Risks are factors of uncertainty that affect human activity on various levels. Manage them is a key issue for the success of projects that may occur. In the field of software development, an area that has inherently a high degree of uncertainty and where there is participation of various professionals involved, manage risk variables has become a necessity. The present study shows the obtaintion of a categorization of risks in software development by means of exploratory factor analysis using the technique of principal component analysis (PCA - Principal Component Analysis). To apply this technique were collected estimates of severity of risk and probability of their occurrence between project managers and developers members of virtual communities on the Internet. The categorization of risk variables allows a greater understanding of their relationship and the possibility of treating them in a higher level, dealing with factors of greater extent rather than spraying efforts managing many small variables simultaneously. Seven factors were identified, and six of them showed good internal reliability: Project Management, Team Development, Scope and Requirements, Knowledge and Technological Uncertainty Relationship with the External Environment, Relationship with the Client / User and value / importance attributed for the Project. The text discusses the results and points to lines of research for improvement of the theoretical area, highlighting opportunities that may be implemented by Project Managers and Software Developers in their daily processes of risk management.
Project Management; Risk Management; Software Development; Principal Component Analysis; Information Technology
