The study aimed to construct a sacle of self-directed learning skills for students in the tenth and eleventh grade in the Sultanate of Oman, verify its validity and reliability, identify the level of self-directed learning, and verify the significance of differences in the variables of gender and grade. The study adopted a descriptive approach as the study's sample consisted of (2034) male and female students from the tenth and eleventh grade in Muscat Governorate schools for elementary and post-elementary education. A scale of self-directed learning skills was constructed according to the Garrison model to achieve the objectives of the study. The final form of the scale consisted of (53) items with five rating scale: strongly agree, agree, not sure, disagree, and strongly disagree. The scale includes face validity (content validity), criterion-related validity, construct validity, and factorial validity through the exploratory factor analysis and confirmatory factor analysis resulting in three factors; selfmanagement and achieving goals, self-monitoring, and learning planning. These factors explained (37.18 %) of the total variance. As for reliability, the values of Cronbach’s alpha internal consistency coefficients ranged from (0.87) to (0.93) and from (0.88) to (0.94) via the split-half method. The results showed that most of the students in the tenth and eleventh grades had average or above average levels of self-directed learning skills with an (86.33%) percentage. The differences were statistically significant in the gender variable in favor of the female students, and in the grade variable for the tenth grade. In light of these findings, a number of recommendations and proposals have been produced.