Cross Tabulation - Pearson’s Chi-Square Test

Chi-Square Test is used to see if there is an association between categorical variables or not. We formulate two hypothesis, as follows:

• Null Hypothesis: There is no significant relationship between variables.
• Alternative Hypothesis: There is a significant relationship between variables.

Click to download video

Chi-Square calculation is done using the below-mentioned formula:

where:

• O - Observed Frequency
• E - Expected Frequency
• Significance Levels - α = 0.01, α = 0.05, α = 0.10

Chi-Square Critical value is obtained from Chi-table using degrees of freedom and Significance level α.

If Chi-Square (Calculated) > the critical value of Chi-Square at ‘α’ level of significance, where α (alpha) is either 0.01 or 0.05 or 0.1; or if p-value < level of significance ‘α’, the null hypothesis is rejected.

Furthermore, we state that the association between the variables in the cross table (dimension P x Q) is significant, at 100*(1- α ) % confidence.

Hence we accept the Null hypothesis and state that there is no significant relationship between two variables (Gender and credit cards owned).