---
title: "Strength of Association — AP Stats Definition & Guide"
description: "Strength of association measures how tightly points cluster around a linear pattern, quantified by |r|. Key for AP Stats Topic 2.5 and scatterplot FRQs."
canonical: "https://fiveable.me/ap-stats/key-terms/strength-of-association"
type: "key-term"
subject: "AP Statistics"
unit: "Unit 2"
---

# Strength of Association — AP Stats Definition & Guide

## Definition

In AP Statistics, strength of association is how closely data points cluster around a linear pattern, quantified by the absolute value of the correlation coefficient r. Values of |r| near 1 mean a strong linear association; values near 0 mean a weak one. The sign of r tells you direction, not strength.

## What It Is

Strength of association answers one question about a [scatterplot](/ap-stats/key-terms/scatterplot "fv-autolink"): how tightly do the points hug the [linear](/ap-stats/unit-2/representing-relationship-between-two-quantitative-variables/study-guide/3rWWsKXcnbYlqY64hQ1j "fv-autolink") pattern? If the points fall almost exactly on a line, the association is strong. If they form a loose, fuzzy cloud that only vaguely trends one way, the association is weak.

On the AP exam, strength gets quantified by the correlation coefficient r, which is always between -1 and 1 (EK 2.5.B). Here's the part that trips people up. Strength comes from the *absolute value* of r, not its sign. An r of -0.85 describes a stronger linear relationship than an r of 0.65, because |-0.85| > |0.65|. The negative sign just means the [variables](/ap-stats/unit-1/language-variation-variables/study-guide/nKpeaxi1H3Ht9aFhTHKt "fv-autolink") move in opposite directions. Also, r is unit-free, so converting kilograms to pounds or centimeters to inches changes nothing about the strength. One more CED warning worth memorizing: an r close to 1 or -1 does not prove a linear model is appropriate. A curved pattern can still produce a high r, which is why you check the scatterplot and residual plot, not just the number.

## Why It Matters

Strength of association lives in Topic 2.5 (Correlation) within [Unit 2](/ap-stats/unit-2 "fv-autolink"): Exploring Two-Variable Data. It directly supports learning objective 2.5.A (determine the correlation for a linear relationship) and especially 2.5.B (interpret the correlation). When [AP Stats](/ap-stats "fv-autolink") asks you to 'describe the relationship' in a scatterplot, the rubric expects four things: direction, form, strength, and unusual features. Strength is one of those four non-negotiable pieces. It also sets up Unit 2's biggest conceptual trap, which is that a strong association still does not imply causation. Lurking variables can produce tight clustering with zero causal link, and the exam loves testing whether you know the difference.

## Connections

### [Correlation Coefficient (Unit 2)](/ap-stats/key-terms/correlation-coefficient)

The [correlation coefficient](/ap-stats/key-terms/correlation-coefficient "fv-autolink") r is how strength gets a number. Strength of association is the concept; |r| is the measurement. Think of r as packaging two facts into one value, where the sign gives direction and the distance from 0 gives strength.

### [Direction of Association (Unit 2)](/ap-stats/key-terms/direction-of-association)

[Direction](/ap-stats/key-terms/direction "fv-autolink") and strength are the two separate jobs r does at once. Direction is positive or negative (the sign); strength is weak to strong (the absolute value). The exam tests whether you can keep these jobs separate, like recognizing that r = -0.9 is strong, not 'less than' r = 0.5.

### [Z-scores (Unit 1)](/ap-stats/key-terms/z-scores)

The formula for r is essentially the average of the products of [z-scores](/ap-stats/key-terms/z-scores "fv-autolink") for x and y. That's why r is unit-free. Standardizing strips away kilograms, inches, or dollars, so strength measures pure clustering, not measurement scale.

### [Mean (Unit 1)](/ap-stats/key-terms/mean)

The r formula is built on each point's distance from x̄ and ȳ. Points that are consistently above both means or below both means push r toward +1, which is the mechanical reason tight clustering produces strong correlation.

## On the AP Exam

Multiple-choice questions usually test strength one of two ways. First, the comparison trap, where two studies report r = -0.85 and r = 0.65 and the correct answer is that the negative one is stronger because strength depends on |r|. Second, the scatterplot match, where you see points loosely clustered around a downward-sloping line and have to pick a moderately negative r (something like -0.5, not -0.95). Unit conversions also show up; changing kilograms to pounds leaves r exactly the same. On free-response, strength is part of the standard scatterplot description. The 2022 FRQ on bullfrog length and mass and the 2023 FRQ on tule elk both asked about bivariate relationships where describing direction, form, strength, and unusual features earned the points. A safe sentence template is: 'There is a strong, positive, linear association between length and mass.'

## strength of association vs Direction of association

Strength and direction are two different properties of the same relationship, and r encodes both. Direction is the sign of r (positive means the variables rise together, negative means one rises as the other falls). Strength is the absolute value of r (how close to 1 or -1 it is). The classic MCQ trap is treating r = -0.85 as 'weaker' than r = 0.65 because -0.85 is a smaller number. It's the opposite. The relationship with r = -0.85 is stronger; it just runs downhill instead of uphill.

