台灣傳統農業銷售管道階層過多，影響農民收入，透過網路販賣農產品，使消費者與農民中間的阻礙變少，可增加農民的收益；4G時代的來臨，智慧型手機的蓬勃發展，使得網路購物交易量大幅的成長；2014年食安危機後，消費者愈來愈重視購買食物的來源，農產品網路商店可提供消費者更多商品資訊。作為農產品電子商務研究的雛形，本研究標的為可常溫配送的水果，本研究採用聯合分析法，探討消費者在網路商店購買水果時所考量的屬性偏好及組合，並研究在不同市場區隔下，消費者的偏好是否有差異。
研究結果顯示，受測者在網路商店購買水果時特別重視顧客評價及退貨服務，其他的屬性偏好依序是平台使用順暢度、農家知名度及獨家折扣，而獨家折扣因為運費緣故對消費者實質利益有限，影響力極小。在所有問卷受測者屬性水準的偏好上，網路商店要有提供顧客評價的資訊、退貨方式為物流業者上門收貨、農家知名度來源為親友推薦、獨家折扣為超低價組合包。許多分群呈現不一樣的結果，水果電子商務業者可根據這些資訊發展不同差異化的服務，吸引消費者從實體通路轉向網路通路購買水果。
另外，研究發現不同於一般網路販售商品，消費者重視價格因素及避免退貨的可能性，顧客在水果網路商店重視顧客評價及退貨方式的便利性，卻不重視折扣促銷，這是因為水果非標準化商品、保存有時限性及運送途中容易受損的特質。為了讓消費者在網路商店安心購買水果，業者宜在網路商店設計簡單明瞭的顧客評價系統及加強物流配送及退換貨的效率。

SUMMARY

In Taiwan, multi-channels of agricultural products selling result to the deficiency of farmers’ income. Through Internet, the path from farmers to customers becomes shorter and farmers could enhance revenues. Along with the technology develops, it is prevalent that consumers shop with smart-phones today. Therefore, internet shopping grows rapidly with the convenience of portable devices. However, agricultural e-commerce in Taiwan develops slowly. The consumers are used to choosing agricultural goods on the spot. In 2014, a crisis of food safety aroused the public’s awareness of food resources. Agricultural products sold in on-line shops could offer customers more product information to reduce risks of contaminated food. As a preliminary research of agricultural e-commerce, fruit was chosen as the research objective. This study conducted conjoint analysis to inspect what attributes influence customers to purchase at fruit online shops. The results showed that customer review and product return occupied great importance on the consumers’ minds, and other attributes listed in sequence below: smoothness of shopping platform, farmhouse reputation, and exclusive discounts. Notably, discounts barely influence customers to buy fruits in on-line shops. According to results, fruit e-commerce owners should optimize the design of customer reviews, improve the service of product return, cooperate with honest farmers and maintain good quality of fruits. Therefore, it naturally creates web WOM which attracts more consumers to buy fruits from off-line to on-line.

INTRODUCTION
According to the data of Market Intelligence & Consulting Institute, Taiwan's B2C online shopping market scale was expected to reach NT 413.9 billion in 2014, and maintained a high growth rate about 15% in the coming years. Along with the technology develops, it is prevalent that consumers shop with smart-phones today. The report also mentioned consumers’ future mobile shopping lists, and the top six categories included "activity tickets (41%), travel tickets (39.3%), 3C products (28.1%), food (23.3%), mobile games (22.1%), and clothing accessories (22%).” Although food selling is popular in internet shopping, agricultural e-commerce in Taiwan develops slowly. Agricultural products are easily damaged in the transportation, and the consumers in Taiwan are used to choosing agricultural goods on the spot. However, a crisis of food safety aroused the public’s awareness of food resources in 2014. With the transparency of Internet, agricultural products sold in on-line shops could offer customers more product information to reduce risks of contaminated food. On the other hand, multi-channels of agricultural products selling result to the deficiency of farmers’ income in Taiwan. Through Internet, the path from farmers to customers becomes shorter and farmers could enhance revenues.
From literatures, agricultural e-commerce encountered some plights and continued adjusting managements to cater for customers. When customers purchase fruits at on-line shops, some factors are taken into consideration. However, it is difficult to customize fruit on-line shops for individual because everyone has different needs. As a preliminary research of agricultural e-commerce, fruit was chosen as the research objective. The objective is to find out what specific attributes and factors of fruit on-lines shops can satisfy consumers.

MATERIALS AND METHODS
This study conducted conjoint analysis, a technique used by marketing managers to gain an insight into consumers' preferences for products and services and to predict buyers’ behaviors, to distinguish what attributes customers favor when they buy fruits at on-line shops.
There were five steps in this study: (1) Collect attributes from the literatures. (2) Reduce the number of attributes by the pretest. (3) Design the formal questionnaire. (4) Verify and analyze the collected data. (5) Make conclusions and suggestions.
According to literatures, it could sort out nine attributes of fruit e-commerce at first and then reduce to five attributes through a pretest. The last five attributes included customer review, farmhouse reputation, smoothness of shopping platform, exclusive discounts and product return. Therefore, this study would utilize attributes listed above to design a formal questionnaire.

