统计学实验八分析季节变动：懿说学区4

统计学实验八分析季节变动：懿说学区4The question now is how to judge which assumption needs to be rejected and which assumption needs to be accepted. Here the concepts of rejection region and critical point are extended. When the test statistic takes the value of a variable and rejects the original hypothesis the region is called rejection region and the boundary point of rejection region is called critical point.那么现在的问题在于，如何判断哪一

懿说学区（4） | SPSS统计分析（13）假设检验（二）

Yishuo school district (4) | SPSS statistical analysis (13) hypothesis test (2)

“分享兴趣，传播快乐，增长见闻，留下美好！ 大家好，这里是小编。欢迎大家继续访问学苑内容，我们将竭诚为您带来更多更好的内容分享。

"Share your interests spread happiness increase your knowledge and leave something beautiful! Hello everyone this is Xiaobian. Welcome to continue to visit the content of the school and we will wholeheartedly bring you more and better content sharing.

今天，我们继续讲述统计学中的一个非常重要的概念——假设检验， 上期我们说到，在进行假设检验的时候我们需要设置两个假设：原假设和备择假设，原假设和备择假设互相排斥，并且同时只有一个正确。

Today we continue to talk about a very important concept in Statistics - hypothesis testing. In the last issue we said that we need to set up two hypotheses during hypothesis testing: the original hypothesis and the alternative hypothesis. The original hypothesis and the alternative hypothesis are mutually exclusive and only one is correct at the same time.

那么现在的问题在于，如何判断哪一个假设是需要拒绝的，而哪一个假设是需要接受的。这里就引申出了拒绝域和临界点的概念，当检验统计量取某个变量中的值的时候，拒绝原假设，则称该区域为拒绝域，称拒绝域的边界点为临界点。

The question now is how to judge which assumption needs to be rejected and which assumption needs to be accepted. Here the concepts of rejection region and critical point are extended. When the test statistic takes the value of a variable and rejects the original hypothesis the region is called rejection region and the boundary point of rejection region is called critical point.

由于统计学是基于概率论的一门学科，在概率的领域中，一切都不是绝对的。因此，假设检验偶尔也会发生错误。

Because statistics is a subject based on probability theory everything is not absolute in the field of probability. Therefore the hypothesis test occasionally makes mistakes.

我们将假设检验的错误归结于两类：

（1） 第一类错误：在假设检验中拒绝了本来是正确的原假设，我们称这类错误为α错误或第一类错误，即当原假设正确的时候我们却认为它错了，因此拒绝了原假设，也称“弃真错误”。

（2） 第二类错误：在假设检验中接受了错误的原假设，我们称这类错误为β错误或第二类错误，即当原假设是错误的时候，我们通过检验却认为它是正确的，因此接受了原假设，也称“取伪错误”。

We attribute the errors of hypothesis testing to two categories:

（1） The first kind of error: the original hypothesis that is correct is rejected in the hypothesis test. We call this kind of error α Error or the first kind of error that is when the original assumption is correct we think it is wrong so we reject the original assumption also known as "false error".

（2） The second kind of error: we accept the wrong original hypothesis in the hypothesis test. We call this kind of error β Error or the second kind of error that is when the original assumption is wrong we think it is correct through testing so we accept the original assumption also known as "false error".

而在进行假设检验的时候，我们可以接受的犯第一类错误的最大概率称为检验的显著性水平，这个概率常记为α。显著性水平α对假设检验的结论有直接影响，通常在抽样前就要求指定好，得到的结果才不影响我们的选择。α为置信度或置信水平。

In hypothesis testing the maximum probability that we can accept to make the first type of error is called the significance level of the test and this probability is often recorded as α。 Significance level α It has a direct impact on the conclusion of hypothesis test. It is usually required to specify before sampling so that the results obtained will not affect our choice. α Is the confidence level or confidence level.

SPSS

在进行假设检验后，通常我们能够得到一个概率P值，所谓的概率P值是当原假设正确的时候，观测到的样本信息出现的概率。P值越小，拒绝原假设的理由就越充分。我们判断P值是否足够的小，通常是和预先设定的显著性水平α值比较，若α值为0.05，则P值小于0.05才认为概率值很小，即原假设“不显著”，应当拒绝原假设。

After the hypothesis test we can usually get a probability p value. The so-called probability p value is the probability of the observed sample information when the original hypothesis is correct. The smaller the p value the more sufficient the reason to reject the original hypothesis. We determine whether the p value is small enough usually at a predetermined significance level α Value comparison if α If the value is 0.05 the probability value is considered to be very small when the p value is less than 0.05 that is the original assumption is "not significant" and the original assumption should be rejected.

假设检验同时具有方向性，这点的差异将假设检验分为两种类型：

（1） 双侧检验：只强调差异而不强调方向性的检验。

（2） 单侧检验：强调某一方向的检验。

Hypothesis tests have directionality at the same time. This difference divides hypothesis tests into two types:

（1） Bilateral test: a test that emphasizes only differences but not directionality.

（2） One side inspection: it emphasizes the inspection in a certain direction.

下期预告：本节我们继续深入了解假设

检验的理论知识，下一期，我们将为您

总结假设检验的一般步骤，以及用案例

亲自演示假设检验如何进行。

Preview for the next issue: in this section we will continue to understand the theoretical knowledge of hypothesis testing. In the next issue we will summarize the general steps of hypothesis testing and personally demonstrate how to conduct hypothesis testing with cases.

今天的分享就到这里了，如果您对今天的文章有独特的想法，欢迎给我们留言，让我们相约明天，祝您今天过得开心快乐！

That's all for today's sharing. If you have unique ideas about today's article please leave us a message. Let's meet tomorrow. I wish you a happy day today!

参考资料：百度百科，《SPSS 23统计分析实用教程》

翻译：百度翻译

本文由LearningYard新学苑原创，部分文字图片来源于他处，如有侵权，请联系删除。