隨「唯手機族」增加，結合市話與手機之雙底冊電話調查已成趨勢。本研究探討雙底冊調查於全國家犬貓總數與寵物戶數比之推估應用，採用 112 年度調查之 10,348 份有效家戶問卷，對比樣本比例估計法與洪永泰比例估計法於三層與四層事後分層估計之推估效果。
實證顯示，樣本比例估計法有低估唯手機族之傾向；改採洪永泰比例估計法後，唯手機族占比增加，家犬貓總數及寵物戶數比亦普遍呈微幅增加趨勢，表示洪永泰比例估計法對修正唯手機族群之涵蓋偏差可能具有些許助益。模擬研究證實，洪永泰比例估計法對樣本配置變動具較佳穩定性；而家犬總數推估之變異係數則呈凹形波動特徵，其較佳之樣本配比受母體結構影響，其中高唯手機族比例之情境以市話樣本30%與手機樣本70%之配置為佳，中低唯手機族比例之情境則以市話樣本50%與手機樣本50%之配置為宜。
然而，受限於手機號碼缺乏地理區域資訊，若增加手機樣本之比例，易造成直轄市手機樣本過多、非直轄市手機樣本過少，導致縣市推估結果有較大之波動。因此，實務上採取市話樣本與手機樣本約七比三之配置，實為兼顧全國總體推估效率與縣市推估結果穩定性之折衷與考量。

With the increase in the cell-phone-only (CPO) population, combining landline and cell-phone frames in dual-frame telephone surveys has become increasingly common. This study examines the application of dual-frame surveys to estimating national dog and cat populations and the proportion of pet-owning households. Using 10,348 valid household questionnaires from the 2023 survey, we compare the Sample Proportion Method and Hong's Proportion Method under three-layer and four-layer post-stratification models.
Empirical results show that the Sample Proportion Method tends to underestimate the proportion of CPO households. Hong's Proportion Method increases the estimated CPO proportion, with the estimated national dog and cat populations and the proportion of pet-owning households generally showing slight increases. This suggests that Hong's method may provide some benefit in correcting CPO coverage bias. Simulation results show that Hong's Proportion Method is more stable under different sample allocations. The coefficient of variation of dog population estimates exhibits a U-shaped pattern. A 30/70 landline-to-cell-phone sample allocation ratio performs best under high-CPO scenarios, while a 50/50 sample allocation ratio is preferable under medium-to-low CPO scenarios. However, because cell-phone numbers lack geographic information, increasing the proportion of cell-phone samples may cause excessive sampling in special municipalities and insufficient sampling elsewhere, resulting in greater county-level fluctuations. Therefore, a 70/30 landline-to-cell-phone sample allocation ratio is a practical compromise between national estimation efficiency and county-level stability.

侯佩君（2023），雙底冊電話調查合併組合加權處理之比較分析，選舉研究，30（1），43–88。
洪永泰（2021），雙底冊電話調查的估計成效評估：以選前調查為例，選舉研究，28（2），95–126。
國家通訊傳播委員會（2022），111年度通訊市場調查結果報告，https://www.ncc.gov.tw/
國家通訊傳播委員會（2023），112年度通訊市場調查結果報告，https://www.ncc.gov.tw/
國家通訊傳播委員會（2024），113年度通訊市場調查結果報告，https://www.ncc.gov.tw/
國家通訊傳播委員會（2025），114年度通訊市場調查結果報告，https://www.ncc.gov.tw/
趙昌泰（2021），110年全國犬貓數量調查期末報告，行政院農業委員會。
趙昌泰（2023），112年度全國犬貓數量調查期末報告，農業部。
趙昌泰（2025），114年度全國犬貓數量調查期末報告，農業部。
Brick, J. M., Dipko, S., Presser, S., Tucker, C., & Yuan, Y. (2006). Nonresponse bias in a dual frame sample of cell and landline numbers. Public Opinion Quarterly, 70(5), 780–793.
Fuller, W. A., & Burmeister, L. F. (1972). Estimators for samples selected from two overlapping frames. Proceedings of the Social Statistics Section, American Statistical Association, 245–249.
Hartley, H. O. (1962). Multiple frame surveys. Proceedings of the Social Statistics Section, American Statistical Association, 203–206.