RESULTS AND DISCUSSION
The results revealed that customer review was the most important attribute for customers when it came to buy fruits on on-line shops, and its importance percentage was 37.78%. It showed that customers doubted the quality of fruits and wanted to know more product information from other buyers. The second attribute customers preferred was product return with 31.95 importance percentage. It told that consumers cared the conveniences of product return because they feared they might receive damaged fruits. The third and fourth attribute were smoothness of shopping platform and farmhouse reputation, as well as their important percentage were 17.93% and 9.53% respectively. Smoothness of shopping platform meant the smoothness customers felt when they visited fruit on-line shops and finished their purchases. The result reflected customers favored user-friendly websites so that they could save time in internet shopping. Farmhouse reputation stood for how customers know the fruit shops and most customers would like to adopt the suggestions from their family or friends. The last attribute was exclusive discounts with 2.81 importance percentage. Unexpectedly, price discounts rarely influenced consumers. It was assumed that considering the fee of shipping, customers chose to buy fruits from internet for other reasons so price was not their main concern.

CONCLUSION
From the results, consumers valued customer reviews and product return service for fruit on-line shops. On the other hand, the attribute of exclusive discounts had least influence. It is unusual because on-line shoppers always pay attention to price promotion and avoid product return. It is assumed that fruit is not similar to the other standard goods and easily get damaged so that consumers care the conveniences of product return. As for the exclusive discounts, consumers can not feel price discounts because of the fee of shipping. For fruit internet sellers, they need to notice the managements of customer reviews, services of product return, and good quality of fruits and shopping websites. Therefore, it naturally creates web WOM which attracts more consumers to buy fruits from off-line to on-line.

