Recent empirical studies of software have shown a strong correlation between change history of files and their fault-proneness. Statistical data analysis techniques, such as regression analysis, have been applied to validate this finding. While these regression-based models show a correlation between selected software attributes and defect-proneness, in most cases, they are inadequate in terms of demonstrating causality. For this reason, we introduce the Software Development Profile Model (SDPM) as a causal model for identifying defect-prone software artifacts based on their change history and software development activities. The SDPM is based on the assumption that human error during software development is the sole cause for defects leading to software failures. The SDPM assumes that when a software construct is touched, it has a chance to become defective. Software development activities such as inspection, testing, and rework further affect the remaining number of software defects. Under this assumption, the SDPM estimates the defect content of software artifacts based on software change history and software development activities. SDPM is an improvement over existing defect estimation models because it not only uses evidence from current project to estimate defect content, it also allows software managers to manage software projects quantitatively by making risk informed decisions early in software development life cycle. We apply the SDPM in several real life software development projects, showing how it is used and analyzing its accuracy in predicting defect-prone files and compare the results with the Poisson regression model.