Type: Empirical

Purpose: The purpose of this article was twofold. The first part described the Chain of Cognitive Accomplishments. The second part described a set of studies of middle school programming classes.

Findings: A chain of cognitive accomplishments was described as a model of the learning outcomes of programming courses. This chain consists of 3 links. The first link requires understanding the language features (e.g., GOTO, IF...THEN) in using various programs. The second link are design skills that include: using templates to organize knowledge and the use of procedural skill to plan, test and reformulate problems. The third link was program solving skills that applied generalized procedural skills and generalized templates to new systems Typical instructional classrooms were selected if teachers participated in one in-service computer workshop, there were at least 8 computers in the classroom and the duration of instruction was at least 12 week. Exemplary instructional classrooms explicitly taught how to design programs and teacher had more experience in programming. Findings from 12 weeks of instruction in a typical classroom concluded that middle school students learned some language features of BASIC but a range of 20% to 68% on the language feature items. A range between 11% and 33% were scored on the design items. Because of the low range on the design items, the problem-solving ability items were not assessed. Student in the exemplary classrooms mastered the language features items. The range was 58% to 81% in designing programs. These students were able to design programs with up to 50 lines of code.

Recommendations: Teachers should emphasize "relationships between skills used to solve problems in each discipline" (p. 28). The gap of students without access to computers at home is an issue. The author suggests that improvements are needed to the curriculum and preparation for teachers for programming classes.