P-Value Calculator for Z-Test

Use this calculator to find the p-value from a z-test statistic. This is useful after you’ve calculated your z-statistic and need to determine if results are statistically significant.

When to Use

  • After calculating z-test statistics (one proportion, two proportions, one mean with known σ)
  • Testing claims about population proportions or means (when σ is known)
  • Large sample sizes (z-distribution applies with n ≥ 30 or normal population)
  • Determining statistical significance by comparing p-value to significance level
  • Examples: Testing if conversion rate differs, checking product quality claims

How to Use

Step 1: Enter your calculated z-test statistic (can be positive or negative)

Step 2: Select tail type:

  • Left-tailed: Testing if parameter is less than hypothesized value
  • Right-tailed: Testing if parameter is greater than hypothesized value
  • Two-tailed: Testing if parameter differs from hypothesized value

Step 3: Click “Calculate”

Step 4: Interpret the result:

  • If p-value < 0.05 (typical threshold): Reject H₀, results are statistically significant
  • If p-value ≥ 0.05: Fail to reject H₀, insufficient evidence of difference
Z-test p Value Calculator
Z-value : ($z$)
Tail : Left tailedRight tailedTwo tailed
Results
p-value:

Understanding Z-Test P-Values

The p-value for a z-test represents the probability of observing a test statistic as extreme (or more extreme) as yours, assuming the null hypothesis is true. It uses the standard normal distribution (Z-distribution).

Key Formula

p-value calculation using standard normal distribution:

  • Left-tailed: p-value = P(Z ≤ z_obs) = Φ(z_obs)
  • Right-tailed: p-value = P(Z ≥ z_obs) = 1 - Φ(z_obs)
  • Two-tailed: p-value = 2 × P(|Z| ≥ |z_obs|)

Where Φ(z) is the cumulative standard normal distribution function.

P-Value Interpretation

p-value Range Decision Strength of Evidence
p < 0.001 Strong rejection of H₀ Extremely significant
0.001 ≤ p < 0.01 Reject H₀ Very significant
0.01 ≤ p < 0.05 Reject H₀ Significant
0.05 ≤ p < 0.10 Borderline Marginally significant
p ≥ 0.10 Fail to reject H₀ Not significant

Common Z-Test P-Value Misconceptions

❌ WRONG: “p-value = 0.02 means 2% chance our result is a false positive” ✓ RIGHT: “p-value = 0.02 means if H₀ is true, we’d observe this result 2% of the time”

❌ WRONG: “Significant result (p < 0.05) proves the hypothesis is true” ✓ RIGHT: “Significant result means we have sufficient evidence to reject H₀”

❌ WRONG: “p-value measures the size/importance of the effect” ✓ RIGHT: “p-value measures statistical significance; effect size is separate”

Worked Example

Scenario: Testing if website conversion rate differs from industry standard (12%)

Given:

  • Sample size: n = 500
  • Number of conversions: 72
  • Industry standard (p₀): 0.12
  • H₁: p ≠ 0.12 (two-tailed)

Calculation:

  • Sample proportion: p̂ = 72/500 = 0.144
  • SE = √(0.12 × 0.88 / 500) = 0.0145
  • z = (0.144 - 0.12) / 0.0145 = 1.655
  • Two-tailed p-value: p ≈ 0.098

Decision: Since p-value (0.098) > α (0.05), fail to reject H₀. Marginal evidence that conversion rate differs from 12%.

When to Use Each Tail Type

Left-tailed (p = P(Z ≤ z_obs)):

  • Testing if parameter is less than claim
  • Example: “Is defect rate less than 5%?”
  • H₁: p < p₀ or μ < μ₀

Right-tailed (p = P(Z ≥ z_obs)):

  • Testing if parameter is greater than claim
  • Example: “Is success rate above 80%?”
  • H₁: p > p₀ or μ > μ₀

Two-tailed (p = 2 × P(|Z| ≥ |z_obs|)):

  • Testing if parameter differs from claim
  • Example: “Does treatment change outcome?”
  • H₁: p ≠ p₀ or μ ≠ μ₀

Z-Distribution Properties

The standard normal distribution (Z-distribution) used here:

  • Mean = 0, SD = 1 (standardized)
  • Symmetric: P(Z < -z) = P(Z > z)
  • Assumes large samples or normal population
  • Critical values:
    • α = 0.05 (two-tailed) → |z| = 1.96
    • α = 0.01 (two-tailed) → |z| = 2.576

Tips for Using This Calculator

  1. Calculate z-statistic first using appropriate formula:

    • For proportions: z = (p̂ - p₀) / √(p₀(1-p₀)/n)
    • For means (σ known): z = (x̄ - μ₀) / (σ/√n)

  2. Choose correct tail type based on your alternative hypothesis direction

  3. Large sample requirement: Use z-test when:

    • For proportions: np₀ ≥ 5 AND n(1-p₀) ≥ 5
    • For means: n ≥ 30 (or population is normal)

  4. Remember: p-value is conditional on H₀ being true

  5. Consider effect size: Combine p-value with practical significance

Related: Z-Test for One Mean, Z-Test for Proportion, Z-Test for Two Proportions, Tutorial