Chi-squared Test - Hands-on

Step1of4

Installations

Let's get the installations done, prior to creating tasks.

Run the command given below.

pip install --user numpy scipy

prettytable

Editor Usage

Instructions

1.Open app.py file using vim editorvim app.py

2.Pressifor inserting content into file.

3.Write the required python code.

4.Save the file by pressingEsckey and typing:wqin editor.

5.Execute the script with commandpython app.py | tee output.txt

6.Click on Continue to move to next Task.

Do the Hypothesis Testing

The table shows the contingency table of marital status by education. Use Chi-Square test for testing Homogenity.contingency table of marital status by education.

View the table by executing the following command

python

from prettytable import PrettyTable

t = PrettyTable(['Marital Status','Middle school', 'High School','Bachelor','Masters','PhD'])

t.add_row(['Single',18,36,21,9,6])

t.add_row(['Married',12,36,45,36,21])

t.add_row(['Divorced',6,9,9,3,3])

t.add_row(['Widowed',3,9,9,6,3])

print (t)

exit()

Hypothesis

Null Hypothesis: There is no difference in distribution between the types of education level in terms of marital status.

Alternate Hypothesis: There is a Difference

Code it

Coding

1.Importchi2_contingencyandchi2fromscipy.statspackage.

2.Declare a 2D array with the values mentioned in the contingency table of marital status by education.

3.Calculate and print the values of

- Chi-Square Statistic - Degree of Freedom - P value - Hint: Use chi2_contingency() function 

4.Assume the alpha value to be 0.05

5.Compare the P value with alpha and decide whether or not to reject the null hypothesis.

- If Rejected print "Reject the Null Hypothesis" - Else print "Failed to reject the Null Hypothesis" 

Execute the script with commandpython app.py | tee output.txt

Sample output 2.33 4.5 8.9

Reject the Null Hypothesis