral student supervisors constituted the sample. Varied numbers of participants were used for each analysis after missing or extreme data were deleted. Supervisors' usage of power bases and supervisory styles were measured by the adopted version of Interpersonal Power Inventory (Raven, Schwarzwald, & Koslowsky, 1998) and Supervisory Style Inventory (SSI; Friedlander & Ward, 1984), respectively. In part 1, results of factor analyses revealed four first-order power factors and two higher-order power factors (Soft & Harsh). Schmeid-Leiman's (1957) solution was also applied. In part 2, result of correlation analysis in revealed that supervisors' usage of Soft or Soft-type power factor (Idealized Expert) was moderately positively correlated to all three supervisory styles but that usage of Harsh or Harsh-type factors (Compensatory Obligation, Relational Power, & Collaborative Alliance) was only weakly correlated to supervisory styles, for majority of supervisors. Similarly, results of regression analyses revealed that supervisory styles did not significantly predict supervisors' usage of Harsh factor, but both supervisory styles and usage of Harsh factor significantly predicted usage of Soft factor at moderate and strong level, respectively. The interpersonally-sensitive styles predicted Soft factor slightly more strongly than the other styles. It was concluded that supervisors who engaged in supervision with any one of three supervisory styles also tended to use more Soft or Soft-type factor when there are disagreements, but rarely used Harsh or Harsh types. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]