Example programs are well known as an important tool to learn computer programming. Realizing the significance of example programs, this study has been conducted with a goal to measure and evaluate the quality of examples used in academia. We make a distinction between good and bad examples, as badly designed examples may prove harmful for novice learners.
In general, students differ from expert programmers in their approach to read and comprehend a program. How do students understand example programs is explored in the light of classical theories and models of program comprehension. Key factors that impact program quality and comprehension are identified. To evaluate as well as improve the quality of examples, a set of quality attributes is proposed. Relationship between program complexity and quality is examined.
We rate readability as a prime quality attribute and hypothesize that example programs with low readability are difficult to understand. Software Reading Ease Score (SRES), a program readability metric proposed by Börstler et al. is implemented to provide a readability measurement tool. SRES is based on lexical tokens and is easy to compute using static code analysis techniques. To validate SRES metric, results are statistically analyzed in correlation to earlier existing well acknowledged software metrics.
Source: Umeå University
Author: Abbas, Nadeem