Correcting a Significance Test for Clustering  

Oleh:

Hedges, Larry V.

Jenis: Article from Journal - ilmiah internasional
Dalam koleksi: Journal Of Educational And Behavioral Statistics vol. 32 no. 2 (Jun. 2007), page 151-179.
Topik: cluster-randomized trials; significance tests; intraclass correlations; multilevel models
Fulltext: 151.pdf (392.74KB)

Isi artikel

A common mistake in analysis of cluster randomized trials is to ignore the effect of clustering and analyze the data as if each treatment group were a simple random sample. This typically leads to an overstatement of the precision of results and anticonservative conclusions about precision and statistical significance of treatment effects. This article gives a simple correction to the t statistic that would be computed if clustering were (incorrectly) ignored. The correction is a multiplicative factor depending on the total sample size, the cluster size, and the intraclass correlation ?. The corrected t statistic has Student’s t distribution with reduced degrees of freedom. The corrected statistic reduces to the t statistic computed by ignoring clustering when ? ¼ 0. It reduces to the t statistic computed using cluster means when ? ¼ 1. If 0

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali