[绝密版第八章t检验.ppt

第8章 t 检 验 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 主要内容 一、样本均数与总体均数的比较 三、 两样本均数的比较 四、正态性检验与方差齐性检验 五、t′检验 二、配对设计均数的比较 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. t检验和z检验 t 检验的应用条件： z 检验应用条件： ⑴ 总体标准差 未知； ⑵ 样本含量n 较小（n 100） ； ⑶ 样本来自正态总体； ⑷ 两样本均数比较时方差齐， 即 样本含量n 较大（ n≥100） (2) n 虽小但总体标准差 已知 （不常见）。 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 应用类型： 样本均数与总体均数的比较 配对t 检验 成组设计两样本均数的比较 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 一、样本均数与总体均数的比较 ( One-sample test ) 目的：推断样本均数代表的未知总体均数 µ 与已知总体均数 µ0 （一般为理论值、 标准值或经大量观察所得的稳定值等） 有无差别 条件：理论上要求资料来自正态分布总体 Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 在 H0 成立的前提条件下，检验统计量计算公式： ① σ已知或σ未知但n足够大: ② σ未知且n较小： Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. Evaluation only. Created with Aspose.Slides for .NET 3.5 Client Profile 5.2.0.0. Copyright 2004-2011 Aspose Pty Ltd. 例8-1 根据大量调查得知，某地20岁健康成年男子平均身高为170cm，标准差为8.0cm。今随机抽查了该地25名健康成年男子，求得其身高均数为172cm，标准差为8.6cm，能否据此认为该地现在20岁成年男子平均身高与以往不同？ [分析]根据题意，实际是观察2