A success rate analysis of methodologies in software development: A correlational and comparative study

The high failure rate of software projects impels researchers to explore factors that could improve the project success rate by minimizing development costs, shortening delivery schedules, and meeting the users' requirements. This study focused on the correlation between each factor of the triple constraints (development cost, delivery timing, and product quality) and project success, as well as the differences of each factor under two software development methodologies, traditional structured and object-oriented. The existing survey instrument was utilized fo r the online survey, and data collection was conducted through SurveyMonkey. The respondents consisted of 224 project managers in the United States supervising the software projects with medium-to-high complexity levels. Among them, 139 respondents selected the traditional structured method, while 85 chose the object-oriented method. Due to the skewness of the data, Spearman's Rho was utilized for the correlational study, and analysis of variance (ANOVA) nonparametric Kruskal-Wallis hypothesis tests were performed to test the null hypotheses for the difference between two development methodologies impacting the triple constraints and project success. The results of the study indicated that there was a weak correlation between the development cost and project success, and a moderate correlation between the delivery schedule or the users' requirements and project success. The results of the study also revealed that the correlational relationship between the users' requirements and project success was stronger in the object-oriented development method than in the structured development method. The hypothesis tests supported the null hypotheses that there was no difference between traditional, structured development methodology and the object-oriented development method impacting each factor of the triple constraints and project success. The knowledge discovered could help project managers better estimate budgets, set delivery schedules, and meet users' requirements without considering which development methodology could adversely affect project success.