AB Test Calculator

Calculate statistical significance for two-proportion AB testing with confidence intervals and Z-score analysis

AB Test Statistical Significance

Group 1 (Control)

Conversion Rate
0.00%

Group 2 (Treatment)

Conversion Rate
0.00%

AB Test Results

0.000
Z-Score
1.96
Critical Value (Zα/2)
100.000%
P-Value
0.0000
Overall Proportion (p̄)
0.0000
Effect Size (Cohen's h)
✗ NOT STATISTICALLY SIGNIFICANT
|Z| = 0.000 < 1.96 (Critical Value)

Interpretation: With 95% confidence, we fail to reject the null hypothesis. There is no statistically significant difference between the two groups.

Conversion Rate Difference: 0.00 percentage points (Group 2 higher)

Test Validity Checks

⚠️ Warning: Sample sizes below 30 may not follow normal distribution assumptions.

Example AB Test

Website Button Test

Group 1 (Red Button):

• Sample size: 1,000 visitors

• Clicks: 85 conversions

• Conversion rate: 8.5%

Group 2 (Blue Button)

• Sample size: 1,000 visitors

• Clicks: 112 conversions

• Conversion rate: 11.2%

Result

Z-score: -2.28

P-value: 2.27%

Significant at 95% confidence

Blue button performs better!

AB Test Requirements

1

Sample Size

≥30 samples per group

Required for normal distribution

2

Random Sampling

Representative samples

Avoid selection bias

3

Similar Sizes

Balanced group sizes

For reliable results

Confidence Levels

90%Z = 1.645
95%Z = 1.96
98%Z = 2.326
99%Z = 2.576

Understanding AB Testing

What is an AB Test?

An AB test is a statistical method to compare two versions of something to determine which performs better. It uses a two-proportion Z-test to determine if observed differences between groups are statistically significant or just due to random chance.

When to Use AB Testing

  • Website design and user interface testing
  • Marketing campaign effectiveness
  • Product feature comparison
  • Medical treatment efficacy

Z-Score Formula

Z = (p₁ - p₂) / √[p̄(1-p̄)(1/n₁ + 1/n₂)]

where p̄ = (t₁ + t₂)/(n₁ + n₂)

  • p₁, p₂: Sample proportions for groups 1 and 2
  • p̄: Overall sample proportion
  • n₁, n₂: Sample sizes for groups 1 and 2
  • t₁, t₂: Number of positive results in each group

Statistical Significance

Statistical significance indicates that the observed difference between groups is unlikely to have occurred by chance alone. We reject the null hypothesis when |Z| > Zα/2.

P-Value Interpretation

p < 0.01: Very strong evidence
p < 0.05: Strong evidence
p < 0.10: Some evidence
p ≥ 0.10: Little/no evidence

Effect Size (Cohen's h)

|h| < 0.2: Small effect
|h| = 0.5: Medium effect
|h| = 0.8: Large effect
|h| > 1.2: Very large effect