Applying Rasch Analysis Modeling to Assess the Reliability and Validity of Change Order Impacts (COI) on Performance of Libyan Construction Projects

الملخص

Change order is a set of instructions that allow occurring modifications or changes on an earlier agreement, changes have become one of main and serious problems in construction projects, such changes may affect negatively on original scope, cost, and schedule of work or whole performance of the project. The current study is concerned with producing empirical evidence on the validity and reliability of the items that will be used in the questionnaire to evaluate the effects of changes on the performance of construction projects. In North West of Libya, 104 questionnaires were distributed between professional people from consulting firms, construction firms and government organization which associated to building construction projects. Analyzing was done by Rasch analysis Model via Mini steps software 5.8.0.0. The outcomes indicated that (PMC) PT-measure correlation value for 14 items range from 0.24 to 0.54 for the construct of change order impacts, except three items were -0.03, 0.01, and 0.09, which means below 0.19. In unidimensional measure, raw variance explained was 55.9% with the unexplained variance in the first factor was 7.8%. Meanwhile, the findings of the summary statistic indicate that item’s reliability for constructs was 0.99. Otherwise, the items separation index of change order effect construct was (8.86). With exception to the three items (EQ_CP5, EQ_SP14, and EQ_SP13), The analytical discussion suggests that the 14 items on questionnaire effectively measure the concept of change orders impacts (COI). These analytical measures support the validation of the questionnaire's construct validity, making it suitable for further analysis within the conceptual framework. Such findings can be employed in future research to develop advanced models that provide in-depth explanations of these effects and their interrelationships, as well as their connections with other factors.