Three different methods for predicting the Seebeck coefficient and power generation of a commercial available thermoelectric cooler (TEC) module are used and compared to the experimental data. Method 1 and 2 are developed based on mathematical models and Method 3 is established in terms of experimental measurements. Method 3 considers the effect of cooling condition, whereas Method 1 and 2 don't. Two different temperatures at the cold side of the TEC module are also considered to account for the influence of cooling condition on the performance of the TEC. The power generation of the TEC module with low-temperature cooling is at least 5% higher than that with normal cooling. Method 3 gives the best prediction in open circuit voltage and power generation. Basically, the three methods are able to evaluate the properties and performance of a TEC easily, thereby providing useful tools for designing and constructing a TE generation system.