Why Is The Intercept Not Significant?

1772

收藏 2014-01-09

I'm working on a two level model. Analyzing null model revealed that random intercept is not significant (it's -0.0059).So, I want to know:

How can I interpret it? (when the intraclass correlation coefficient is 5.2%,sigma-square 0.9554 (0.028),tau 0.0522(.013))
What should I do next?

My main goal is to compare a multilevel item response theory(MLIRT) model with traditional multilevel model(ML) using TIMSS data. For MLIRT I use mlirt package under R developed by Fox (2003) and for ML I use HLM software. Because ability estimates used
in MLIRT are on a scale with mean of 0 and SD of 1, I transformed plausible values to a scale with mean of 0 and SD of 1 then the estimates could be comparable. Sample size is 2362 students nested within 116 schools.
With "non significant random intercept" I meant fixed part of model.The variance component of level-2 (Tau) is significant based on the
chi-square test provided in the output by HLM and also fiexd part. But random intercept (Gamma00 or G00) using MLIRT is not significant (G00=-0.0059 (0.0298)).FYI, model specifications and data are the same across methods.

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

全部回复

Trevor

2014-1-9 11:50:01

Authors:	Jean-Paul Fox
Title:	[download] (9148)Multilevel IRT Modeling in Practice with the Package mlirt
Reference:	Vol. 20, Issue 5, Feb 2007Submitted 2006-10-01, Accepted 2007-02-22
Type:	Article
Abstract:	Variance component models are generally accepted for the analysis of hierarchical structured data. A shortcoming is that outcome variables are still treated as measured without an error. Unreliable variables produce biases in the estimates of the other model parameters. The variability of the relationships across groups and the group-effects on individuals' outcomes differ substantially when taking the measurement error in the dependent variable of the model into account. The multilevel model can be extended to handle measurement error using an item response theory (IRT) model, leading to a multilevel IRT model. This extended multilevel model is in particular suitable for the analysis of educational response data where students are nested in schools and schools are nested within cities/countries.
Paper:	[download] (9148)Multilevel IRT Modeling in Practice with the Package mlirt (application/pdf, 309.5 KB)
Supplements:	[download] (2498)Data.zip: Data sets in SPSS format (application/zip, 2 MB)
	[download] (2567)mlirt_1.0.tar.gz: R source package (application/x-gzip, 682.5 KB)
	[download] (2434)v20i05.R: R example code from the paper (application/zip, 2 KB)
Resources:	BibTeX \| OAI

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

扫码加我 拉你入群

扫码加我 拉你入群

分享

扫码加好友，拉您进群

扫码加我拉你入群

扫码加我拉你入群