標題: | 線上購物零售業之消費者退貨影響因素分析 An Analysis of the Factors Affecting Consumers' Propensity to Return in E-Retailing |
作者: | 孫馨璍 馮正民 黃昱凱 Cheng-Min Feng Yu-Kai Huang 運輸與物流管理學系 |
關鍵字: | 線上購物零售業;退貨傾向;決策樹分類器;E-retailing;Return Propensity;Decision Tree |
公開日期: | 2007 |
摘要: | 網際網路的代表著一新興市場,電子商務成為重新整合供應者與消費者關係的新商業模式。資策會預期2008年臺灣B2C市場規模可達1385億,相較於去年度的市場規模有38%的成長。由此可見,線上購物的規模與成長不容小覷。然而,隨著市場的成熟,退貨也逐漸成為線上購物零售商所需克服的議題之ㄧ。過去的文獻探討到退貨議題,往往集中在供應商與零售商間,但隨著線上購物的興盛,零售商與終端消費者間的退貨議題,逐漸被受重視。 本研究擬針對線上購物之交易,利用實證資料以決策樹分類模型判別交易是否退貨,藉以了解何判別變數影響顧客之退貨傾向。判別變數分為三大類別,分別是顧客屬性變數、商品變數以及服務變數。 研究結果發現,變數中-商品種類、價格與配送天數能夠較有效地判別消費者退貨傾向的高低,最後,針對決策樹所歸納出的規則與變數範圍,研擬對應策略以作為網站管理者控制線上購物退貨之參考。 The internet represents a growing and huge market. The development of e-commerce is an efficient business model which enables new relationship between consumers and suppliers. In particular, the B2C market in Taiwan is expected to reach NT $138.5 billion with 38% increase in 2008. The E-retailing is obviously becoming a noticeable market. However, as the market grows and matures, “return” becomes one of the challenges for E-tailers. In the past, most of the literature on return issues focused on the wholesaler-retailer relationship. Recently, due to the advent of Internet-based retailing within the past decade, attention is shifting to the issue of returns in the retailer-consumer relationship. In this study, we use empirical data and conduct a Decision Tree model to analyze the critical variables revealing the customer return propensity. There are 3 dimensions of variables in our data set- customer demographic variables, merchandise variables and service variables. We find that three variables- category, price and delivery days could be used to distinguishing customer return propensity more effectively. In accordance with these variables, we propose some strategies for website managers to control returns in E-retailing. |
URI: | http://140.113.39.130/cdrfb3/record/nctu/#GT009536502 http://hdl.handle.net/11536/39254 |
Appears in Collections: | Thesis |
Files in This Item:
If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.