This study examines the effect of different melody designs for children's songs on preschoolers' language comprehension. The dependent variable in this study is the language comprehension score obtained by the preschoolers. Pitch and tempo differences of the melody designs for the children's songs are the independent variables. Two experimental designs are conducted separately for examining the effects of right/wrong pitches and the right/wrong tempos on the preschoolers' understanding to the song scores. The researcher is able to design two versions (right or wrong) of the pitch for one song as well as two versions (right or wrong) of the tempo for another song. In addition, according to the literature review, an instrument can be easily developed by the researcher by using Sentence Variation Technique to measure the children's language comprehension. As the data collection is completed, ANOVA is conducted to analyze the preschoolers' obtained scores in different experimental designs. As a result, the study finds that: (1) Pitch differences of the melody designs for the children's songs have significant differences on the preschoolers' obtained language comprehension scores; (2) Tempo differences of the melody designs for the preschoolers have no effect on their language comprehension. Finally, this study concludes that children's song composers must accurately design the right pitch for the lyrics of the children's songs. Thus, the preschoolers are able to effectively comprehend song lyrics. It is also suggested that the composers can be brave enough to design special tempos for the children's songs.