Spatial regression models provide the opportunity to analyze spatial data and spatial Dependence. Yet, several model specifications can be used, all assuming different types of spatial dependence. This interactional Multi dimensionality lead to the problem of model identification in spatial modeling. This study is use to identify the best suited spatial model by using classical hypothesis testing approach. While Cross-sectional spatial econometric approach is use to check the spatial dependence between income and educational inequality at district level in Pakistan. Results of the analysis specify the SLM as an optimal spatial model comparative to SEM through classical approach while spatial dependence has found significant in the regional income and educational inequality data. Educational inequality significantly and positively contributed in reducing the regional income inequality due to spatial interactional effect. Although educational attainment also contributed significantly but educational distribution outer performed than educational attainment in the model. It is concluded that investing for equitable distribution of education will be very effective policy strategy to overcome the problem of regional income disparity in Pakistan. Supervisor:- Dr. Hafsa Hina