Purpose To effectively develop leaders, human resources and career development professionals need an effective method of determining leadership skill levels. For example, sending a novice leader to training meant for experts would likely be ineffective and frustrating for both the instructor and individual. Promoting a novice leader to a position requiring expert leadership skills could be disastrous. The purpose of this study was to determine if the Dreyfus (2004) model of skills acquisition could be applied to general leadership. Design/methodology/approach A total of 124 surveys were collected from five employers. Participants self-assessed their leadership skill level from novice to expert using Dreyfus level descriptions in 18 leadership self-efficacy dimensions identified by Anderson, Krajewski, Goffin and Jackson (2008). For comparison, leadership self-efficacy (LSE) was also measured with a self-assessment of proficiency in 88 specific leadership and management behaviors also identified in the Anderson et al. (2008) study. Findings Pearson correlation coefficient computations between total LSE and average Dreyfus level dimensions reported a strong positive correlation [r (124) = 0.644, p < 0.001] between total leadership self-efficacy and the average participant Dreyfus level self-assessments in each of the 18 leadership self-efficacy dimensions. Of the 18 LSE dimensions participants assessed their skill levels, 4 were found to be significant predictors of LSE [F (4,119) = 67.6887, p < 0.001] with an R2 = 0.482. Predicted leadership self-efficacy is equal to 187.14 + 16.327 (Project Credibility) + 8.046 (Mentor) + 6.971 (Build) + 9.342 (Solve). Research limitations/implications The majority of the individuals in the sample in this research study were from one employer, a local college (n = 88). The entire sample was from one small, somewhat isolated community. The majority of this sample was female (n = 81, 65 per cent) and white (n = 118, 95.2 per cent). A larger and more diverse sample may provide differing results. It also possible that other factors affected overall LSE, but using that score as a comparison, a clear correlation was shown between LSE and Dreyfus levels. Practical implications Based on the results of this study an individual who self-categorizes his or her leadership skill as novice, advanced beginner, competent, proficient or expert is likely correct. Should the findings of this study prove generalizable, an individual’s perceived leadership skill level could be closely approximated with a simple, four-item instrument. Originality/value The concept of leadership levels has appeared in many studies and popular press publications. However, quantifying leadership skill levels or determining an individual’s leadership level has not been often addressed. This study attempts to apply a skills acquisition model and apply it to general leadership. The results appear to show that leadership levels can be quantified and accurately self-determined. This study also attempted to validate a leadership self-efficacy model.