In this study a model is conceptualized to measure the child mortality under different parity of women such that a better strategy can be formulated to bring down mortality rates. In the estimation of probability of child mortality some socio demographic variables are taken in consideration. The estimates are obtained under Bayesian procedure. Two different models are formulated for it and model fitting is observed by graphical approach along with the chi square test. First model is betabinomial and second is binomial regression model. Second model shows the better fit on the data. The estimate of probability of child mortality at higher parities namely, parity 3, parity 4 and parity 5 were obtained as 0.06, 0.09 and 0.13 respectively on the basis of the second model.