Please use this identifier to cite or link to this item:
Title: 隨機化試驗在教育研究中的應用
Other Titles: Randomized Trials: Usage in Educational Research
Authors: 譚克平
Issue Date: Dec-2010
Publisher: 國立臺灣師範大學研究發展處
Office of Research and Development
Abstract: 隨機化試驗是一種行之有年的研究方法,但近年來在不同社會因素的影響之下,隨機化試驗在教育界開始備受注意, 並且引起多方的討論甚至爭議。本文旨在介紹隨機化試驗的內容,包括隨機分派與因果關係的建立;並介紹兩種隨機化試驗的形式, 包括隨機化對照試驗與集群隨機化試驗,以及進行時考量的要點。本文並報導三個隨機化試驗的實例以供參考,從中可看出進行隨機化 試驗時,因實務所需而會有或大或小的調整,最後則以綜合討論作結。本文希望能讓讀者對隨機化試驗有比較全面的初步瞭解,如日後 要從事隨機化試驗時,在設計考量上能更加全面。
Randomized trials have been around for many years as a research design. However, many factors have contributed to an emphasis on their use recently in the field of educational research and this has stirred much discussion from different aspects. The purpose of this paper is to offer an extended introduction to randomized trial design for educational researchers. It begins with an introduction to the social background that contributed to the current stress on the design. Next, the components that constitute a randomized trial will be explained with particular attention to the relationship between random assignment and causation. Two formats of randomized trials, namely, randomized controlled trials and cluster randomized trials will then be described. Afterwards, examples are given to illustrate the intricacies and the kinds of considerations needed for real world applications. Next, some concerns about randomized trials are briefly summarized followed by a presentation of counterarguments against them. The text closes with a discussion of relevant issues pertaining to randomized trial and observational studies. With awareness of the features of and the issues in such designs, interested researchers will be more comprehensive in planning a randomized trial.
Other Identifiers: 3561D29C-6D11-B97B-F7E8-98F1147F449C
Appears in Collections:教育科學研究期刊

Files in This Item:
File SizeFormat 
ntnulib_ja_L0807_5504_069.pdf867.72 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.