既存文獻對於廠商技術擴散的研究，多以「雙元模型」與「期待模型」來探討廠
商間對新技術採用的決定因素以及等待時間的長短。本文旨在對傳統模型應用在創新擴散早
期階段的適用性提出質疑。我們以光電技術在傳統工業廠商間擴散的情況為研究對象加以驗
證，並同時探討光電技術擴散的決定因素。實證結果顯示：廠商成立久暫，成長率，研究發
展活動及資訊為影響廠商採用光電技術以及等待時間的重要決定因素，而資訊則為影響決策
者計劃採用的唯一因素。模型中的解釋變數對不同狀態之影響有顯著性差異，足以推定傳統
模型處理光電技術的早期擴散是值得商榷的。

Studies of technology diffusion have traditionally applied "binary
model" and "waiting model" to study the determinants of adoption and the length
of waiting time. This paper is aimed at questioning the appropriateness of
applying these traditional models to study the early stage of technology
adoption. Instead, "trinomial logit model" and "Tobit model" are used in this
study. Empirical results indicate that firm's age, growth rate, R&D activity,
and information significantly influence the firm's decision-making on adopting
the optoelectronic technology and the length of waiting time, whereas
information is the only deteminant for the firm's decision on planning to
adoption.