We consider uplink WiFi data offloading in a heterogeneous network consisting of one LTE macrocell and one overlaid 802.11n WiFi hotspot to maximize the minimum energy efficiency (EE) of the users. We propose a closed-form expression to compute Nw, i.e., the number of offloaded users, and then offload the data traffic of Nw users with the lowest received SINR at the cellular base station (BS) to the WiFi network. Our simulation results show that our proposed offloading algorithm performs nearly the same as the exhaustive search and greedy algorithms, but requires much lower computational complexity. We also compare the EE performance of our proposed algorithm with naive schemes, i.e., no offloading, all offloading, and random selection, by taking into account the practical energy cost such as the transmit-power dependent energy consumption in LTE and the contention energy consumption in WiFi. We show that our proposed algorithm always outperforms naive schemes, especially when the user density in the hotspot is high.