A Meta-analysis of Empirical Studies on the Effects of Extensive Reading

就實證性研究探討廣泛閱讀成效之整合分析

王儷蓉(Li-Jung Wang)

指導教授 ： 卓江

國立清華大學/人文社會學院/外國語文學系/碩士(2010年)

https://doi.org/10.6843/NTHU.2010.00655

摘要

This study aimed at investigating the effect of extensive reading by meta-analysis. In order to collect experimental or quasi-experimental studies for effect size calculation, different sources were searched with four key words, “extensive reading”, “Free Voluntary Reading”, “Sustained Silent Reading”, and “self-selected silent reading” and the search period was limited to 1977 to 2007. Systematic search started from databases, PsycINFO, PsycARTICLES, and Linguistics and Language Behavior Abstracts-CSA (LLBA). In addition, CSA Linguistics and Language Behavior Abstracts as well as ERIC were the main databases used in LLBA. These databases contained the 14 academic journals that Norris and Ortega (2000) had browsed. Moreover, local and international theses and dissertations were searched through Electronic Theses and Dissertations System in Taiwan as well as ProQuest Dissertations and Theses (PQDT). In order to avoid publication bias, 16 volumes of proceedings of the International Symposium on English Teaching in Taiwan were browsed. Online journals such as The International Journal of Foreign Language Teaching, International TESL Journal, Reading in a Foreign Language, Reading Matrix, Reading Online, TESL-EJ, and TESL Reporter were all browsed. Studies were also retrieved through useful bibliographies such as Krashen (2007) and the references of each study.
Thirty five studies which offered sufficient statistical data were included to calculate Cohen’s d. The effects of extensive reading on (1) reading speed, (2) vocabulary acquisition, (3) reading comprehension (including cloze test and reading comprehension test), (4) writing fluency, (5) accuracy, and (6) language proficiency were examined based on the effect sizes. In some dependent variables, treatment period and age were further investigated to see if the effect was influenced.
The findings show that in uncontrolled experiments extensive reading have large effects on improving reading speed, reading comprehension, writing fluency, accuracy and language proficiency and moderate effect was found on vocabulary acquisition. Although participants who received extensive reading improved after the treatment, some uncertainties still remained on whether they outperformed their control counterparts. In controlled experiments, large effect of extensive reading is found on improving reading speed and moderate effect on reading comprehension with reading comprehension test. Except accuracy, positive effects on the other variables were also found but no significance was achieved.
When examining gain scores, experimental group exceeds control group to a great extent on reading speed, but on vocabulary acquisition and reading comprehension with cloze test, experimental group is found to be only slightly better than the control group. For the rest variables, positive effects of extensive reading were found but uncertainty remained since their confidence intervals included zero.
In addition to analyses on different designs, age and treatment period are not found to be important factors that affect the efficacy of extensive reading. In terms of age difference, only on accuracy were college learners found to benefit more than younger learners. As for the length of treatment, it was found to be related to the effect of extensive reading on writing fluency and accuracy but not on other variables. The results of the current study show the positive effects of extensive reading, which are beneficial to language learning. Therefore, extensive reading is worth being adopted in the language classroom for achieving better learning efficacy.

關鍵字

廣泛閱讀 ； 自由閱讀 ； 持續性靜讀 ； 自我選讀

並列摘要

本論文主要是探討廣泛閱讀在英語教學領域之成效。廣泛閱讀、持續性靜讀、自由閱讀、自我選讀，是四個主要用來搜尋統合分析所需實證性研究的關鍵字。由於Norris和Ortega (2000)曾瀏覽過的十四本期刊均包含在資料庫當中，因此本研究著手於資料庫的搜尋(PsycARTICLES、PsycINFO及Linguistics and Language Behavior Abstracts-CSA (LLBA)，其中LLBA 包含了CSA Linguistics and Language Behavior Abstracts以及ERIC)以及台灣的全國碩博士論文資訊網、數位化論文典藏聯盟也是主要的資料來源。此外，為了避免發表偏倚的現象，本論文也涵蓋來自於台灣英語文教學國際研討會的十六本論文集的實證性研究以及七本線上期刊(The International Journal of Foreign Language Teaching, International TESL Journal,Reading in a Foreign Language,Reading Matrix, Reading Online,TESL-EJ,TESL Reporter)，以上均為本研究提供不少可分析的資料。
　　在西元1977到2007這三十年間，共三十五篇實證性研究可提供有效程度的分析(本研究以Cohen’s d呈現有效程度)。除了檢視廣泛閱讀對於：（一）閱讀速度、（二）單字習得、（三）閱讀理解能力(含克漏字及閱讀測驗)、（四）寫作流暢度、（五）精確度及（六）語言能力的影響之外，學習年齡和實驗時間的長短也是分析重點。
　　研究結果發現廣泛閱讀對於強化學生的閱讀速度、閱讀理解、寫作流暢度、精確度及語言能力的改善都有很大的提升；而對於單字習得也有一些助益。
　　即使學生在接受廣泛閱讀之後，能力進步許多，但他們的表現是否優於對照組則仍存在一些變數。但是相較之下，廣泛閱讀還是能有效提升學生閱讀速度及閱讀理解的程度。大體而言，除了精確度之外，實驗組在其他應變項的成果，皆有優於對照組的趨勢；進步分數的分析中，實驗組除了在閱讀速度上優於對照組，在單字習得和克漏字的表現上也都有些微的優勢。但是這個結論尚未明確，乃因所得之信賴區間可信度仍沒有顯著差異。雖然在其他因變數的檢視上，廣泛閱讀似乎有所幫助，但統計結果尚無法幫助我們下定論－－廣泛閱讀的成效，相對於提升閱讀速度、單字習得和克漏字的表現，其他應變項能否也有顯著的成果。
　　此外，本研究同時也發現廣泛閱讀對於大學生而言，只有在準確度上的助益較大，而實驗時間也只對寫作流暢度及準確度的改善有所影響。在其他變數的分析上，廣泛閱讀成效並沒有受到學習年齡及實驗時間長短的影響。由以上研究結果可得知，廣泛閱讀對於改善語言學習是有其成效的。因此，若可以將廣泛閱讀與傳統教學合併使用，相信對於語言學習的成效將會有所提升。

並列關鍵字

extensive reading ； Free Voluntary Reading ； Sustained Silent Reading ； self-selected silent reading

參考文獻

Bus, A.D., et al. (1995). Joint book reading makes for success in learning to read: A meta-analysis on intergenerational transmission of literacy. Review of Educational Research, 6 (1), 1-21.

Carrell, P.L., & Carson, J.G. (1997). Extensive & intensive reading in an EAP setting. English for Specific Purposes, 16 (1), 47-60.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112 (1), 155-159.

Cook, T.D., & Leviton, L.C. (1980). Reviewing the literature: A comparison of the traditional methods with meta-analysis. Journal of Personality, 48 (4), 449-472.

Davis, C. (1995). EA: an expensive extravagance, ELT Journal, 49 (4), 329-336.