The New Procedure to Code and analyze the Open-ended Questions' Data in Social Research

社会研究中开放式问题数据编码和分析的新程序

• 批准号: 08551003
• 负责人: KAWABATA Akira
• 金额: $ 4.22万
• 依托单位: Osaka University (1998)Kyoto Koka Women's University (1996-1997)
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (A)
• 财政年份: 1996
• 资助国家: 日本
• 起止时间: 1996 至 1998
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-08551003/
• 关键词: code; textual data; content analysis; AUTOCODE; KTCoder; open-ended question; 社会調査法

The purpose of this research is to show the new computer-aided techniques for the analysis of textual data. Only few attempts have so far been made at content analysis with personal computer in Japan. The main reason is that Japanese have a difficult problem to solve. Japanese have no explicit break to classify words according to a part of speech. Then, our concern is to make new computer programs to classify words.There are three steps in this research. The first is how large amounts of text are converted to machine-readable format. The second is how to code textual data. The third is the methods of analyzing it.The point is to make the software for coding textual materials in Japanese on personal computers. We have completed two programs : AUTOCODE program and KTCoder.AUTOCODE program was originally developed to support the coding of open-ended questions in survey questionnaires. Its basic function is 'cut-and paste' for Windows. It aids to collect similar text segments from the raw textual data and AUTOCODE makes a coding-rule-file. This file can be used to confirm the precision of coding. AUTOCODE program codes textual data in accordance with the coding-rule-file and outputs the file that can be analyzed with statistical package soft such as SPSS. This program has graphical user interface and a mouse can be used.KTCoder divides the raw textual data into separate segments and outputs the files that include the locations of each code. All files used and created by the program that contain code with information of location data are stored in standard database format. This offers users extensive possibilities for using other software tools, for instance, Access. This program has also graphical user interface and a mouse can be used.

本研究的目的是展示用于文本数据分析的新的计算机辅助技术。到目前为止，在日本，只有很少的尝试在个人计算机上进行内容分析。主要原因是日本人有一个很难解决的问题。日本人没有明确的根据词性对词进行分类的界限。然后，我们的关注点是制作新的计算机程序来对词进行分类。首先是如何将大量文本转换为机器可读格式。第二是如何对文本数据进行编码。三是分析方法，重点是在个人计算机上制作日文文本资料编码软件。我们已经完成了两个程序：AUTOCODE程序和KTCoder。AUTOCODE程序最初是为了支持调查问卷中开放式问题的编码而开发的。它的基本功能是“剪切和粘贴”的Windows。它有助于从原始文本数据中收集类似的文本段，AUTOCODE生成一个编码规则文件。此文件可用于确认编码的精度。AUTOCODE程序根据编码规则文件对文本数据进行编码，并输出可供SPSS等统计软件分析的文件。该程序具有图形用户界面，可以使用鼠标。KTCoder将原始文本数据划分为单独的段，并输出包含每个代码位置的文件。程序使用和创建的所有文件（包含带有位置数据信息的代码）都以标准数据库格式存储。这为用户使用其他软件工具提供了广泛的可能性，例如Access。这个程序也有图形用户界面和鼠标可以使用。