科技接受模式(TAM)被廣泛用來測試使用者的科技接受方式，但這理論 隨著被運用在不同問題、背景和研究方法，也導致構面關係顯著性、方向 與數值大小有些分歧。面對TAM的發展|國內外也開始有學者針對TAM 進行後設分析，然而這些國內外相關後設分析之研究卻沒有針對台灣使用 者進一步討論，因此本研究彙總了國內2002至2013年以台灣使用者為對 象的研究，共36篇文章，40筆資料1將原本TAM理論擴充至6個構面、 12組關係。另外本研究亦將可能會造成研究差異的因素，如1資料來源」、 「問卷型」、「受訪對象」、「採用自由度」與「樣本大小」等研究方法上 的變項視為調節變數進行後設分析，以便能夠更正確的驗證不同的研究方 法上的選擇是否影響該理論中各構面之關係的大小，以供後續應用TAM時 之參考，結果發現在12組關係中有5組關係是比較穩定的，而且本研究所 探討之調節變數的確對某些關係的效果大小具有調節作用。

The technology acceptance model (TAM) is one of the most widely used test models of technology acceptance. Many TAM studies are characterized by different question, area and methodology, resulting in conflicting and somewhat confusing findings which vary considerably in terms of statistical significance, direction and magnitude. Faced with the development of TAM, domestic and foreign scholars have begun to discuss a meta-analysis of the TAM. However, those studies have no flirther analysis on Taiwanese users. This paper conducted a meta analysis based on 36 selected empirical articles which focus on the data from Taiwanese users between 2002 and 2013 (including 36 studies, 40 research data set, 6 integrated constructs and 12 relations between constructs). Furthermore, we also conducted a moderator analysis for the effort of different source of article, types of questionnaires, respondents, freedom of acceptance and sample size. The result suggests that 5 relations are more stable in our research, and the moderator indeed affects effect size of some relations in TAM.