首頁Home › 數位介面與使用者體驗 Digital Interface & UX
A/B測試
1這是什麼What is it?
A/B 測試是將使用者隨機分派到兩種設計版本,比較其行為指標(如點擊率、轉換率)的實驗方法。它能提供因果證據,但需足夠樣本與統計顯著性;持續實驗與個人化使其成為產品優化流程,同時也引發倫理與短期指標偏誤的討論。
A/B testing randomly assigns users to two design versions and compares behavioral metrics such as click-through and conversion rates. It can provide causal evidence but requires sufficient sample and statistical significance, and continuous experimentation and personalization make it a product optimization process while also prompting discussion of ethics and short-term metric bias.
2常被用在哪種研究問題Typical research questions
常見問法:哪個版本能提高轉換或點擊?
A/B 測試能以真實行為比較設計。本研究設計兩個版本並蒐集點擊與轉換資料,檢驗設計差異是否顯著影響使用者行為。
A/B testing compares designs with real behavior. This study designs two versions, collects click and conversion data, and tests whether design differences significantly affect behavior.
撰寫內容,非真實研究結果,請與指導教授討論後改寫。
Written as a model example, not a real finding. Please revise with your advisor.
3怎麼測它How to measure it
線上對照實驗+統計檢定
Measurement notes are written in Chinese. Switch to 中/EN to read them.
4代表文獻Key references
英文論文(2000 年後)English paper (post-2000)
Kohavi, R., Longbotham, R., Sommerfield, D., & Henne, R. M. (2009). Controlled experiments on the web: survey and practical guide. Data Mining and Knowledge Discovery, 18(1).
中文論文(2000 年後)
中文論文為預填檢索連結,請自行挑選一篇 2000 年後的期刊論文並核對作者、年份、出處後再引用。
英文引用來自模型整理,引用前請務必核對並閱讀原文。English citations are compiled by a model. Verify and read the original before citing.