A multi-national, multi-institutional study of assessment of programming skills of first-year CS students

Abstract

In computer science, an expected outcome of a student's education is programming skill. This working group investigated the programming competency students have as they complete their first one or two courses in computer science. In order to explore options for assessing students, the working group developed a trial assessment of whether students can program. The underlying goal of this work was to initiate dialog in the Computer Science community on how to develop these types of assessments. Several universities participated in our trial assessment and the disappointing results suggest that many students do not know how to program at the conclusion of their introductory courses. For a combined sample of 216 students from four universities, the average score was 22.89 out of 110 points on the general evaluation criteria developed for this study. From this trial assessment we developed a framework of expectations for first-year courses and suggestions for further work to develop more comprehensive assessments.

Multi-Institutional Computer Science Education Research: A Review of Recent Studies of Novice Understanding

“…Those students with a DoC of 1 who turned in an empty file did not mention factors related to the process of solving the exercise and often attributed blame to factors outside their control. In contrast DoC level 5 students "competently described the difficulties they experienced in the process of creating a solution" [13].…”

“…Its "disappointing results" [13] perhaps quantified what many educators' felt was true -that students completing introductory programming courses can't program at the level we expect they should. Students at the end of their first or second semester of programming instruction were given a restricted-time (1 or 1.5 hours), lab-based programming task -a type of charette assessment.…”

What about a simple language? Analyzing the difficulties in learning to program

“…Programming has traditionally been considered a difficult topic, and research has shown that novices face several problems when learning to program (Spohrer and Soloway, 1986;Lister et al, 2004;McCracken et al, 2001;Robins, Rountree, & Rountre, 2003). Winslow (1996) refers to Linn and Dalbey, who have suggested an ''ideal chain'' for learning to program: (1) learn one language feature at a time (both syntax and semantics), (2) learn to combine the newly learned feature with previously known skills, and (3) develop general problem-solving skills.…”

Teaching the novice programmer: A study of instructional sequences and perception

“…The literature contains many studies citing the poor performance of students after completing their first programming courses (e.g., Kurland et al 1989;Linn and Dalbey 1989;McCracken et al 2001;Putnam et al 1989;Soloway et al 1982;Winslow 1996). Programming courses are also regarded by students as difficult and have high dropout rates (Robins et al 2003).…”