圖書館環境脈絡與個人特性對圖書館服務品質主觀評估之影響

Abstract

圖書館導向之客觀評估與使用者導向之主觀評估是圖書館服務品質評估的兩大研究取向，然而，過去多數評估研究僅採用其中一種取向，或受限於資料分析方法，不僅無法完整呈現圖書館主客觀服務品質現況，亦缺乏深入探討兩者評估結果之間的關係。本研究目的在分析使用者對圖書館服務品質主觀評估之個人與館際差異，探討個人特性與圖書館環境脈絡 (客觀服務品質與結構屬性) 對使用者主觀評估圖書館服務品質的影響，並檢驗使用者的服務品質知覺是否作為其使用圖書館的狀況影響滿意度之中間機制，以及此中間機制是否受圖書館環境脈絡影響。縱合相關文獻，本研究建構「圖書館服務品質整合評估理論模型」，並提出五個研究假設。 本研究採用橫斷式問卷調查，取得國內51所大學圖書館為圖書館層次樣本，並以此51所大學之29,916位師生為使用者層次樣本。圖書館層次收集51所大學圖書館之資源投入與服務使用量資料，並透過評等及加權予以標準化，成為客觀服務品質或圖書館服務脈絡，包含人力、館藏、設備等資源投入品質，以及造訪使用、館藏使用及支援服務使用等輸出品質。此外，以圖書館之服務規模及公私立別等屬性為結構脈絡。使用者層次主要收集51所大學師生的圖書館使用頻率與廣度，以及對圖書館的內容、系統及支援等服務品質知覺、滿意度等主觀評估資料，以及使用者之身分類別、學科領域類別及自我效能等個人特性資料。資料分析採用層級線性模式/多層次分析法，逐一檢驗研究假設。 由多層次多變量回歸分析獲得以下五項研究結果：1. 不考慮使用者與圖書館層次任何因素下，使用者的使用頻率、使用廣度、服務品質知覺及滿意度等主觀評估均存在個人與館際差異，但超過88% 的差異來自使用者層次，圖書館層次能解釋差異的比例相對在12% 左右。2. 考慮使用者與圖書館層次所有變因後，使用者對圖書館服務品質的主觀評估會因使用者的身份類別、領域類別及自我效能等個人特性的不同而有差別，但滿意度不受自我效能的影響。3. 在排除使用者層次解釋變因的影響效果後，圖書館環境脈絡八個因素中，有一半對使用者的主觀評估有作用，分別為服務規模、人力品質、設備品質及造訪使用品質。而六個主觀評估品質，以使用者滿意受較多環環境脈絡因素影響，分別為服務規模、設備品質及造訪使用品質。使用者的支援服務品質知覺則完全不受圖書館環境脈絡影響。4. 使用者的服務品質知覺是其使用狀況影響滿意度之淨中間機制，即存在直接與間接影響關係，使用頻率越高，滿意度越高，透過服務品知覺增強其滿意度，而使用廣度越廣，滿意度卻越低，透過服務品知覺減弱其滿意度。5. 使用者的使用狀況直接及間接透過服務品質知覺影響滿意度之淨中間機制，多數的影響關係受圖書館環境脈絡影響，但使用狀況直接影響滿意度之效果不因圖書館環境脈絡的不同而有差異。 由結果顯示，本研究之研究假設與問題大部分得到支持和解答，但部分因研究方法之限制，可能影響研究結果，建議後續研究進一步探討與修正，且本研究所建構之「圖書館服務品質整合評估理論模型」，乃首次應用於國內大學圖書館的評估環境，亦建議後續重複驗證，以精煉評估理論模型。本研究結果亦提供大學圖書館參考，建議各館持續進行服務品質評估並深入了解使用者，使圖書館的服務更貼近使用者的需求。

Library-oriented or objective evaluation and user-oriented or subjective evaluation are two major approaches to evaluate library service quality. However, less is known about the relationship between the results of the two evaluation approaches, particularly about the effects of library contexts on users’ subjective evaluation. The present study employs a combination of library- and user-oriented evaluation approaches to develop an integrated evaluation model of library service quality with multi-level and multi-dimension. Based on this evaluation model, the purposes of the present study are (1) to analyze whether users’ subjective evaluation of library service quality varies among user individuals and libraries, (2) to examine the effects of individual attributes and library contexts (service and structural context) on users’ subjective evaluation, (3) to diagnose if the relationship between users’ self-reported usage of and user satisfaction with a library is mediated by users’ perceived library service quality, and (4) to test how the mediated effects are influenced by the library contexts, if the relationship between users’ self-reported usage and user satisfaction is mediated by users’ perceived library service quality. The present study utilizes a cross-sectional survey designed to collect library and user level data respectively in 51 university libraries in Taiwan. There are eight factors of library level as library contexts. Six are indicators of objective, quantitative quality of library service context and two are variables of structural context. Library input and output data are standardized and weighted as six indicators of objective, quantitative quality of library service context: (1) manpower quality, (2) collection quality, (3) facility quality, (4) library visit quality, (5) circulation quality, and (6) support service quality. Two variables of library structural context are population size and public/private sector of a library. For the user level, there are six indicators of subjective evaluation and three individual attributes. Six indicators of subjective evaluation are (1) self-reported library use frequency, (2) library use width, (3) perceived content service quality, (4) perceived system service quality, (5) perceived support service quality, and (6) satisfaction. Individual attributes are user type (faculty and student), discipline (humanity, social science, and science) and self-efficacy. The sample size for the library level is 51 university libraries and for user level is 29,916 respondents from the faculty and students of the 51 universities. The hierarchical linear model (HLM) or multilevel analysis is utilized to estimate the effects of user-level and library-level factors on users’ six subjective evaluations. The findings show that users’ six indicators of subjective evaluation of library service quality vary among user individuals and libraries respectively, when there are no factors of user level or library level considered. More than 88% of the variance in users’ subjective evaluation is at the user level, while about 12% is at the library level. The findings also show that users’ six indicators of subjective evaluation differ by the level of users’ type, discipline and self-efficacy; but there is no self-efficacy difference in user satisfaction. Four of eight factors of library contexts (population size, manpower quality, facility quality and library visit quality) have significant effects on most of the users’ six indicators of subjective evaluation. User satisfaction is the most affected indicator by population size, facility quality and library visit quality. However, users’ perceived support service quality does not vary by the library contexts. Further analyses show that users’ perceived service quality (perceived content quality, perceived system quality and perceived support service quality) is the partial mediation between self-reported usage (library use frequency and use width) and user satisfaction. In other words, self-reported usage has direct and indirect effects on user satisfaction through perceived service quality. Users’ library use frequency positively affects user satisfaction, while users’ library use width has a negative effect on user satisfaction. Moreover, seven factors of the library contexts, except for collection quality, have effects on some of the partially mediated effects. The effect of users’ library usage on user satisfaction does not differ by the library contexts. Finally, the implications of these findings and needs for future research are discussed.