Software development practices and software maintenance performance: a field study
Sanora Ann Slaughter · 1996
Software maintenance continues to claim a large and growing proportion of organizational resources. It is thought that many maintenance problems can be attributed to inadequate software design and development practices. Poor design choices can result in complex software that may be costly to support and difficult to change. Because the effects of software design and development practices are realized over the life cycle of the software, it is difficult to determine their actual maintenance performance effects. To estimate the impact of software design and development activities on software maintenance, this study develops a theoretical framework for software maintenance performance that builds on Wood's (1986) psychological theory for task complexity. A two stage model is formulated, with software complexity as an intermediate variable that links design and development decisions and their downstream effects on software maintenance performance. To assess the external validity of the theoretical framework, the research employs econometric techniques to analyze data collected during an intensive field study. The field study was conducted over a period of three years and examined one hundred forty-seven software enhancement projects and sixty-four software applications in two commercial organizations. Results indicate that increased software enhancement effort is significantly related to higher levels of software complexity. Furthermore, increased software complexity is associated with the use of purchased software and the use of a software development productivity tool for generating code. Programming practices such as structured techniques and the use of report generators are associated with lower levels of software complexity and reduced enhancement effort. These results suggest an important link between software design and development practices and software maintenance performance.