Finding reliable ways of assessing female mate choice is an essential prerequisite to understanding variation in female preferences and its impact on sexual selection. The zebra finch, Taeniopgyia guttata, is a classic avian model species for investigating mate choice and sexual selection, but to date virtually all work has been carried out using domesticated birds, and there are inconsistencies in the findings between studies. We tested three different methods for measuring female preferences in this species using both wild and domesticated birds. The first method was a traditional two-way mate choice chamber, where we measured the time spent with each male. The second method was a no-choice chamber where a single male and female were placed in the same cage for 5 min, such that the pair could physically interact, and the female's sexual response was recorded. The third method was an aviary set-up, comprising eight birds (four males and four females), and we recorded the number of pair-bonding behaviours observed over a 24-h period. In the aviary, we found that birds formed pair bonds almost exclusively with their own type (wild males paired with wild females, and domestic males paired with domestic females). In the two-way choice chamber, this assortative mating preference was only shown by domestic females, and in the no-choice test, only wild females showed this assortative mating preference. We discuss the possible reasons for these differences between mate choice tests and make recommendations for future studies on mate choice.