An Evaluation of Software Measurement Processes in Pakistani Software Industry

Abstract

Implementing a successful measurement process is a challenging task. Most measurement studies report measurement models, experiences, and lessons learned based on pilot projects or case studies in a limited scope to overcome specific challenges. This paper identifies 14 basic measurement practices and proposes a model of 18 success factors for implementing measurement processes with respect to the identified measurement theories in our systematic literature review (SLR), i.e., A systematic literature review on software measurement programs, by Tahir et al. , 2016. In addition, a survey is conducted to evaluate the state of measurement practices and to validate the proposed model based on the feedback from 200 software professionals working in Pakistani software industry. The state of measurement practices in the industry is mostly not according to the identified measurement theories in the SLR. For instance, more than 50 measurement models reported in the literature but only 10% software organizations follow any measurement model. 75% of organizations do not follow any measurement standard. 80% software organizations do not use any measurement tool. The proposed model is validated by applying structural equation modeling on the survey data. Furthermore, among 18 success factors, it is statistically significant that Pakistani software professionals strongly believe in necessity of three factors for successful implementation of a measurement process, i.e., synchronization between measurement process and software process improvement, use of measurement standards, and use of measurement models. In addition, they also believe that a successful measurement process will improve prediction, monitoring and management of software projects, and support in achievement of individual and organization-wide objectives. Software organizations might consider this paper in planning and improving their measurement processes.