Distinction between tests of data or theory: Null versus disconfirming result  

Oleh:

Harcum, E. Rae

Jenis: Article from Journal - ilmiah internasional
Dalam koleksi: The American Journal of Psychology vol. 103 no. 3 (1990), halaman 359.
Topik: Theory; Test; Data
Fulltext: 1423215.pdf (836,21KB)

Isi artikel

An important distinction has been identified between a research hypothesis, which is statistically tested from the data, and the theory from which that hypothesis is derived. Statistical support or nonsupport for the research hypothesis does not therefore necessarily mean the same level of evidence for or against the theory. Because the research hypothesis can predict an empirical difference (significant results), or no difference (null results), the terms significanta nd null results should be restricted to conclusions about empirical data, and the terms confirminga nd disconfirminrge sults should be used for outcomes that influence decisions with respect to a hypothesis and theory. Because of the inherent confusability,t he terms positivea nd negative should not be used in either context.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali