多元計分三參數試題選項分析變動效應模式之模擬研究
碩士 === 臺中師範學院 === 教育測驗統計研究所 === 90 === Abstract The purpose of this study is to combine non-parameter ability estimation and parameter model to estimate item parameters in large-scale test. This study use simulation testing method on the basis of item numbers and sample sizes....
Main Author: | |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2002
|
Online Access: | http://ndltd.ncl.edu.tw/handle/22556938231993973433 |
id |
ndltd-TW-090NTCTC629007 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-090NTCTC6290072015-10-13T10:31:15Z http://ndltd.ncl.edu.tw/handle/22556938231993973433 多元計分三參數試題選項分析變動效應模式之模擬研究 丁振祥 碩士 臺中師範學院 教育測驗統計研究所 90 Abstract The purpose of this study is to combine non-parameter ability estimation and parameter model to estimate item parameters in large-scale test. This study use simulation testing method on the basis of item numbers and sample sizes. According to results, some conclusions are listed below: 1. Under the same item numbers, the more examinees make discrimination parameter and parameter of guessing determination power more accurate. 2. Under the same sample size, the more item numbers make parameter of guessing determination power the more accurate. And discrimination parameter, guessing parameter, correct power parameter, and distractibility parameter are no discrimination between good and bad. 3. Parameters used in this study are more than general parameter model. There has covariance between parameter estimation. The results reveal the large MSD of all parameter estimation. So it’s hard to find out the general estimation under the more item numbers and examinees. Keywords: non-parameter estimation for ability, parameter model, item parameters, number of examines 許天維 2002 學位論文 ; thesis 0 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 臺中師範學院 === 教育測驗統計研究所 === 90 === Abstract
The purpose of this study is to combine non-parameter ability estimation and parameter model to estimate item parameters in large-scale test. This study use simulation testing method on the basis of item numbers and sample sizes.
According to results, some conclusions are listed below:
1. Under the same item numbers, the more examinees make discrimination parameter and parameter of guessing determination power more accurate.
2. Under the same sample size, the more item numbers make parameter of guessing determination power the more accurate. And discrimination parameter, guessing parameter, correct power parameter, and distractibility parameter are no discrimination between good and bad.
3. Parameters used in this study are more than general parameter model. There has covariance between parameter estimation. The results reveal the large MSD of all parameter estimation. So it’s hard to find out the general estimation under the more item numbers and examinees.
Keywords: non-parameter estimation for ability, parameter model, item parameters, number of examines
|
author2 |
許天維 |
author_facet |
許天維 丁振祥 |
author |
丁振祥 |
spellingShingle |
丁振祥 多元計分三參數試題選項分析變動效應模式之模擬研究 |
author_sort |
丁振祥 |
title |
多元計分三參數試題選項分析變動效應模式之模擬研究 |
title_short |
多元計分三參數試題選項分析變動效應模式之模擬研究 |
title_full |
多元計分三參數試題選項分析變動效應模式之模擬研究 |
title_fullStr |
多元計分三參數試題選項分析變動效應模式之模擬研究 |
title_full_unstemmed |
多元計分三參數試題選項分析變動效應模式之模擬研究 |
title_sort |
多元計分三參數試題選項分析變動效應模式之模擬研究 |
publishDate |
2002 |
url |
http://ndltd.ncl.edu.tw/handle/22556938231993973433 |
work_keys_str_mv |
AT dīngzhènxiáng duōyuánjìfēnsāncānshùshìtíxuǎnxiàngfēnxībiàndòngxiàoyīngmóshìzhīmónǐyánjiū |
_version_ |
1716828767139135488 |