The poor economic performance and growth of ECOWAS economies have become so challenging and this has necessitated research and discussions. Experts have argued that the relationship between trade and economic growth in the ECOWAS region is unclear and the question of whether trade promotes growth and development is controversial. In this paper, an attempt is made to analyse the impact of trade on economic growth both from a theoretical perspective and using econometric evidence from ECOWAS countries. In particular, an attempt is made to econometrically test the hypothesis of a positive relationship between trade and growth in ECOWAS countries during the 1990-2013 period. This is done using panel data regression analysis. Utilizing data for the 15 ECOWAS countries, a log-linear regression equation of per capita real GDP is fitted on exports, the exchange rate, investment, human capital, inflation and population growth. For completeness and to test the robustness of the econometric results, 4 estimators (pooled OLS, Fixed effects model, Random effects model, and dynamic panel regression model) were utilized. The dynamic panel data estimator is preferred as it is able to handle the problems arising from 'endogeneity' or reverse causality. All the 4 estimated regression equations had high coefficients of determination and F-statistic. In all the equations, exports, exchange rate and investment were significant determinants of per capita real income growth. Exports were consistently positively related to growth, thus confirming the hypothesis of trade having a significant positive impact on economic growth in ECOWAS countries.