Visual and Annotation Effects on User Behaviors of Selecting and Answering Web Survey

• Main Authors: Han-Wen Huang, 黃漢文
• Other Authors: Shiau-Huey Wang
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/38717455456160945490

碩士 === 中原大學 === 資訊管理研究所 === 101 === Web survey becomes increasingly popular because internet platform can offer the advantages of fast survey creation, wide sample collection and real-time data analysis. However, web survey also faces challenges, mainly from unbalanced sample distribution and low participation willingness. Unbalanced Distribution includes Unbalanced Answering Rate of listed surveys and Improper Sampling. Therefore, we propose Explicit Visual Aid (EVA) and Implicit Annotation Aid (IAA) for directing users to answer the specific surveys assigned. We investigate the effects of EVA and IAA with a 2x2 cross-section study design through four different survey listing pages. Our research results show the synergy of EVA and IAA can successfully lead users to select the surveys recommended to them with a 60% acceptance rate. The results also show that 60% acceptance rate, compared with list without aids, can lower Distribution Deviation of listed surveys to one-third. Furthermore, in terms of user psychology, EVA and IAA improve users’ willingness of answering surveys. In brief, this thesis focuses on how to improve web survey sample distribution and users’ willingness of participation. Using ANOVA Analysis, the synergy of EVA and IAA has the best effect, while EVA has even better performance than IVA through post-hoc tests. As regression test is applied, the raise of willingness can lower users’ burden when they are selecting and answering surveys.