本文從家計生產模型出發，將Ben -Porath與Welch探討性別偏好之風險模型予以簡化，建立可供檢定之命題，隨接利用台灣之家計資料，分析台灣地區是否存在性別偏好？如果答案是肯定的，則進一步分析性別偏好屬於何種型態？性別偏好之原因又是何？最後，也是最重要的，則在檢視性別偏好對生育決策的影響。 實證結果發現：(1)台灣地區「至少一個男嬰」的偏好明顯；(2)雖然「至少一個男嬰」偏好形成原因中，源於嗜好之不同與源於養育男女相對價格之不同均有其影響，惟異質嗜好所帶來之影響程度較大；(3)「至少一個男嬰」之偏好對生育率有著顯著的正影響。

The quality of children has been mostly discussed in as far as it can affect fertility. The sex of children is a particularly important subject in this context. Because a child’s sex is stochastically determined, Ben-Porath and Welch used the house­hold production approach to identify risky decision points. The purpose of this paper is two-fold. One is to simplify the Ben-Porath and Welch’s model. The other is to use KAP data in Taiwan to examine the causes of preferring to have at least one boy and to test in what direction and the degree to which such preference affects fertility. The results can be summarized as follows: (1)Preferring to have at least one boy is significant in Taiwan, (2) The data support both the assumption of inelastic demand and mild learning, (3) Concerning the dummy variable for the existence of at least one boy preference, it can be shown that the regression coefficients for each of family structure, wife's education and husband's occupation were statistically significant at the .01 significance level, and (4) When the impact of “at least one boy preference”on completed family size is concerned, the regression coefficient for the existence of at least one boy preference is positive as expected and also the most significant variable included.