Fertility has been falling continuously since the late 1980s, but has been rebounding in Ghana, and, unlike in Latin America and Asia, the country has not reached the post-transition phase. While the distal and proximate factors associated with this stall have been described, there is less research understanding how the type of place of residence affects fertility rates and patterns in Ghana in the long term. This article uses data from the women's files of the 2003, 2008 and 2014 Ghana Demographic and Health Surveys (GDHS) to explore urban-rural differences in children ever born (CEB) and its proxitive (marital status, use of contraceptive, experience of abortion, postpartum amenorrhoea) and distal (age, education, occupation, religion, ethnicity, wealth quintile, region) correlates, in order to consider how this can be explained. In IBM SPSS Statistics version 20.0, univariate, bivariate (chi-square/cross-tabulation) and trivariate analyses were performed. Results indicate that rural women have consistently higher mean CEB over all three survey periods and by age, education, occupation, religion, ethnicity, wealth quintile, region, marital status, contraceptive use, abortion experience or amenorrhoeic status when compared to urban women. The gap between rural and urban, however, contracted slightly over the time period, in line with the spread of modern reproductive values from urban regions to rural backwaters. The study findings suggest that fertility differentials remain but are slowly narrowing in Ghana and urges the policy of fertility reduction to explicitly address rural underdevelopment as well as urban poverty.