In this study we used the data from Afghanistan Health Survey 2006. For this study, 8278 households were randomly selected in which 8281 women aged 10-49 years were interviewed by survey teams using a structured questionnaire. The information was also collected for all children aged 5 years or less from all these households. The sample includes 7843 (13.8%) children under the age of 5 years old. Literacy of mothers (ability to read), age of mother at marriage, number of children, exposure to mass media (listening to radio or watching TV) were the independent variables and BCG vaccination, initiation of breastfeeding (within first hour of life or after first hour); and use of bed net (to protect a child from Malaria) were dependent variables. Chi square and Odd Ratio test was used to test significance of the associations. Logistic Regression test was used to control for the confounders. In this study we found that those listening to radio at least once a week were more likely to start breastfeeding during the first hour of life. Those watching TV at least once a week were more likely to vaccinate their children for BCG. These associations were significant after controlling for confounders (economic status of the family and distance to health facility). The fact that why the other independent variables did not have association with BCG vaccination, initiation of breastfeeding and use of bed net can be either due to limitation of the study or there are other reasons that require further investigations.