We demonstrate how a non-nested statistical test developed by Vuong (1989) can be used to assess the suitability of alternate order-of-entry assumptions used for identification purposes in empirical entry models. As an example, we estimate an entry model of McDonald’s and Burger King restaurant outlets in United States. The data set focuses on relatively small “isolated” markets. For these markets, the non-nested tests suggest that order-of-entry assumptions that give Burger King outlets a first-mover advantage are statistically preferred. Last, a Monte Carlo experiment provides encouraging results suggesting that the Vuong-type test yields reliable results within the entry model framework.