懿说学区（5） | SPSS统计分析（14）假设检验（三）
Yishuo school district (5) | SPSS statistical analysis (14) hypothesis test (III)
“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.
在前几期之中，我们了解到了假设检验的一些理论知识，今天，我们将接触到如何正确地进行假设检验的一些操作。
假设检验是对给定总体参数值，利用样本数据对其推断，并给出接受或拒绝的过程。假设检验依据“小概率事件实际不可能原理”，如果发生了小概率事件，我们就有理由怀疑假设的正确性，从而拒绝假设检验的原假设。
In the previous issues, we learned some theoretical knowledge about hypothesis testing. Today, we will be exposed to some operations on how to correctly conduct hypothesis testing.
Hypothesis testing is a process of inferring a given population parameter value using sample data and giving an acceptance or rejection. The hypothesis test is based on the "principle of the actual impossibility of small probability events". If a small probability event occurs, we have reason to doubt the correctness of the hypothesis and reject the original hypothesis of the hypothesis test.
在实际操作之中，我们首先应该定义所谓的小概率，一般取0.01或0.05，即显著性水平。显著性水平取值太小容易发生取伪错误，取值太大容易发生弃真错误。所以我们可以归纳总结出进行假设检验的一般性步骤：
In practice, we should first define the so-called small probability, which is generally taken as 0.01 or 0.05, that is, the significance level. If the value of significance level is too small, it is likely to be false, and if the value is too large, it is likely to be false. Therefore, we can summarize the general steps for hypothesis testing:
1
给出检验问题的原假设和备择假设
The first step is to give the original hypothesis and alternative hypothesis of the test problem
根据检验问题的要求，将需要检验的最终结果作为原假设。例如：需要检验某学校的高考数学平均成绩是否同往年的平均成绩一样，都为75，由此可做出原假设H0：μ=75。
For example, it is necessary to test whether the average score of mathematics in the college entrance examination of a school is the same as that in previous years, which is 75. Therefore, the original assumption H0 can be made: μ= 75。
2
选择检验统计量
Step 2: select test statistics
在统计推断中，总是通过构造样本的统计量的概率值进行推断，一般构造的统计量应该服从或近似服从常用的已知分布，例如均值检验中常用的T分布和F分布等。
In statistical inference, it is always inferred by constructing the probability value of the sample statistic. Generally, the constructed statistic should obey or approximately obey the commonly used known distribution, such as the t distribution and F distribution commonly used in the mean test.
3
规定显著性水平
Step 3: specify the significance level
显著性水平指的是当假设正确被拒绝时的概率，即弃真概率，一般取0.01或0.05。
The significance level refers to the probability when the hypothesis is rejected, i.e. the probability of discarding the truth, which is generally taken as 0.01 or 0.05.
4
计算检验统计量的观测值及其发生
的概率值
The fourth step is to calculate the observed value of the test statistic and its occurrence probability.
在给定原假设前提下，计算统计量的观测值和相应概率P值。概率P值就是在原假设H0成立时检验统计量的观测值发生的概率，该概率值间接地给出了样本值在原假设成立前提下出现的概率，对此可以依据一定的标准来判断其发生的概率是否为小概率。
Given the original hypothesis, the observed value of the statistic and the corresponding probability p value are calculated. The probability p value is the probability of occurrence of the observed value of the test statistic when the original hypothesis H0 is established. This probability value indirectly gives the probability of occurrence of the sample value on the premise that the original hypothesis is established. For this, we can judge whether the probability of occurrence is a small probability according to certain standards.
5
在给定显著性水平条件下，作出统计
推断结果
Step 5: make statistical inference results under the given significance level.
当检验统计量的概率P值小于显著性水平时，则认为拒绝原假设而犯弃真错误的概率小于显著性水平，即低于预先给定的水平，也就是说，犯错误的概率小到我们能容忍的范围，这时可以拒绝原假设；反之，当检验统计量的概率P值大于显著性水平时，则认为拒绝原假设而犯弃真错误的概率大于预先给定的容忍水平，这时不应该拒绝原假设。
When the probability p value of the test statistic is less than the significance level, it is considered that the probability of rejecting the original hypothesis and making the false error is less than the significance level, that is, it is lower than the predetermined level, that is, the probability of making the error is small enough to be tolerated, and the original hypothesis can be rejected; On the contrary, when the probability p value of the test statistic is greater than the significance level, it is considered that the probability of rejecting the original hypothesis and making the false error is greater than the preset tolerance level, and the original hypothesis should not be rejected.
下期预告：本节我们深入了解了假设检验的一般步骤和实际操作，我们对于假设检验的探讨就到此结束了，下一节，我们将讲述单样本T检验的内容。
Forecast for next issue: in this section, we have thoroughly understood the general steps and actual operations of hypothesis testing. This concludes our discussion on hypothesis testing. In the next section, we will talk about the content of single sample t-test.
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!
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。
#
如若转载，请注明出处：https://www.dasum.com/107223.html