A Mapping Study on Software Process Self-Assessment Methods
Authors: Thaísa C. Lacerda, Christiane Gresse von Wangenheim, Jean C. R. Hauck
Abstract: Assessing processes is one of the best ways for an organization to start a software process improvement program. An alternative for organizations seeking for lighter assessments methods is to perform self-assessments, which can be carried out by an organization to assess its own process. In this context, the question that arises is which software process self-assessment methods exist and which kind of support they provide? To answer this question, a mapping study on software process self-assessment methods was performed. As result, a total of 33 methods were identified and analyzed, synthesizing information on their measurement framework, process reference model and assessment process. We observed that most self-assessment methods are based on consolidated models, such as CMMI or ISO/IEC 15504 with a trend to develop self-assessment methods specifically for SMEs. In general, they use simplified assessment processes, focusing on data collection and analysis. Most of the methods propose to collect data through questionnaires that are answered by managers or other team members related to the process being assessed. However, we noted a lack of information on how most of the assessment methods (AMs) have been developed and validated, which leaves their validity questionable. The results of our study may help practitioners, interested in conducting software process self-assessments, to choose a self-assessment method. This research is also relevant for researchers, as it provides a better understanding of the existing self-assessment methods and their strengths and weaknesses.
Tools for this paper
Ask questions about this paper to our AI assistant
You can also chat with multiple papers at once here.
The previous summary was created more than a year ago and can be re-run (if necessary) by clicking on the Run button below.
Look for similar papers (in beta version)
By clicking on the button above, our algorithm compares the title and abstract of this paper with all the papers of the same arXiv category that we have summarized, and finds the closest ones. You can rerun it from time to time: the results improve as our database grows.
Disclaimer: The AI-based summarization tool and virtual assistant provided on this website may not always provide accurate and complete summaries or responses. We encourage you to carefully review and evaluate the generated content to ensure its quality and relevance to your needs.