- 學位論文
探討涉入程度與網路書評對偏好穩定性的影響：以奇幻小說為例
Testing the intermediary effect of product involvement on the influence of online book review: a case study of mystery readers
摘要
偏好可分成內在穩定以及受外界資訊刺激所建構出的兩種型態。不同情境刺激，或是顧客對於產品的涉入程度，也就是熟悉瞭解程度，皆有可能影響顧客對於產品選擇的偏好穩定性。過去對於涉入程度及偏好穩定的研究較少針對愉悅閱讀書籍此類文化性商品；因此本研究希望能藉由實驗情境的操弄，瞭解讀者對於愉悅閱讀書籍的不同涉入程度，是否會因網路書評所傳達的口碑資訊，進而影響偏好穩定性。 本研究採用實驗法，並以學者所提出的涉入量表作為愉悅閱讀讀者的涉入程度測量。在實驗選書情境中，請讀者依平時閱讀喜好形成不同閱讀興趣，區別出具正、負面初始印象的奇幻小說。稍後給予讀閱讀與其初始印象相衝突的書評。藉由直接與認知相衝突的形成，瞭解正負書評對於偏好穩定性的影響程度。並藉由問卷填答，分析面對正負書評及其他相關訊息時，不同涉入程度讀者所考量的資訊是否有所不同。 研究結果如下：一、讀者涉入程度影響偏好穩定性，又以高涉入者偏好穩定性較佳；但本次所招募的低涉入者非全無閱讀經驗者，部分低涉入者是其他文學作品的高度涉入者，在實驗時可能會套用其個人閱讀經驗。二、正負書評對閱讀前後評分具影響力，尤以負評易形成損害趨避想法，評分變動大。三、高衝突較低衝突書評易造成偏好穩定性變動，低涉入者變動較為明顯。四、涉入程度影響選書資訊的考量：高涉入者以文字資訊為主，低涉入者偏重圖像式資訊。 進一步研究建議上，未來在不同涉入程度受試者招募可配合閱讀的頻率、興趣，再搭配涉入量表的涉入程度分數，利用多元面向瞭解受試者間高低涉入程度，使兩個族群間差異更明顯。此外，影響受試者資訊的選擇可以高低強度進行資料收集，以瞭解不同階段受試者所考量到各資訊的影響及重要程度。
並列摘要
The judgment of readings for leisure inevitably involves certain degree of uncertainty as it is difficult to determine the quality of a work before its consumption. Potential readers often have to rely on decision cues such as reviews and word of mouths for judgment. The influence of reviews is arguably more salient for readers who have a less degree of involvement with a certain genre of leisure reading. The study set out to explore the influence of genre readers’ degree of involvement on their preference stability. To test our hypothesis that highly involved readers has a more stable preference, an experiment was conducted where 41 high and low involvement readers in fantasy novels were exposed to online book reviews opposite to their own. Preference stability was then measured by the extent to which the participants’ opinions of a set of eight previously unknown titles vary before and after they were exposed to contradictory reviews. Major findings are as following: firstly, high involvement readers had a higher degree of preference stability; secondly, negative reviews had a more significant impact on readers’ judgment. Furthermore, it was also found that readers with high and low involvement relied on different decision cues for judgment; while highly involved readers relied more on introductory texts and editorial reviews, readers with low involvement tended to rely heavily on star rating, which demands less cognitive effort.