The corrected midparental height method was introduced by Tanner in 1970 (Tanner method) and is commonly used to estimate target height in children to evaluate the effectiveness of growth-promoting therapies. It has not been established if the equation used to compute target height should be the same for children with short, normal, or tall parents. In this study, we examined the predicted target height values by parental heights in a large population-based study (n = 2402). A simple linear function of midparental height (x) was proposed to estimate target height (y): y = 45.99 + 0.78x (boys), y = 37.85+0.75x (girls), with a 95% predicted interval of about +/-10 cm. The prediction model was similar for boys and girls in SD scores (SDS), and was not affected by assortative mating or difference in parental heights. The model may underestimate the potential stature by about 2 cm for children with midparental height below -2 SDS, or 163 cm. In comparison, the Tanner method may lead to a 6-cm error in underestimating target height for these children. The function would be a better choice than the Tanner method for estimating target height in the clinical evaluation of growth promotion treatments because it is common that short children also have short parents. Children with very short parents will usually be much taller than their parents in adult stature, and we believe that a different function should be developed. The results support the proposed nondominant, non-sex-linked, polygenic inheritance in stature. The estimated heritability values were 0.75-0.78 in cm or 0.55-0.60 in SDS.