The aim of this study is to make a awareness to enhance the rate of adult literacy specially of the rural area in Bangladesh. To fulfill the purpose of this study, age specific adult literacy rates (ASALRs) for male, female and both sexes population of rural area in Bangladesh are estimated as well as to fit some mathematical models to these estimated ASALRs and their forward cumulative distribution. For this, the data are taken from Population Census 2001. Quasi Newton Method is employed to fit these models. F-test, t-test and cross validity prediction power (CVPP) are also used to check the accuracy as well as validation of the model. In this study, it is found that age specific adult literacy rates for male, female and both sexes population of rural area follow third degree polynomial model. And forward cumulative distribution for male, female and both sexes follow second degree polynomial model. It is observed that models are well fitted in accordance with t-test, F-test and CVPP. The proportions of variation of these models are more than 96%.