碩士 === 國立彰化師範大學 === 數學系 === 106 === Differential item functioning (DIF) is one of the important indices on detecting item’s fairness. This study applies sign test to detect nonuniform DIF generated by a cognitive diagnostic DINA model. DINA model separates examinees into masters and non-masters. Non...

Full description

Bibliographic Details
Main Author: 林家瑜
Other Authors: 李信宏
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/ru4ap3
id ndltd-TW-106NCUE5479004
record_format oai_dc
spelling ndltd-TW-106NCUE54790042019-05-30T03:57:14Z http://ndltd.ncl.edu.tw/handle/ru4ap3 無 認知診斷DINA模式之差異試題功能(nonuniform DIF)分析 林家瑜 碩士 國立彰化師範大學 數學系 106 Differential item functioning (DIF) is one of the important indices on detecting item’s fairness. This study applies sign test to detect nonuniform DIF generated by a cognitive diagnostic DINA model. DINA model separates examinees into masters and non-masters. Nonuniform DIF is found to occur when an item is beneficial to both masters from the reference group and non-masters from the focal group. It’s implied that the directions of the influence of DIF are not the same. A nonparametric sign test based on the odds ratios of item responses of the focal and reference groups is used to investigate the nonuniform DIF. The factors of simulation study contain the correlation of attributes, item parameters, sample sizes and the amount of DIF. The average type I error rates are approximated to the significant level of 0.05. Meanwhile, the powers rise significantly as the sample size and the amount of DIF increase. When the required attributes of the item increase, the differences between two groups cannot be well identified and the results of the sign test are often viewed as uniform DIF instead of nonuniform DIF. 李信宏 2018 學位論文 ; thesis 36 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立彰化師範大學 === 數學系 === 106 === Differential item functioning (DIF) is one of the important indices on detecting item’s fairness. This study applies sign test to detect nonuniform DIF generated by a cognitive diagnostic DINA model. DINA model separates examinees into masters and non-masters. Nonuniform DIF is found to occur when an item is beneficial to both masters from the reference group and non-masters from the focal group. It’s implied that the directions of the influence of DIF are not the same. A nonparametric sign test based on the odds ratios of item responses of the focal and reference groups is used to investigate the nonuniform DIF. The factors of simulation study contain the correlation of attributes, item parameters, sample sizes and the amount of DIF. The average type I error rates are approximated to the significant level of 0.05. Meanwhile, the powers rise significantly as the sample size and the amount of DIF increase. When the required attributes of the item increase, the differences between two groups cannot be well identified and the results of the sign test are often viewed as uniform DIF instead of nonuniform DIF.
author2 李信宏
author_facet 李信宏
林家瑜
author 林家瑜
spellingShingle 林家瑜

author_sort 林家瑜
title
title_short
title_full
title_fullStr
title_full_unstemmed
title_sort
publishDate 2018
url http://ndltd.ncl.edu.tw/handle/ru4ap3
work_keys_str_mv AT línjiāyú wú
AT línjiāyú rènzhīzhěnduàndinamóshìzhīchàyìshìtígōngnéngnonuniformdiffēnxī
_version_ 1719195730487279616