Detail Misrepresentative Language Samples and 'totl   Oleh:  Miller, G.R. Jenis: Article from Bulletin/Magazine Dalam koleksi: Journal of Verbal Learning and Verbal Behavior vol. 21 no. 6 (Dec. 1982), page 666-669. Ketersediaan Perpustakaan PKBB Nomor Panggil: 405/JVL/21 Non-tandon: tidak ada Tandon: 1  Lihat Detail Induk Isi artikel Coleman (1979) noted that if one considers any single language experiment to be a finite sample from a larger population then the correlation between the population Treatment effects and the population Treatment-by-Language effects can cause the actual probability of either Type 1 or Type 2 errors to be greater than the calculated value. The type of error is determined by whether the correlation is positive or negative. Marler (1981) thought he had discovered a flaw in Coleman's logic by computing two correlation between Treatment effects and Treatment-by-language effects which always equal zero. The present article demonstrates that both of Marler's correlations are irrelevant. Opini Anda Klik untuk menuliskan opini Anda tentang koleksi ini! Kembali

Copyright © 2006, 2007 Unika Atma Jaya, all rights reserved