## Key Takeaways

- Strength of association describes how tightly points cluster around a linear pattern, and it's measured by the absolute value of r, not the sign.
- An r of -0.85 indicates a stronger linear association than an r of 0.65, because |-0.85| is closer to 1.
- Correlation r is unit-free, so converting kilograms to pounds or centimeters to inches does not change the strength of association at all.
- An r close to 1 or -1 does not prove a linear model is appropriate; you still need to check the scatterplot or residual plot for curvature.
- A strong association does not imply causation, because lurking variables can create tight clustering between variables that don't cause each other.
- When describing a scatterplot on an FRQ, always mention direction, form, strength, and unusual features to earn full credit.

## FAQs

### What is strength of association in AP Stats?

It's how closely data points cluster around a linear pattern in a scatterplot, measured by the absolute value of the correlation coefficient r. Values of |r| near 1 are strong, near 0 are weak, and r = 0 means no linear association at all.

### Is r = -0.85 stronger than r = 0.65?

Yes. Strength depends only on the absolute value of r, and |-0.85| = 0.85 is closer to 1 than 0.65. The negative sign only tells you the direction of the relationship, not how strong it is.

### Does a strong correlation mean one variable causes the other?

No. The CED states directly that correlation does not necessarily imply causation. A strong association can come from a lurking variable, so only a well-designed experiment can support a causal claim.

### What's the difference between strength and direction of association?

Direction is the sign of r: positive means both variables increase together, negative means one decreases as the other increases. Strength is the magnitude of r, meaning how close it sits to 1 or -1. An r of -0.9 is strong and negative; an r of 0.2 is weak and positive.

### Does changing the units of my variables change the strength of association?

No. Because r is built from z-scores, it's completely unit-free. If r = 0.75 for height in centimeters versus weight in kilograms, it stays exactly 0.75 after converting to inches and pounds.

## Related Study Guides

- [2.5 Correlation](/ap-stats/unit-2/correlation/study-guide/LlS81pC6QricXgIKNuFM)

## Structured Data

```json
{"@context":"https://schema.org","@graph":[{"@type":"LearningResource","@id":"https://fiveable.me/ap-stats/key-terms/strength-of-association#resource","name":"Strength of Association — AP Stats Definition & Guide","url":"https://fiveable.me/ap-stats/key-terms/strength-of-association","learningResourceType":"Concept explainer","educationalLevel":"AP® / High School","about":{"@id":"https://fiveable.me/ap-stats/key-terms/strength-of-association#term"},"audience":{"@type":"EducationalAudience","educationalRole":"student"},"dateModified":"2026-06-11T05:22:54.153Z","isPartOf":{"@type":"Collection","name":"AP Statistics Key Terms","url":"https://fiveable.me/ap-stats/key-terms"},"publisher":{"@type":"Organization","name":"Fiveable","url":"https://fiveable.me"}},{"@type":"DefinedTerm","@id":"https://fiveable.me/ap-stats/key-terms/strength-of-association#term","name":"strength of association","description":"In AP Statistics, strength of association is how closely data points cluster around a linear pattern, quantified by the absolute value of the correlation coefficient r. Values of |r| near 1 mean a strong linear association; values near 0 mean a weak one. The sign of r tells you direction, not strength.","url":"https://fiveable.me/ap-stats/key-terms/strength-of-association","inDefinedTermSet":{"@type":"DefinedTermSet","name":"AP Statistics Key Terms","url":"https://fiveable.me/ap-stats/key-terms"}},{"@type":"FAQPage","mainEntity":[{"@type":"Question","name":"What is strength of association in AP Stats?","acceptedAnswer":{"@type":"Answer","text":"It's how closely data points cluster around a linear pattern in a scatterplot, measured by the absolute value of the correlation coefficient r. Values of |r| near 1 are strong, near 0 are weak, and r = 0 means no linear association at all."}},{"@type":"Question","name":"Is r = -0.85 stronger than r = 0.65?","acceptedAnswer":{"@type":"Answer","text":"Yes. Strength depends only on the absolute value of r, and |-0.85| = 0.85 is closer to 1 than 0.65. The negative sign only tells you the direction of the relationship, not how strong it is."}},{"@type":"Question","name":"Does a strong correlation mean one variable causes the other?","acceptedAnswer":{"@type":"Answer","text":"No. The CED states directly that correlation does not necessarily imply causation. A strong association can come from a lurking variable, so only a well-designed experiment can support a causal claim."}},{"@type":"Question","name":"What's the difference between strength and direction of association?","acceptedAnswer":{"@type":"Answer","text":"Direction is the sign of r: positive means both variables increase together, negative means one decreases as the other increases. Strength is the magnitude of r, meaning how close it sits to 1 or -1. An r of -0.9 is strong and negative; an r of 0.2 is weak and positive."}},{"@type":"Question","name":"Does changing the units of my variables change the strength of association?","acceptedAnswer":{"@type":"Answer","text":"No. Because r is built from z-scores, it's completely unit-free. If r = 0.75 for height in centimeters versus weight in kilograms, it stays exactly 0.75 after converting to inches and pounds."}}]},{"@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"AP Statistics","item":"https://fiveable.me/ap-stats"},{"@type":"ListItem","position":2,"name":"Key Terms","item":"https://fiveable.me/ap-stats/key-terms"},{"@type":"ListItem","position":3,"name":"Unit 2","item":"https://fiveable.me/ap-stats/unit-2"},{"@type":"ListItem","position":4,"name":"strength of association"}]}]}
```