Objective: To test the significance of differences between the two have normal population mean of contact exists.
Conditions: There is contact, normal population, the sample size should be the same. They can generally be divided into four categories:
① Comparative same subject before and after treatment : as for diabetes, and the same group of patients measured blood glucose levels prior to use new methods of treatment, blood glucose levels measured again after use, is formed between two groups of samples; to indicate whether there is a general effect .
② different portion of the same subject data
③ results for the same samples tested by two methods (instruments, etc.);
④ two subjects receiving data respectively paired two treatment: diabetic pair of body weight ( two pairs of 60, two pairs 65 ......) and different methods of treatment of patients pairing .
case analysis:
Case Description: test before tea and after tea if the mean body weight of the occurrence of a significant change to determine weight loss weight loss tea. ( Source: "Statistical Analysis with SPSS Chapter V of the application" Xue Wei )
Topic Analysis: before and after the change in body weight and about tea, while overall approximate normal distribution, the use of two paired samples t-test.
Interface Procedure: Open Data - - Comparative Analysis Means - Paired t test - setting parameters - output
Screenshot key steps:
How can a pair of variables
Result analysis:
Paired sample statistics |
|||||
|
Means |
N |
Standard deviation |
The standard error of the mean |
|
Of 1 |
Weight before tea |
89.2571 |
35 |
5.33767 |
.90223 |
After drinking weight |
70.0286 |
35 |
5.66457 |
.95749 |
Paired sample correlation coefficient |
||||
|
N |
The correlation coefficient |
Sig. |
|
Of 1 |
Weight before tea & drink after weight |
35 |
-.052 |
.768 |
= 0.768 Sig> 0.05 , shows that body weight before and after taking slimming tea, and no significant changes linearly, the degree of linear correlation is weak.
Paired sample test |
|||||||||
|
Differential pairs |
t |
df |
Sig. ( Bilateral ) |
|||||
Means |
Standard deviation |
The standard error of the mean |
The difference 95% confidence interval |
||||||
Lower limit |
Upper limit |
||||||||
Of 1 |
Tea weight before - after drinking weight |
19.22857 |
7.98191 |
1.34919 |
16.48669 |
21.97045 |
14.252 |
34 |
.000 |
Analysis single sample t test similar.
Note: The higher correlation coefficient before and after tea, explained the role of tea is actually not that much.
Reference books:
"Statistical Analysis with SPSS applications" (Fifth Edition) Xue Wei
" SPSS statistical analysis from scratch," Wu Jun
" SPSS statistical analysis based tutorial" Zhang Wentong