摘要 近年來隨著台灣主體研究的蓬勃發展，客家語文得以納入國民中 小學正式課程，所以不論正式教材或各類補充教材及字詞典等，皆開 始了蓬勃的編輯出版。然而不論在撰寫教材或編輯詞目的過程中，卻 始終缺乏科學客觀的客語字頻、詞頻數據可供參考依循，成為亟待研 究解決之問題。因此本論文提出建立客語語料庫的方法，進行字頻、 詞頻的統計；並以其中得出的高頻字、高頻詞，作為客語各式教材編 輯出版的依據。相信以建立客語語料庫這種科學客觀的方法，應該可 以為客語教材及各式出版品， 展開新頁。

Abstract For the surge of Taiwanese identity research, Hakka have become a formal curriculum in compulsory education, and the education policy has brought the mushroom of editing and publishing of text books, dictionaries and complementary teaching materials. To develop Hakka corpus as the basis of editing and publishing all kinds of Hakka teaching materials.The scientific and objective method of corpus development could launch the new leaf of Hakka teaching materials and all kinds of Hakka publication. Therefore, the thesis, gathering character and term frequency statistics and extracting high frequent characters and terms to develop Hakka corpus as the basis of editing and publishing all kinds Hakka teaching materials. The scientific and objective method of corpus development could launch the new leaf of Hakka teaching materials and all kinds Hakka publication.