June 24, 2017
June 24, 2017
June 28, 2017
Computers in Education
Software Engineering (SE) students learn various concepts in academia that prepares them for working in software industry. To fulfil the demand of delivering quality software, industries focus on inspection technique, introduced by Fagan at IBM in 1976, for early detection and removal of faults that saves rework effort and cost. Due to the importance of inspections, academic institutions have begun training students on inspection technique. This helps them to understand the nature of faults made during development and help them avoid committing the same defects. Despite the importance of inspections as an effective verification technique, training students on becoming effective inspectors is especially hard because the background factors (education, experience, personalities) of inspectors do not impact their individual fault detection effectiveness. On that note, our prior results showed that students tend to perceive & process information presented in Natural Language (NL) requirements document in different ways and is known as Learning Styles (LS). To gain insights into the reading patterns of individual inspectors during the requirements inspection, this research utilizes the eye tracking concept along with the psychological research on LS to identify the reading pattern of students as they inspect NL requirements document. The paper reports results from an academic study where inspection data of thirty-nine students were used to understand how educator can improve their inspection training technique. First, LS of each student was collected and students were trained on fault-checklist based inspection technique with sample requirements to guide inspections. Next, each student inspected requirements document (with seeded faults) and reported faults. The results were handed back with seeded fault list to introspect which helps students to understand inspection technique. Next, each student performed inspection on a second document (with seeded faults) in an eye tracking laboratory settings. Each student read requirements document on a computer monitor and eye movements were recorded by an eye tracker sitting at the bottom of the monitor. The faults reported were checked for fault positives by one of the researchers to generate final fault list for each student. The eye movement data was utilized to visualize scanpaths (linear/random reading trend) of students with fixations (where eye movements are relatively stationary) on the document along with the fixations at fault region. The data was then analyzed to evaluate the reading patterns in general and with different LS preference. Results show that inspection performance is significantly affected by the fixations especially at the fault regions. Results also showed that students had higher degree of linear reading pattern than random jumps during inspection and LS preferences had varying eye movement factors that affected inspection performance. Results provided insights into the design and execution of inspection training program that could be utilized to train students to focus more at fault prone areas for improved inspection outcome. Results also provided insights into the specific sections of requirements document that demonstrated regressive fixations and how the improvement in the requirements organization can improve the readability of requirements which in turn would improve their inspection performance.
Goswami, A., & Walia, G. S., & Padmanabhan, G., & McCourt, M. E. (2017, June), Improving the Requirements Inspection Abilities of Computer Science Students through Analysis of their Reading and Learning Styles Paper presented at 2017 ASEE Annual Conference & Exposition, Columbus, Ohio. 10.18260/1-2--28498
ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2017 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015