日本医科大学医学会雑誌 Online Journal メインナビゲーションを飛ばす
総目次 > 号目次 > Abstract ホームへ戻る
Abstract

第10巻 2014年4月 第2号

全文PDF (437K)

■基礎科学から医学・医療を見る

検定の考え方,独立性の検定
儀我 真理子
日本医科大学基礎科学数学

Statistical Tests, Test of Independence
Mariko Giga
Department of mathematics, Nippon Medical School

In this paper we describe statistical test. This is the most important statistical field for examining the mean, variance and other characteristics of population by using sample. First, we propose a null hypothesis, for example, that the population mean is a certain value. With statistical analysis, if we conclude that the null hypothesis is to be rejected, we judge that the hypothesis is probably not true. When we conclude that the null hypothesis is to be accepted, we can only say that we cannot reject it. In this paper, we also describe a test of independence. This is a method of testing, for example, whether we can recognize a difference in the numbers of male and female patients with certain diseases. The main method of testing for independence is not applicable to small samples, for which other methods must be used.

日医大医会誌 2014; 10(2), 115-119

Correspondence to
Mariko Giga, Department of Mathematics, Nippon Medical School, 2-297-2 Kosugi-cho, Nakaharaku, Kawasaki, Kanagawa 211-0063, Japan

受付:2014年1月16日 受理:2014年3月17日

メインナビゲーションへ戻る このページのトップへ戻る