本研究參酌PISA閱讀評量架構分析90至98年基測國文科題組選文類 型和試題閱讀歷程的組成，並以Kruskal-Wallis單因子等級變異數檢定考生 不同閱讀層面的表現差異，主要發現摘述如後。在文本類型分布上，文言 文和白話文的題組數和配置試題數比例約1:2。在選文類型分布上，文言文 以敘事文出現最多，約51%，其次分別為韻文;記述文和議論文，約介於 14%到19%間，說明文和指示都不到3%。白話文以說明文;議論文和敘事 文三類居多，約介於21%到25%間，其次分別為記述文和韻文，約在13% 和14%，指示不到4%。在閱讀歷程分布上，文言文配置試題比例依序為解 釋文本（67%）;提取資訊（15%）;反思和評鑑（12%）;語文知識的 認知（6%）。白話文配置試題比例依序為解釋文本（46%）;提取資訊 （29%）;反思和評鑑（11%）;語文知識的認知（14%）。在全體考生表 現方面，考生在文言文的平均通過率顯著低於白話文，不同選文類型的平均 通過率沒有顯著差異，不同閱讀歷程的平均通過率有顯著差異，提取資訊的 平均通過率顯著高於解釋文本。在不同成就水準學生的表現方面，高分組在 文言文和白話文的表現無顯著差異，中間考生和低分組則文言文表現低於白 話文，三組考生在不同選文類型的表現皆無顯著差異，三組考生在提取資訊 歷程的表現皆高於解釋文本。篇末根據研究發現提出建議供相關人員參考。

This project consisted of two parts. The first part adopted the PISA reading literacy framework to analyze the structures of reading tasks on the Chinese Basic Competence Test for Junior High School Students (BCTEST) from 2001 to 2009. The second part used Kruskal-Wallis one-way ANOVA by ranks to analyze examinees’ performances on different reading texts and processes. Major findings were stated as follows. In terms of the distributions of reading text, narration was the largest category for classical Chinese tasks, accounting for about 51%. The larger categories for vernacular tasks were exposition, argumentation and narration, ranging from approximately 21% to 25%. As for reading process, the distributions on the classical Chinese tasks were as follows: interpreting texts (67%), retrieving information (15%), reflection and evaluation (12%), and other aspects (6%). The distributions on the vernacular tasks were as follows: interpreting texts (46%), retrieving information (29%), reflection and evaluation (11%), and other aspects (14%). In terms of examinees’ performances, they performed significantly better on vernacular tasks than on classical Chinese tasks, and better on retrieving information items than on interpreting them. However, there was no significant difference between the performances on different reading texts. In the case of the performances of different achievement level examinees, there was significant difference in performance between vernacular and classical Chinese tasks for average (middle 50%) and low achieving group (bottom 25%), but not for high achieving group (top 25%). The performance on retrieving information was significantly higher than that of interpreting texts for different achievement level groups. However, there was no significant difference in the performances on different reading texts for different achievement level groups. Implications based on the findings of this study were proposed for test developers, educators, researchers and police makers.

國中基測;閱讀評量;PISA閱讀評量架構

Basic Competence Test for Junior High School Students, reading assessment, PISA reading literacy framework