Research Paradigm: The approach for this study is based on positivism theory with an intention to obtain evidence through focused observations and identify its likeliness with the actual data collected so that it can be generalized with the findings of other scholars who have worked on the topic (Remenyi et al., 1998). As elaborated by (Gill & Johnson, 2002), the crux of positivism philosophy revolves around the relationship of cause and its effect generated by different constructs, and the best possible exploration of these variables. Research Design: Primary data would be collected through a structured close-ended questionnaire which use to gather the responses from the stakeholders including teachers, coordinators, and school management and human resource personnel in the private schools. Therefore, the research questionnaire adapted from Dogan (2009) and Asiyai (2016) as a research instrument. Sample Size and Sampling technique: The target population for this research study consists of different stakeholders for instance: teachers, coordinators, and school management and HR personnel working in private schools of Karachi -Pakistan. Non-probability purposive sampling is being used for this study. The sample size of 400 different stakeholders from private schools would be adequate to figure out the results (Loehlin, 2004). Since the actual size of the population is not exactly known and neither accessible to conduct random sampling, therefore non probability purposive sampling is chosen. Since the sample consists of different stakeholders for instance: teachers, coordinators, and school management and HR personnel working in private schools of Karachi -Pakistan. Therefore, it is a purposive sampling. Data Collection Tool: In this study we used the quantitative research to measure the perceived effectiveness of in-service training in improving teacher’s performance with one independent variables: in-service training (Training need assessment and training methodology), one dependent variables teachers performance (Task performance and contextual performance), 03 mediators (professional skills, knowledge enhancement and work engagement) and one moderator (reward). Questionnaire has distributed in the several schools by hand or through email. Statistical Technique: Since the model consists of multiple variables with mediation and moderation model, therefore the Structural Equation Modeling (SEM) is used. The interface terms integrated with the model is measured for statistical significance via bootstrapping method. The structural equation modeling has executed by the partial least square approach.