立法院在2008 年第七屆立法委員選舉中，首次採用了單一選區兩票制（Mixed-Member Majoritarian Electoral System, MMM ），其結合了單一選區多數決與政黨名單比例代表制兩種席次分配方式，與過去採用的「複數選區單記非讓渡投票制」（Single Non-Transferable Vote with Multi-Member-District system, SNTV-MMD ）在制度特性上截然不同。本文主要目的是檢視選民對於新選制的瞭解，亦即選制知識的高低，是否將對選民之投票參與產生影響。 從既有的文獻中發現，傳統研究或者假定選民對選制有充分的理解而無須納入分析，或者即便探討選制知識高低對投票行為之影響，往往將選民的選制知識本身視為外因變數（exogenous variable ）。本文與過去既有之研究有三點主要的差異。首先，本文並不貿然假定選民對於兩票制的新選舉制度擁有充分的資訊與瞭解，而是以四個問卷題目來測量選民對新選舉制度的瞭解程度。第二，在選制知識的測量方法上，既有研究常直接加總選民答對題目的數目，作為選制知識高低的變數，本文則進一步運用項目反應理論（item response theory ），建構一個「二參數項目反應模型」（twoparameter item response model ）來估計各個測量項目對選制知識的區辨能力，進而建立一個屬於連續變數的選制知識測量。第三，本文認為選制知識在對投票行為的影響上，應是一個內因變數（endogenous variable ），基於此一理論設定，本文結合兩個方程式同時估計選制知識高低的成因及其對選民投票的影響。分析結果顯示，選民的選制知識的確屬於內因變數，而在充分考量此一內因性後，證實選制知識對投票參有顯著的正效應：選民對新MMM 選制越瞭解，參與投票的機率也越高；反之，對新選制越不瞭解，參與投票的機率也越低。

In the trajectory of Taiwan’s democratization, the judicial system has long been entangled in the political conflict, and the general public does not seem to trust the judiciary to be independent of political influences. Some folk sayings about the courts reveal the negative stereotypes of vote-buying verdicts; for example, “the courts are dominated by the Kuomintang;” “at the first trial a heavy sentence is passed, at the second trial the sentence is halved, and at the third trial the case is quashed;” “those with good social connections won’t have any problem, but those without connections will have big trouble;” and, “those elected will be let off, but those losing the elections will be imprisoned.” This study investigates the impacts of political factors (including partisanship, judicial procedure, sociopolitical connections, and whether a defendant is elected or not) on three levels of court decisions on vote-buying litigation. In a departure from quantitative analyses in the preceding literature, this research employs a qualitative method. The methodology adopted in this study involves two steps. The first approach involves the use of documentary analyses; we intentionally select eight salient vote-buying cases, and review their court verdicts and news reports in order to inquire whether or not the judiciary is politically biased in its judgments. The second method employs face-to-face intensive interviews with eight legal experts (including three lawyers, one judge, three law professors, and one senior legal journalist), and asks the respondents questions designed to evaluate the political influences on vote-buying lawsuits and to assess this negative public impression of judicial verdicts. Contrary to expectations, the interview findings reveal that the effects of political factors have considerably less of an influence than expected on court decisions. In the conclusion, the key findings are reviewed, and suggestions regarding judicial politics are made for future research.