**Type:** Empirical

**Purpose:** The purpose of this study was to investigate if Scratch can be used to teach concepts of computer science. The researchers also investigated the extent of the students' internalization of the concurrency concept.

**Findings:** Statistically significant differences were not found between students' achievement on pre, interim and posttests in the two classes. There were changes in the students' mean grades between pre and post 3 cognitive process categories which were, multistructural understanding, multistructural creating and relational applying. These results contradicted researchers initial conjecture. The students were able to answer the more difficult question of relational applying level (62.5%) than answer the second question of multistructural applying level. There was a decrease in mean grades from pre- to post in multistructural applying. In terms of students' internalization of CS concepts, correct definitions of bounded loops, conditional loops, and message passing was given by more than 56% of the students during posttest.

**Recommendations:** Explicit instruction in each concept is needed for students to acquire a deep understanding of the CS concepts.

**Sample Size:** 46

**Participant Type:** Students were the main focus on this study.

**Notes:** These students are 14-15 years old and are in the ninth grade. This study consisted of two classrooms where the first class had 18 students (11 boys and 7 girls) and the second classroom had 28 students in an all-boys school. The teacher in the first class was a mathematics teacher with no CS experience, and the second teacher had 15 years experience teaching CS. Both teachers were introduced to Scratch environment for the first time.