中文摘要：

背景：条目反应理论越来越多应用于量表或问卷的研发和评价,但其原理深奥、公式繁琐,使得该理论难于推广。学者们为了推广该理论,研制了一系列条目反应理论软件。Bilog-MG软件是处理二分类资料的条目反应理论软件,可通过书写程序或页面的点击的方式实行操作。目的：采用实例介绍如何使用Bilog-MG软件。方法：选择鼻咽癌患者SF-36量表生理职能的4个条目,共纳入532例患者。Bilog-MG软件的分析步骤主要包括定义基本情况和分析模型,定义分析条目和估计方法,定义测试评分,定义数据基本情况,定义条目的正确答案,保存数据和生成语法。结果与结论：（1）分析过程中需要重点定义整个模型的基本情况（测试数量、分组情况、测试条目等）,定义分析模型（单参数、双参数、三参数模型）,定义数据基本情况和数据格式;（2）分析结果包括条目统计量、条目参数估计值和整个量表的信度系数;还给出以下几个图形：条目特征曲线、条目信息曲线、总体信息曲线等;（3）Bilog-MG软件优势明显,简单易学,能够精确估计条目参数和能力参数;希望Bilog-MG软件能够被医学研究者广泛应用,也对推广条目反应理论起到一定的作用。

英文摘要：

BACKGROUND： Item response theory is gradually used in the scale and questionnaire. However, the principles are profound and complicated, which makes it difficult to popularize. Therefore, the researchers have developed a series of softwares such as Bilog-MG software to deal with two-category data through writing program or clicking on the page. OBJECTIVE： To introduce the Bilog-MG software using an example. METHODS： Four items of physical role in the SF-36 scale were selected, and a total of 532 patients with nasopharyngeal carcinoma were enrolled for analysis. The main steps of Bilog-MG software included： basic definition and analysis model, definition of item analysis and estimation methods, definition of test score, definition of the data, definition of the correct answer, saving the data, and generating grammar. RESULTS AND CONCLUSION：（1） The steps of Bilog-MG software focused on the definition of the model（the number of test, grouping, items of test）, definition of the analysis model（single parameter, double parameters, three parameters model）, and definition of data and format.（2） The results of analysis included： item statistics, item parameter estimation values and the reliability coefficient. It also included the following graphics： item characteristic curves, item information curve and overall information curve.（3） In conclusion, the Bilog-MG software is easy to learn, and can accurately estimate the item and ability parameters. We hope Bilog-MG software can be widely used in medical research.