Hypothesis Testing is a statistical method to make decisions based on data. It helps determine if a result is statistically significant or due to random chance. In business analytics, it powers A/B testing, product improvements, and decision-making.
Hypothesis testing starts with two competing assumptions about a population:
Example: H₀ = “Average sales are equal for Campaign A and B”; H₁ = “Campaign B generates higher average sales.”
A/B test on two website designs: p-value = 0.03, α = 0.05. Since 0.03 < 0.05, we reject H₀ → new design performs significantly better.
Compares means of two groups.
Example: Average sales before vs. after a campaign.
Tests relationships between categorical variables.
Example: Customer preference across age groups.
Compares means across more than two groups.
Example: Sales performance across multiple regions.
Used for large samples when population variance is known.
Example: Proportion of defect rates vs. expected rates.
Balancing Type I & II errors is crucial — smaller α reduces false positives but may increase false negatives.