壹、 中文文獻
朱崇佑(2015)，「虛實整合平台個案分析」。國立政治大學科技管理與智慧財產研究所碩士論文。
李佩穎(2014)，「有機生鮮農產品線上銷售創新商業模式」，臺灣大學商學研究所碩士論文。
李宗儒、黃丹青、顧婉瀛(2011)，「蔬果於電子商務銷售模式之探討」，農林學報，第六十卷第四期，頁263-274。
李宗儒、鐘秀欣、劉曼貞、陳麒文(2005)，「以消費者的觀點來進行農産品網路商店功能之分析」，電子商務學報，第七卷第三期，頁215-238。
周文賢、張欽富(2000)，聯合分析在產品設計之運用(p.26)，台北:華泰文化。
林子靖(2010)，「消費者虛實通路選擇決定因素與消費行為之研究-以農產品為例」，國立中興大學應用經濟學系所碩士論文。
林震岩(2012)，多變量分析：SPSS的操作與應用(p.695)，台北：智勝文化事業有限公司。
胡迪凡 (2006)，「正常性折扣和贈品促銷，其與品牌知名度間對網際網路消費者購買意願影響之研究-以無線電話機為例」，國立交通大學管理學系在職專班國際經貿組碩士論文。
徐振軒(1998)，「網際網路上促銷模式之研究」，國立中山大學資訊管理學系碩士論文。
張佩穎(2009)，「網站品質對消費者知覺風險與網購意願之影響-以農產品為例」，國立成功大學國際企業研究所碩士論文。
陳乃維(2015)(譯)，商用統計學(pp. 274-351)，台北：華泰文化。
陳育棠(2004)，「網站環境之認知、情感對消費者態度及行為意圖之影響─以農產品網站為例」。國立中興大學行銷學系碩士論文。
黃俊英(1995)，多變量分析，台北：華泰文化事業公司。
廖國鋒、張詠筑(2011)，組合價格促銷類型對網路購物衝動性購買行為影響研究－以女性內衣為例，輔仁民生學誌，第十七卷第二期，頁25-43。
劉興榮(2010)，「影響消費者對有機農產品網路銷售價格容忍度之因素」，花蓮區農業改良場研究彙報，第二十八卷，頁45-60。
鄧振源(2005)，計畫評估─方法與運用，基隆：華梵大學運籌規劃與管理研究中心。
簡立賢、莊雁喬、王俊彬（2002），「電子交易環境下農產品運銷通路新模式的探討」，台灣土地金融季刊，第三十九卷第一期，頁65-79。
魏佑霖(2013)，「以組織經濟學角度研究農產品電子商務營運模式」，國立臺灣大學農業經濟學研究所碩士論文。
貳、 英文文獻
Aaker, D. A. (1991), Managing Brand Equity: Capitalizing on the Value of a Brand Name, New York: The Free Press.
Berry, L. L., Seiders, K., & Grewal, D. (2002). “Understanding service convenience,” Journal of Marketing, 66(3), pp.1–17.
Best, R. J. (2009), Market-based management: Strategies for growing customer value and profitability (5th ed.), New Jersey: Pearson Education.
Bond, Cathy (1995), “Premiums with Pep,” Marketing, (Spring), pp.12.
Campbell, Leland and William D. Diamond (1990), “Framing and sales promotions: The characteristics of a good deal,” Journal of Consumer Marketing, 7(4), pp.25-31.
Cattin, P., & Wittink, D. R. (1982), “ Commercial use of conjoint analysis: A survey,” Journal of Marketing, 46(3), pp. 44-53.
Coltman & Devinney & Latukefu & Midgley (2001), “E-business: Revolution, evolution, or hype”, California Management Review, 44, pp.57-86.
D.K. Gauri, A. Bhatnagar, R. Rao (2008), “Role of word of mouth in online store loyalty: Comparing online store ratings with other e-store loyalty factors,” Communications of the ACM, 51(3), pp.89–91.
Dowling, G. R.,Staelin, R.(1994), “A Model of perceived risk and intended risk-handling activity,” Journal of Consumer Research,21,pp.119-134.
Filieri, R. (2015), “What makes online reviews helpful? A diagnosticity-adoption framework to explain informational and normative influences in e-WOM,” Journal of Business Research, 68(6), pp. 1261-1270.
Green, P. E. (1974). “On the design of choice experiments involving multifactor alternatives,” Journal of Consumer Research, 1(2), 61-68.
Green, P. E., & Wind, Y. (1973), Multiattribute decisions in marketing: A measurement approach, Hinsdale, IL: The Dryden Press.
Green, P. E., and Srinivasan, V. (1978), “Conjoint analysis in consumer research: Issues and outlook,” Journal of consumer research, pp.103-123.
Hennig-Thurau, T., Gwinner, K.P., Walsh, G., & Gremler, D.D. (2004), “Electronic word-of mouth via consumer-opinion platforms: What motivates consumers to articulate themselves on the internet?” Journal of Interactive Marketing, 18(1), pp.38–52.
Jiang, L., Yang, Z. L., & Jun, M. (2013), “Measuring consumer perceptions of online shopping convenience,” Journal of Service Management, 24(2), pp.191-214.
Jiang, Z., & Benbasat, I. (2007), “Investigating the influence of the functional mechanisms of online product presentations,” Information Systems Research, 18(4), pp.221–244.
Kaplan, L. B., Szybillo, G. J. & Jacoby, J. (1974), “Components of perceived risk in product purchase: A cross-validation,” Journal of Applied psychology, 59(3), pp.287-291.
Keller, K. L. (1993), “Conceptualizing, measuring, and managing customer-based brand equity,” Journal of Marketing, 57, pp.1-22.
Kim, H. W., Xu, Y. J., & Gupta, S. (2012), “Which is more important in Internet shopping, perceived price or trust?” Electronic Commerce Research and Applications, 11(3), pp.241-252.
Kim, K., & Prabhakar, B. (2000), “Initial Trust, Perceived Risk, and the Adoption of Internet Banking”, Proceedings of the Twenty-First International Conference on Information Systems, Cairo, Egypt, December, pp.537-543.
Ku, E. C. S. (2012), “Beyond price: how does trust encourage online group's buying intention?” Internet Research, 22(5), pp.569-590.
Lai, J. Y., Ulhas, K. R., & Lin, J. D. (2014), “Assessing and managing e-commerce service convenience,” Information Systems Frontiers, 16(2), pp.273-289.
Lichtenstein, D.R., Ridgway, N. M. and Netemeter, R.G. (1993), “Price perceptions and consumer shopping behavior: A field study,” Journal of Marketing Research, 30, pp.234-245.
Lohse, G.L. and Spiller, P. (1998), “Electronic shopping,” Communications of the ACM, 41(7), pp.81-87.
Schmalensee, R. (1984), “Commodity Bundling by Single-Product Monopolies,” Journal of Law and Economic, 25, pp. 67-71.
Segal, M. N. (1982), “Reliability of conjoint analysis: Contrasting data collection procedures”, Journal of Marketing Research, 19(1), pp.139-143.
Simonin, B. L. and Julie A. R. (1995), “Bundling as a strategy for new product introduction: Effects on consumer’s reservation prices for the bundle, the new product, and its tie-in,” Journal of Business Research, 33(3), pp. 219-230.
Steckel, J. H., DeSarbo, W. S., and Mahajan, V. (1991), “On the creation of acceptable conjoint analysis experimental designs,” Decision Sciences, 22(2), pp.435-442.
Yu, W. Y., Yan, C. G., Ding, Z. J., Jiang, C. J., & Zhou, M. C. (2014), “ Modeling and Validating E-Commerce Business Process Based on Petri Nets,” Ieee Transactions on Systems Man Cybernetics-Systems, 44(3), pp.327-341.