近年來台灣地區年輕人呈現遲婚與不婚現象，不僅導致生育率的下降，「頂客族」名稱也成為一個新族群的標記。究竟「頂客族」有多少?本研究使用1992年至1998年「臺灣地區家庭收支調查」來推佑，並嘗試釐清以戶為單位的調查資料，「頂客族」可能的意涵。夫婦均在工作且家戶中只有夫婦二人，作為「頂客族」之操作定義，1998年資料顯示，「頂客族」，佔總家戶約百分之三點五，約為22萬戶。值得注意的，透過家戶型態來推，估「頂客族」，呈現一批夫妻年紀較輕、教育程度高、多為白領階級之「現代頂客族」，以及另一批夫妻年紀較大、教育程度偏低、且多為務農者之「空巢頂客族」，前者約佔總家戶之3%，為19萬戶。可是，從都市化的觀點，「現代頂客族」在台灣應只是方興未艾，尚未形成隨地區都市化程度而增，加的趨勢。若以個人或家庭特質來解釋成為「現代頂客族」的可能性，本研究採用對數迴歸多變項分析，結果顯示，丈夫年齡越高與恩格爾係數越，高越不利於成為「現代頂客族」，而妻子為專業白領工作者成為「現代頂客族」的機會高，個人與家庭特質的解釋在1992與1998年資料中都得到支持。

One of the main demographic features in recent years in Taiwan is the decreasing birth rate. The young are more likely to marry later than older cohorts and to delay pregnancy. This research focuses on investigating the number of DINKs (couples with double income and no kids) and its distribution in Taiwan. Using data from the 1992-98 Surveys of Family Income and Expenditure, the authors identified two kinds of DINKs differentiated on the basis of their living arrangements. The Empty Nest DINK category consists mainly of older couples with low education levels and largely engaged in agriculture. Modem DINKs are mainly relatively young couples, most with white collar jobs. The latter, because of their low birth rates, captures our concern. The number of Modem DINKs increased slightly between 1992-1998, when it accounted for 3 percent of total households. Multivariate analysis shows that wives with white-collar jobs, or households with lower Engler coefficient of expenditure style are more likely to become Modem DINKs. That the net effect of urbanization was not significant in 1992 but was in 1998 indicates that Modem DINKs are becoming a more significant category.