本研究旨在探討Web2.0時代下社群網站的服務品質要素，以剛進入國內的世界最大社群網站「Facebook」為研究對象。以焦點團體訪談（Focus group）方式，輔以文獻支持，針對Facebook的使用者做深度訪談，探索使用者所在意的社群網站服務品質要素。訪談結果共整理出24個B2C，及12個C2C等共36個社群網站服務品質要素。再將36個服務品質要素結合Kano模式，探討每個要素的二維品質特性。同時 加入涉入程度、網站社交強度及使用者經驗為調解變數，探討不同涉入程度、網站社交強度及使用者經驗下，服務品質要素的二維品質分類特性。研究結果歸類出4個魅力品質、8個一維品質、16個必須品質及8個無差異品質。另外在不同涉入程度、網站社交程度及使用者經驗下，對於網站服務品質的二維品質皆有不同的分類。

在當時大家認為產品品質與顧客滿意呈現一維的線性關係時，狩也紀昭首先將二因子的概念應用在產品品質上，發現了品質也有二維的特性，因此可以辨別出何種品質較其他品質重要，藉以形成品質改善的優先順序，其後亦有學者將服務品質概念應用在Kano模式上，發現服務品質也有二維品質的特性，然而Kano模式僅能提供品質的屬性分類，卻無法瞭解目前使用者的狀態，因此後來有學者加入IPA( Importance-Performance Analysis )，藉以瞭解目前使用者主觀的感受如何，因此Kano模式是將品質作「定性」，也就是瞭解品質的特性，而IPA是將品質「定量」，也就是瞭解目前使用者的感受。然而將Kano模式與IPA結合後所提出之品質改善建議往往流於模式上的字面解釋，有些甚至不符合公司當時實際在管理品質的情境，因此本研究試著導入核心策略與戰略行動的概念，試著重新提出Kano模式與IPA結合後的品質改善建議。 接著本研究以E-S-QUAL網路服務品質為基礎，結合Kano模式與IPA分析方式，並以Facebook為研究對象，探討Facebook網路服務品質的特性是否具有二維品質模式，並以不同人口統計變項將樣本分群，探討不同人口統計變項下，使用者是否對Facebook網路服務品質特性之分類有所不同，最後驗證本研究所提出的品質改善建議之合理性，並同時與Kano模式與IPA結合的品質改善建議作一對比。 結果發現，Facebook的網路服務品質的確有二維品質的屬性，但在32項品質項目中，沒有任何一項為魅力品質，代表E-S-QUAL所提出的所有品質項目，對於Facebook的使用者而言，沒有一項品質會令他們愛慕或嚮往。而對於不同人口統計變項的使用者而言，對於Facebook的網路服務品質特性之分類，在某些構面上的確有所不同。最後利用分析過後的資料，實證本研究所提出之品質改善建議，同時比較Kano模式與IPA結合的品質改善建議，發現後者確實在某些品質改善建議上有所遺漏，並於本研究所提出之改善建議中加以彌補，以形成更為周全的品質改善建議。 / At the time we think that product quality and customer satisfaction is a linear relationship, hunting Zhao Ji first applied the concept of two factors theory on quality and found that quality and customer satisfaction also has two-dimensional relationship. So you can identify what kinds of quality are more important than others to the formation of quality improvement priorities. Other researches also applied the concept of the Kano model on service quality and found that service quality also has the two-dimensional quality attribute. However, Kano model only can provide the classification of quality attribute, but cannot understand the current experience of the user. So some researches subsequently joined the IPA( Importance-Performance Analysis ), in order to understand current user subjective experience. We can know Kano model is used to “qualitative”, that is, to understand the attributes of quality, while the IPA is used to “quantitative”, that is, to understand the users current experience. However, the quality improvement suggestions from the combination with Kano model and IPA tend to fellow the literal interpretation from two models. Some suggestions even mismatch the actual company management situation. Therefore, the core of this study tried to import the concept of strategy and strategic action into the combination of Kano model and IPA to submit new recommendations of quality improvement. Then this study, based on E-S-QUAL, combining with analysis methods of Kano model and IPA, and using Facebook as an example, explores whether the characteristics of Facebook social network service qualities show two-dimensional quality attributes. Grouping the samples by different demographic variables to explore whether different users will affect the classification of quality attributes. Finally, this study examines the reasonableness of quality improvement suggestions from itself and also makes a comparison with the suggestion combining with Kano model and IPA. The results showed that, Facebook’s network service qualities do have the two dimensional quality attributes. But in 32 network service quality items, there do not have any attractive quality attribute. It shows that, there is no one service quality item proposed from E-S-QUAL will be loved or desired by Facebook users. For the different demographic variable users, there are some different in classifying Facebook’s network service quality attributes in some dimension. Finally, this study uses the result from the data analyzed through Kano model and IPA to examine the reasonableness of quality improvement suggestions from itself, and also makes a comparison with the suggestion combining with Kano model and IPA. The examination discovered that the latter is indeed miss the quality improvement suggestion in some situation. So the suggestions of this study compensated the missing part and also proposed a more comprehensive suggestion for Facebook’s quality improvement.