Organizational learning (OL) has been widely acknowledged as vital for contracting organizations to sustain continuous improvement and competitive advantages in a turbulent construction market. In this connection, this topic has gained much attention of construction researchers in recent years. Despite the fact that previous studies have made great progress towards understanding the essence of OL in construction, research contributions from these studies are yet to be reviewed in a holistic manner. A critical review on the topic of OL found that construction researchers primarily conceptualize OL in construction at either an intra-organizational or inter-organizational level. Intra-organizational learning (Intra-OL) has been defined as a process of applying the imbibed knowledge which is integrated from individuals within an organization for performance improvement. Inter-organizational learning (Inter-OL) has been defined as learning that occurs among contracting organizations. It has been suggested from previous studies that the practice of intra-OL can be evaluated by the practice of 3 OL types: Single-loop learning (SLL), Double-loop learning (DLL) and Deutero-learning (DeuL). Research findings from previous case studies also revealed that the contracting organizations’ engagement in inter-OL is vital for their successful practice of the 3 OL types. In particular, some construction researchers advocated that inter-OL can be facilitated by providing performance feedback to contracting organizations. Thus, various Project Monitoring Systems (PMS) which provide different types of performance feedback for contracting organizations have been developed. Nevertheless, the effect of practicing the OL types on performance was rarely backed with quantitative analyses. Furthermore, support on the link between the engagement of inter-OL and the practice of the OL types is mainly anecdotal. In these contexts, in recent years, researchers have begun to question the extent to which performance change can be attributed by the effect of practicing intra-OL and engaging in inter-OL. Furthermore, they pinpointed that evidence of the effect of OL on performance improvement was rarely supported by empirical evidences. In addition, they argued that the performance change may be affected by a number of learning dimensions. Nevertheless, not many studies have been conducted to examine the effect of these learning dimensions on performance of the contracting organizations in a holistic manner. In response to these observations, this study seeks to: 1. examine the effects of practicing intra-OL on performance improvement; 2. examine the link between intra-OL and inter-OL and how this may affect performance of the contracting organizations; and 3. identify the learning dimensions and explore how they may lead to different performance outcomes. To accomplish research objective (1), a conceptual model was developed. The model hypothesized that performance improvement of contracting organizations is positively correlated with their practice of the 3 OL types: SLL, DLL and DeuL. Furthermore, the successful practice of the 3 OL types is backed with their engagement in inter-OL. The engagement of OL can be identified by 8 attributes: seeking improvement methods from others; addressing changes persistently; revising the way of working; seeking advice and suggestions from others; openness; revising long term strategy; the development of project monitoring system; and the presence of incentive scheme. A questionnaire survey was conducted to collect data for the testing of the hypotheses. A statistical technique called Structural Equation Modeling (SEM) Analysis was employed. Furthermore, the effects of each of the inter-OL attributes on energizing the practice of OL types for performance improvement were analyzed by employing Multiple Moderated Regression (MMR) Analysis. From the results of the SEM analysis, it was found that the practice of SLL and DLL was positively correlated with performance improvement. Furthermore, a significant correlation was found between the engagement of inter-OL and the practice of the 3 OL types. Nevertheless, the study did not find significant positive effect from practicing DeuL on performance improvement. From the results of the MMR analysis, ‘The development of project monitoring system (PMS)’ was found to be the most versatile inter-OL attribute which significantly moderates the effects of practicing SLL and DLL on performance improvement. To accomplish research objective (2), another questionnaire survey was conducted. Respondents were asked to evaluate their performance as well as practice of SLL and DLL upon the receipt of the performance feedback. MMR was used to examine (i) the effects derived from different types of performance feedback on performance improvement; and (ii) how such effects are contingent on the practice of SLL and DLL by the contracting organizations. It was found that when performance feedback in terms of time, cost and quality was received by the contracting organizations, the moderating effect of practicing SLL and DLL on performance improvement became significant. Nevertheless, insignificant moderating effects were found when ‘Updated forecast profit VS original anticipated profit’ and ‘Actual VS anticipated influence to the environment’ were used as the performance feedback. To accomplish research objective (3), an analytical approach to demonstrate the effect of learning on performance change of the contracting organizations was developed in this study. Data was collected from a project monitoring system called Performance Assessment Scoring System (PASS). The system records the contractors’ performance in the public housing projects in Hong Kong at quarterly intervals. The learning effects of the contracting organizations were tracked by employing the Least Square Curve Fitting Analysis (LSCFA) which fitted the collected data with six theoretical learning curve models: log-linear, 2- parameter exponential, 3- parameter exponential, 2-parameter hyperbolic, and 3-parameter hyperbolic models. It was found that the collected data fitted with the 3-parameter hyperbolic model with low errors. Furthermore, by examining the contractors’ performance change in perspective of the underpinning theory of the 3-parameter hyperbolic model and the project features, four clusters of contracting organizations (Competent Learners, Incidental Learners, Indifferent Learners and Optimizers) which exhibited different patterns of performance change were identified. The study then progressed to explore the possible learning dimensions which lead to the different performance change patterns. The third questionnaire survey was then conducted for this study. Contracting organizations from the four clusters as identified above were asked to evaluate the effects of various learning dimensions on their performance change in the project. A statistical method called Multivariate Discriminant Analysis was applied for data analysis. It was found that the Learning Motivation and the learning capability (in terms of practicing DLL and DeuL) of the contracting organizations significantly affected their performance change patterns. In sum, this study identifies that intra-OL is vital for contracting organizations to attain performance improvement. Furthermore, the empirical work supports that contracting organizations’ engagement in inter-OL fosters their practice of the OL types. In particular, the performance feedback in terms of time, cost and quality energizes the contracting organizations’ practice of SLL and DLL. The importance of designing a PMS with an OL dimension so as to make performance improvement sustainable is thus highlighted. This study also demonstrates the use of an analytical approach to track the effect of learning from a properly devised project monitoring system. The findings of this study highlight some interesting performance change patterns displayed by the contracting organizations. Four clusters of contracting organizations were identified, namely, Competent Learners, Incidental Learners, Indifferent Learners and Optimizers. Competent Learners are capable to sustain continuous performance improvement during the course of the project. Incidental Learners typically achieve performance improvement at the initial stage of the projects. Nevertheless, such improvement cannot be sustained. Indifferent Learners was found to have continuous performance deterioration. Optimizers’ performance was maintained at a level so that they would not lose out in future work opportunities. Furthermore, learning dimensions that may affect the performance of the contracting organizations were examined. Learning motivation and Practice of DLL and DeuL were found as the learning dimensions which significantly affect the performance change patterns of the contracting organizations.