1. Case introduction
A hospital used two different methods to diagnose 53 cases of lung cancer patients. The collected results are shown in the table below. Now I would like to know whether there is any difference in the detection results of the two methods.
2. Problem Analysis
The purpose of this case analysis is to compare whether the two methods have different test results for the same batch of samples, and the test results are binary variables (positive or negative). For this kind of situation, paired chi-square test can be used for research, which needs to meet 3 conditions:
Condition 1: The observed variable is a binary categorical variable.
Condition 2: The observed data is a paired design.
This case satisfies these two conditions, so the paired chi-square test can be used for analysis.
3. Software operation and result interpretation
(1) Upload data
First organize the data into the correct format of the paired chi-square test. There are 2 results in method A, 2 results in method B, and 4 combinations in 2*2. The number of samples in each combination is marked in a separate column as the weight value. Finally, the case data is organized into the following format:
Upload the sorted data to the SPSSAU system, click the upload data button in the upper right corner, and upload the data according to the prompts, the operation is as follows:
After the data upload is complete, the paired chi-square test is started.
(2) Software operation
In the SPSSAU medical/experimental research module, select [Paired Chi-Square], drag "Method A" to the analysis box on the right pair 1 (classification), and drag "Method B" to the right pair 2 (classification) ) analysis box, drag "Number of People" to the right "Weighted Item" analysis box, and then click "Start Analysis", the operation is as shown in the figure below:
(3) Interpretation of results
The SPSSAU output paired chi-square analysis results are as follows:
From the results of paired chi-square analysis, it can be seen that p=0.0225<0.05, and at the level of α=0.05, rejecting the null hypothesis and accepting the alternative hypothesis, it can be considered that the difference between the test results of the two methods is statistically significant. The specific differences can be compared through specific numbers, as shown in the following table:
It can be seen from the above table that the positive detection rate of method A is 27/53=50.94%, and the positive detection rate of method B is 36/53=67.92%. It can be considered that the positive detection rate of method B is higher than that of method A. You can also use statistical graphs for visual comparison, as shown in the following figure:
4. Conclusion
In this case, paired chi-square test is used to analyze whether there is any difference in the detection results of lung cancer between methods A and B. According to the study, the corresponding p value of the paired chi-square test is 0.0225, which is less than 0.05, so the null hypothesis is rejected, and the difference between the two test results can be considered to be statistically significant. The specific difference can be obtained by comparing the numbers, and the positive detection rate of method B is 67.92 % higher than the positive detection rate of Method A of 50.94%.
5. Knowledge Tips
1. Data format description:
In addition to using the weighted format of this case, it is also possible to use the original data format for analysis. In raw data format, with a total of 53 lung cancer patients, there would be 53 rows, each representing a patient.
2. Paired chi-square test type:
If the group of the paired data is 2, that is, the paired four tables, use the McNemar test; as in this case, when both methods have only positive and negative results, that is, the 2*2 structure type;
If the group of paired data is greater than 2, that is, paired multi-category, the Bowker test is used.