本論文以政治學的分配理論為基礎，藉由次級資料的統計分析，探究第三、四屆立法委員的分配政策提案行為是否會受到立委所屬選區規模大小所影響？並且也匯集第三、四屆所有區域立委的樣本，一併觀察第三屆跨越至第四屆，選區規模的變化與否（包括經過選區重劃的臺北縣）是否使得立委的分配政策提案行為有所差異？本文假設，相對於中或大型規模選區來說，小規模選區立委因為員額少，所以對於爭取選區利益的功過容易被選民追究，因此他們在分配政策提案上的動機較強。同樣的，當選區規模改變時，員額減少的選區立委將會增加分配政策提案的動機，反之亦然。從統計上的顯著性來說，分析結果發現選區規模對於區域立委分配政策提案數量的影響強度並不明顯，只有在第三屆個別屆次分析當中，小選區立委分配政策提案數量明顯多過於大型選區立委。其他的分析，包括第四屆個別屆次分析，或是匯集資料的選區規模變動對於分配政策提案數量的影響分析，都沒有看出選區規模或是選區規模改變的顯著性影響。不過，雖然影響強度不明顯，但是影響的型態仍然可以從統計分析結果當中加以辨識，而這種影響型態與我們的預期並沒有太大的差距。本文的發現對於臺灣分配政治的研究具有一定程度的啟示。

This thesis explores the motives of legislators seeking distributive benefits for their districts by secondary data. Based on distributive theories, the authors investigate whether the pork-bill-initiation behavior of legislators from the 3rd and 4th legislative Yuan in Taiwan are affected by the district size where legislators are elected from. The authors also observe whether the bill-initiation behavior of legislators would change as a result of the changes of district size across different Legislative Yuan.Ignoring statistical significance, Negative Binomial Regression analyses show that the regression coefficients of district size are as expected both in sign and scale, in the sense that the larger the district size is, the less the incentives of legislators have on bill-initiation. This is true not only in the respective term of the 3rd and 4th Legislative Yuan, but also on the aggregate-level analysis dealing with the changes of district size.This thesis does not find significant association between district size and legislator's bill-initiation behavior on distributive benefit, nevertheless the statistical results still show identifiable pattern. The authors suggest that it is still worthy to study this issue in terms of different kinds f pork barrel behavior other than bill-initiation.