對一個新興民主國家而言,政黨輪替是過去從未發生過的脈絡變遷,公民被迫要適應學習這種新的政治環境,而且會因其選舉結果而出現不同的學習路徑,這些脈絡因素都對公民的民主態度變遷造成影響。2008年台灣發生第二次政黨輪替,如果與2000年第一次政黨輪替相比,公民對民主的態度是否有所轉變?透過統計模型的分析,台灣的民主經驗顯示,隨著台灣民主的持續,直接的民主支持很快獲得社會的認同,間接民主支持的增長則較緩慢。世代差異主要呈現在間接的民主支持度上,而不是在直接的民主支持度。泛綠支持者不論經歷贏家或輸家,在直接民主支持度上,都遠高於泛藍支持者與政黨中立者。另一方面,不論泛藍或泛綠支持者,如果在某次選舉是贏家的話,民主價值的發展有倒退或停滯狀態的現象;另一方面,選舉輸家在民主價值上,則有大幅度的成長。年齡與直接的民主支持,呈現一種「U型」的因果關係。但是年齡與間接的民主支持,卻呈現一種「倒U型」的因果關係。因此,制度因素可以解釋直接民主支持度的變遷,世代差異與生命週期可以解釋間接民主支持度的變遷,而選舉結果則同時影響這兩種民主支持度的變遷。
For new democracies, rotation of power between parties is an experience without precedent. Citizens are forced to learn to adapt to this new type of political environment, and furthermore, will adopt different paths of learning depending on the political environment. These different contextual factors also impact changes in citizens' attitudes towards democracy. We are interested in the question of how changes in citizens' political attitudes after Taiwan's first rotation of power compare with changes after the second rotation, in particular given the reversal of the ”winner” and ”loser” roles. Taiwan's experience reveals that when a democratic system remains in place over time, direct support for democracy quickly achieves high levels, but indirect support for democracy grows at a slower pace. Differences over time are much more pronounced when we measure indirect support for democracy. Direct support for democracy is consistently significantly higher among pan-green supporters than both their pan-blue rivals and unaligned voters, regardless of the success or otherwise of pan-green politicians at the polls. In addition, both pan-blue and pan-green supporters express almost universal satisfaction at the election result when their side wins, and dissatisfaction with opposition checks on the activities of the government. Measuring age against direct support for democracy produces a U-shaped correlation, but against indirect support for democracy we find instead a reverse U-shaped correlation. Therefore, institutional factors can explain changes in direct support for democracy: and generational differences with life cycle can explain changes in indirect support for democracy, while election results affect both direct and indirect support for democracy.