In this paper, we analyzed the perception of Computational Thinking among engineering students from three engineering disciplines (Electrical, Mechanical, and Civil) and correlated their performance with their discipline. The goal of this analysis is to determine whether structuring discipline-specific Computational Thinking courses can improve the retention or having a diverse group of students in this course is more beneficial by allowing multidisciplinary interaction. This analysis was quantitatively verified by assessing the students' performance in over 40 different sections of Computing for Engineers course taught from Fall 2012 to Spring 2014. Our sample consisted of 861 students (142 Civil, 484 Mechanical, and 235 Electrical). Students’ performance was assessed using quizzes, assignments, lab projects, and exams. We statistically analyzed students' performance in this multi-section course to draw our conclusions which can help structuring Computational Thinking courses for engineering students from different engineering disciplines.