Regression Testing Goals and Measures: An industrial approach
2018 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE credits
Student thesis
Abstract [en]
Context: When a software is modified, regression testing is performed to ensure the behaviour of software is not affected because of those modifications. Due to frequent modifications, the regression testing became challenging. Although there are many regression testing techniques are developed in the research, they are not incorporating in the industry. This is because of the differences in regression testing goals and measures in research and industry. The current context of this study is to identify the regression testing goals and measures in the research and industry perspectives and to find the differences and similarities in both perspectives.
Objectives: The primary objective of this study is to identify the similarities and differences in regression testing goals and measure from research and industry perspectives. Additionally, in this study, a general adapted goals list is presented.
Methods: A mixed method approach is used for this study. A literature review has been used to identify the regression testing goals and measures in research. A survey is used to identify the regression testing goals and measures in the industry. Semi-structured interviews and online questionnaire are used as data collection methods in the survey. Thematic analysis and descriptive statistics are used as data analysis methods for the qualitative and quantitative data.
Results: A literature review is conducted using 33 research articles. In the survey, the data is collected from 11 semi-structured interviews which are validated with 45 responses from an online questionnaire. A total of 6 regression testing goals are identified from the literature review and 8 goals are identified in the survey respectively. The measures used to evaluate these goals are identified and tabulated.
Conclusions: From the results, we observed the similarities and differences in the regression testing goals and measures in industry and research. There are few similarities in goals but the major difference is the priority order of these goals. There are various measures used in research but very fewer measures are incorporating in the industry. The respondents from the survey implied that there is a need for generic adaptive goals. Further, a general list of goals is presented.
Keywords: Regression, Regression testing, Goals, Objectives, Measures, Metrics.
Place, publisher, year, edition, pages
2018. , p. 81
Series
Blekinge Tekniska Högskola Forskningsrapport, ISSN 1103-1581
Keywords [en]
Regression, Regression testing, Goals, Objectives, Measures, Metrics.
National Category
Software Engineering
Identifiers
URN: urn:nbn:se:bth-17249OAI: oai:DiVA.org:bth-17249DiVA, id: diva2:1262345
Subject / course
PA2534 Master's Thesis (120 credits) in Software Engineering
Educational program
PAAXA Master of Science Programme in Software Engineering
Presentation
2018-10-04, J1650, Blekinge Tekniska Högskola, 371 79, Karlskrona, 14:00 (English)
Supervisors
Examiners
2018-11-122018-11-112018-11-12Bibliographically approved