Simpson's Paradox: Deceptive | Math Animation & Manim Python

Q: How can a trend be positive in every individual group yet reverse entirely when the groups are combined?

This happens because of a mathematical trap involving uneven sample sizes and weighted averages. If a subgroup with a lower success rate has a vastly larger population size, its raw volume of data will disproportionately dominate the final combined average, completely overriding the higher percentages found in smaller groups.

Q: What is a lurking or confounding variable in this context?

A lurking variable is a hidden or unmeasured factor (such as group population size or underlying demographic differences) that influences both variables being studied, creating a false correlation or masking the true relationship.

Q: How can analysts avoid falling into the trap of Simpson's Paradox?

Analysts must segment their data and control for confounding variables rather than relying solely on aggregate totals, ensuring they evaluate localized performance before drawing global conclusions.

The Statistical Mirage of Aggregated Data

Most people believe that if a treatment is better for men and better for women, it must be better for the population as a whole. In reality, it might be worse. This happens because of “lurking variables” or weights. Simpson’s Paradox occurs when a trend appearing in several groups of data disappears or reverses when those groups are combined. It’s a sobering reminder that data without context isn’t just incomplete—it can be 100% misleading.

Simpson's Paradox Data Visualization

Simpsons Paradox: Visualizing how hidden variables can reverse statistical trends when individual groups are aggregated into a single dataset.

Visual Interpretation in Manim

The Manim animation visualizes this paradox through Vector Slopes. By representing success rates as the steepness of a line, we can see how two “steep” lines can combine to create a “shallow” result.

The Grouped Vectors: Local TruthsTwo distinct colors (e.g., Blue and Red) show individual group performance. In their own localized space, both show a positive upward trend.
The Aggregate Vector: The Global LieA white dashed line represents the combined data. Notice how it tilts downward even though its components tilt upward.
The HUD: Real-Time RatiosThe scoreboard remains stationary in the corner, showing the success ratios (e.g., 5/10 vs 90/100). This highlights the Weighting Bias—the real culprit behind the paradox.

Why it matters:

In medical trials or Berkeley admission cases, failing to account for group size leads to “Common Cause” fallacies.

The Visual Logic:

Slopes represent rates ($Success/Total$). Adding vectors is not the same as adding slopes.

Note: This phenomenon is a critical study in Probability Theory and Causal Inference. It demonstrates how Confounding Variables can distort a Correlation until it no longer reflects the Statistical Significance of the underlying groups.

The Mathematical Proof

Simpson’s Paradox occurs because of weighted averages. A high success rate in a small group cannot overcome a low success rate in a massive group when they are merged.

Mathematically, it is possible for these three inequalities to exist simultaneously:

(a1 / b1) > (c1 / d1)
(a2 / b2) > (c2 / d2)
BUT
(a1 + a2) / (b1 + b2) < (c1 + c2) / (d1 + d2)

The weights (the denominators b and d) are the “lurking variables.” If Group 2 is much larger than Group 1, its lower performance will “drag” the total average down, regardless of how well Group 1 performed.

The Weighting Trap:

Total success is driven by volume, not just percentage. Massive groups dominate the final average.

Causal Direction:

To avoid the paradox, one must ask: “What is the cause?” and split data accordingly.

Real-Life Use Examples

University Admissions and Gender Bias Studies: The famous UC Berkeley admissions case appeared to show bias against female applicants when looking at aggregate numbers. However, when broken down by individual departments, women actually had equal or higher acceptance rates; the aggregate trend was skewed because women applied in higher numbers to more competitive departments with lower baseline acceptance rates.

Medical Clinical Trials and Treatment Efficacy: When testing a new medical treatment, aggregate results might suggest the therapy is ineffective or harmful. However, segmenting the data by patient age or baseline health can reveal that the treatment works exceptionally well for specific sub-groups, its success masked entirely by raw volume disparities.

Business Analytics and Marketing Conversion Rates: E-commerce companies analyzing marketing campaigns might find that a specific ad performs poorly globally. Yet, when segmented by device type (mobile versus desktop) or regional traffic, the ad actually performs brilliantly within every individual customer segment, distorted only by traffic volume imbalances.

Frequently Asked Questions

How can a trend be positive in every individual group yet reverse entirely when the groups are combined?

This happens because of a mathematical trap involving uneven sample sizes and weighted averages. If a subgroup with a lower success rate has a vastly larger population size, its raw volume of data will disproportionately dominate the final combined average, completely overriding the higher percentages found in smaller groups.

What is a lurking or confounding variable in this context?

A lurking variable is a hidden or unmeasured factor (such as group population size or underlying demographic differences) that influences both variables being studied, creating a false correlation or masking the true relationship.

How can analysts avoid falling into the trap of Simpson’s Paradox?

Analysts must segment their data and control for confounding variables rather than relying solely on aggregate totals, ensuring they evaluate localized performance before drawing global conclusions.

Have questions about Simpson’s Paradox: Deceptive Data?

Ask our AI Tutor for step-by-step calculations, concept breakdowns, or formulas.

Name: Source Code: Manim Implementation *

from manim import *

class SimpsonsParadox(Scene):
    def construct(self):
        # 1. THE GRID

🔒 Members code locked — Please log in to view the full code.