The difference between paired and unpaired
https://zhuanlan.zhihu.com/p/68736699
This article is very well written:
The premise of the t test is:
1. T test is a parametric test and is used to test quantitative data (numbers have comparative significance). If the data are all definite numbers, use non-parametric test. [What is fixed data? 】
2. The sample data obeys normal or approximately normal distribution. [This is easy to understand, the sample must obey the normal distribution. 】
(1) One-sample t test
The one-sample T test is used to compare whether there is a difference between the sample data and a specific value.
(2) Independent sample T test (T test)
The independent sample T test is used to analyze the difference between qualitative data (X) and quantitative data (Y).
(3) Paired sample T test
Used to analyze the difference and contrast relationship between paired quantitative data. Compared with the independent sample t test, the paired sample t test requires the samples to be paired. The sample size of the two samples should be the same; the order of the samples is one-to-one.
For example, compare the two contexts (with advertisements and without advertisements); whether there are obvious differences in the purchase intention of the samples. Through the comparative analysis of the two sets of data, it is judged whether the background music will affect the consumption behavior.