The anomalous depression of the maximum plasma frequency in the ionosphere (foF2) appears significantly within 1-5 days before the M ≥ 5.0 earthquakes in the Taiwan area during 1994 - 1999 (Liu et al. 2003). In this paper, we propose two statistical tests against the foF2 anomaly as a candidate for a precursor of earthquakes based on criteria including the success rate, alarm rate, probability gain, and R score. One statistical test is conducted to investigate the significance of the observed foF2 anomalies related to the recorded M ≥ 5.0 earthquakes in the Taiwan area during 1994-1999. The other statistical test is designed to compare the foF2 anomaly-based method with competitive alternatives for predicting the earthquakes under study. The involved alternatives are a naive prediction based on a coin-tossing experiment and a simple prediction method constructed from the current M ≥ 5.0 earthquakes catalog during 1994 - 1999. The simulation results indicate that, contrast to possible foF2 anomalies, the observed foF2 anomalies are signiflcantly earthquake-related. Moreover, comparing with the alternative predictions under study, the foF2 anomaly remains valid for temporal alarming of the M ≥ 5.0 earthquakes in the Taiwan area during 1994 - 1999.