透過您的圖書館登入
IP:54.210.126.232
  • 學位論文

問卷填寫時間與問卷品質關係之研究探討

Investigating the Relationship between Answering Time and Quality of Web Survey

指導教授 : 王筱惠 賴錦慧
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


因為電腦的普及和網際網路的進步,問卷調查已從傳統的紙本問卷轉變到現在普遍使用的網路問卷。但網路問卷仍有許多缺點,其中最嚴重的問題是,許多網路問卷受測者在填寫問卷過程中不專心,導致問卷樣本品質不良。但是要如何判斷樣本品質的好壞至今都還沒有一個具體的解決辦法。 有鑑於此,本研究利用填寫問卷的時間做為判斷問卷樣本品質的依據。本研究建立網路問卷平台,蒐集受測者填寫每題問卷題目的填寫時間,並提出四種過濾樣本的統計方法進行受測樣本過濾。藉由過濾樣本後的問卷信度變化情形,觀察填答時間對於問卷調查品質的影響。本研究提出四種將樣本過濾的統計方法為,(1) 每份問卷樣本的總填答時間(Total filling-out time per sample, TTPS)異常者、 (2)填寫問卷各題的時間變異程度(Variance of per question's filling time of per sample, VQTS)異常大者,(3)問卷的單題填答時間(Single filling time per question in sample sets, STPQ)異常者及(4)問卷填答時間異常的題數多寡(Number of abnormal STPQ in sample sets, NASTPQ)為樣本過濾的準則。 根據本研究的研究數據顯示,利用受測者填寫問卷的時間當作樣本的過濾準則會提高整份問卷的信度(四種過濾方法皆會提高問卷信度),如果想要使問卷信度能夠明顯提高又能使樣本數保持在一定的數量,NASTPQ則是最適合的方法。除此之外, 透過本研究發現,利用本研究所提出的四種過濾方法過濾受測樣本,所獲得的問卷信度比用反向題過濾受測樣本的問卷信度還要高,而且利用反向題過濾樣本後,再用本研究的方法過濾樣本,可以更有效的將較劣質的樣本過濾掉,獲得更高的問卷信度。簡而言之,本研究透過觀察受測者填寫問卷的時間以及問卷品質之間的關係,且在提高問卷品質的部分提出比反向題過濾更好的方法。

並列摘要


Because of the popularity of internet, web surveys replace the roles of paper ones to become the main trend of data collection. Web survey has many advantages including fast sample collection, low cost and ready-for-input data for statistics software. However, the most serious problem of web survey is sample quality since many web users hardly concentrate on answering surveys. Therefore, this research focuses on investigating how the answering time influences survey quality measured by reliability. This research builds a platform to collect sample answering time and proposes four methods based on the answering time to filter out bad samples. Four proposed statistical methods are Total filling-out time per sample (TTPS), Variance of per question's filling time of per sample (VQTS) , Single filling time per question in sample sets (STPQ) and Number of abnormal STPQ in sample sets (NASTPQ). Our results show all filtering methods raise sample reliability successfully. In terms of keeping the original sample size and improving reliability at the same time, NASTPQ has the best performance. Furthermore, in reliability improvement, we found our methods not only have better efficacy than reverse coding, but also can assist reverse coding to have better bad-sample-filtering. In conclusion, this research investigates the relationship between answering time and answering quality of web survey, and proposes better solution than reverse coding in improving survey quality.

並列關鍵字

Box-plot Reliability Answering Time Web Survey

參考文獻


李政忠(民93)。網路調查所面臨的問題與解決建議。資訊社會研究,6,1-24。
蘇蘅、吳淑俊。(民86)。電腦網路調查可行性及回覆者特質的研究。新聞學研究,54,75-100。
Batagelj, Z., Lozar, K., & Vehovar, V. (1998).Technical and methodological issues in WWW surveys. Meeting of the American Association for Public Opinion Research, St. Louis.
Bosnjak, M., & Tuten, T. L. (2001).Classifying Response Behaviors in Web-Based Surveys. Journal of Computer-Mediated Communication, 6(3).
Cobanoglu, C., & Cobanoglu, N. (2003). The effect of incentives in Web surveys: Application and ethical considerations. International Journal of Market Research, 45(4), 475- 488.

被引用紀錄


邱筱涵(2015)。遊戲因子對工作品質影響之研究-以線上問卷填答為例〔碩士論文,中原大學〕。華藝線上圖書館。https://doi.org/10.6840%2fcycu201500373

延伸閱讀