Dimension coefficient yielded by Rasch model to measure scale validity: An example using the patient experience diagnostic tool for England hospitals

CNU IR > College of Smart Living > Dept. of Hospital and Health (including master's program) > Periodical Articles > Item 310902800/27707

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://ir.cnu.edu.tw/handle/310902800/27707

题名:	Dimension coefficient yielded by Rasch model to measure scale validity: An example using the patient experience diagnostic tool for England hospitals
作者:	Chien, Tsair-Wei Wang, Wen-Chung Chou, Ming-Ting
贡献者:	醫務管理系
关键词:	Unidimensionality Principle Component Analysis Rasch Model National Health Service
日期:	2012-09
上传时间:	2014-03-21 16:18:19 (UTC+8)
出版者:	Drunpp-Sarajevo
摘要:	Background: Dimensionality must be determined before adding item scores to represent the overall performance of a measure. The aim of this study is to help researchers identify the dimension tendency for a scale's validity toward a common entity. This in turn will improve inspecting aberrant responses for each item with individual item-by-item box plots of disclosure for patient views on healthcare service quality.Methods: We used the Rasch rating scale model to analyse the 2009 English inpatient questionnaire data regarding patient satisfactory perception, which were collected from 162 hospitals, examined unidimensionality, and developed a visual plot in Excel that depicts the satisfaction level of each hospital across questions and monitors aberrant responses with Rasch standardised residual for each item and outfit statistics for individual hospitals.Results: We found that (1) the Rasch_residual principal component analysis is able to check dimensionality. The dimension coefficient defined as the extent of dimensionality should report to readers. The five key domains of the 2009 English adult inpatient questionnaire data regarding patient experience measure a common entity and earn a dimension coefficient of 0.88, and (2) item-by-item chart plots along with Rasch's fit statistics (i.e., standardised residual and outfit statistics) can help organisations accomplish large improvements on a small number of key areas.Conclusion: Without removing misfit questions, we found that the 20-item inpatient questionnaire measured the same construct across types of hospitals. The visual chart plot in Excel provides an exemplary comparison of quality of healthcare for individual hospitals. The Rasch analysis allows intra- (using standardised residuals for each item) and inter- (using outfit statistics across hospitals) hospital performances to be easily, quickly and clearly compared.
關聯:	Healthmed, v.6 n.5 pp.1512-1522
显示于类别:	[Dept. of Hospital and Health (including master's program)] Periodical Articles

文件中的档案:

没有与此文件相关的档案.

在CNU IR中所有的数据项都受到原著作权保护.

TAIR相关文章

数据加载中.....