Statistical

CORREL Function in Excel

Returns the correlation coefficient between two data sets.

Syntax

  • =CORREL(array1, array2)

Arguments

  • array1 (required): First range of values
  • array2 (required): Second range of values

Examples

  • =CORREL(A1:A100, B1:B100) - Correlation between A and B - Result: -1 to 1

Excel CORREL function documentation

  • CORREL returns the correlation coefficient between two data sets (-1 to 1).
  • Syntax: =CORREL(array1, array2). Measures linear relationship strength.
  • Use in finance (asset correlation), QA paired metrics, and exploratory analysis.
  • Related: [COVARIANCE.P](/functions/covariance-p/), [RSQ](/functions/rsq/), [PEARSON](/functions/pearson/), [SLOPE](/functions/slope/).

CORREL syntax and requirements

  • array1 and array2 same length — mismatched → #N/A.
  • Text ignored; pairs must align row by row.
  • CORREL ≈ PEARSON in Excel — same linear correlation.
  • Nonlinear relationships may show low CORREL despite pattern.

Step-by-step: correlate ad spend and revenue

  • Step 1 — Monthly spend column B, revenue column C, same rows.
  • Step 2 — =CORREL(B2:B24, C2:C24).
  • Step 3 — Interpret: 0.8 strong positive; -0.5 moderate negative.
  • Step 4 — Scatter chart visual check before trusting one number.
  • Step 5 — Outliers skew CORREL — review or use robust methods.

CORREL vs RSQ vs COVARIANCE

  • CORREL — unitless -1..1.
  • RSQ — R-squared = CORREL^2.
  • COVARIANCE — scale dependent.
  • Statistical guides: [/categories/statistical/](/categories/statistical/).

Worked examples to copy

  • Basic: =CORREL(A2:A100, B2:B100).
  • Named ranges: =CORREL(Spend, Revenue).
  • With IFERROR: =IFERROR(CORREL(x,y), "Need more data").
  • Check RSQ: =RSQ(x,y) equals CORREL^2.

People also ask

  • CORREL causation? — No; correlation not causation.
  • CORREL missing data? — Remove rows with blanks pairwise.
  • CORREL non-linear? — May be near zero — try scatter plot.
  • CORREL Google Sheets? — CORREL(range1, range2).

Common errors

  • Arrays must be same size
  • #DIV/0! if std dev is zero

Use cases

  • Relationship analysis
  • Trend identification
  • Regression prep

Frequently asked questions

  • How do I interpret the correlation coefficient? CORREL returns values from -1 to 1. +1 = perfect positive correlation (both increase together), -1 = perfect negative correlation (one increases as other decreases), 0 = no linear relationship. Generally: |r| > 0.7 is strong, 0.4-0.7 is moderate, < 0.4 is weak.
  • Does correlation mean causation? No! Correlation only shows that two variables move together, not that one causes the other. Ice cream sales and drowning deaths are correlated (both increase in summer), but ice cream doesn't cause drowning. Always consider confounding variables.
  • How do I calculate R-squared from correlation? R² = CORREL(...)^2 or use =RSQ(array1, array2) directly. R² tells you what percentage of variation in Y is explained by X. If CORREL = 0.8, then R² = 0.64, meaning 64% of Y's variation is explained by X.

Editorial review

  • Reviewed by Excel.Directory Editorial Team. Updated May 2026.

When to use CORREL

  • Relationship analysis — common Statistical scenario for CORREL.
  • Trend identification — common Statistical scenario for CORREL.
  • Regression prep — common Statistical scenario for CORREL.

CORREL in the Statistical category

  • Browse all Statistical functions at /categories/statistical/ for related formulas.
  • CORREL syntax: =CORREL(array1, array2)
  • array1 (required): First range of values
  • array2 (required): Second range of values
  • Confirm CORREL arguments match the syntax shown above before filling down.
  • Lock table and range references with $ when copying formulas across rows or sheets.

Formula checklist before you copy down

  • Confirm CORREL arguments match the syntax shown above before filling down.
  • Lock table and range references with $ when copying formulas across rows or sheets.
  • If results look wrong, check for text stored as numbers and invisible spaces with TRIM.
  • Spot-check three known input rows manually against expected output.

Related Excel functions

  • COVARIANCE.P (/functions/covariance-p/): Returns population covariance between two data sets.
  • RSQ (/functions/rsq/): Returns the square of the Pearson product moment correlation coefficient (R²).
  • SLOPE (/functions/slope/): Returns the slope of the linear regression line.

Errors to watch for

  • Arrays must be same size — review causes on linked error pages in the directory.
  • #DIV/0! if std dev is zero — review causes on linked error pages in the directory.

Copy-paste audit workflow

  • Enter CORREL on three test rows with known expected output documented on a QA tab.
  • Fill down only after absolute references are locked on lookup tables and rate tables.
  • Compare against manual calculation or a calculator for financial and statistical functions.
  • Search this directory for comparison guides when choosing between similar functions in the same category.

CORREL worked examples to copy

  • =CORREL(A1:A100, B1:B100) — Correlation between A and B. Expected result: -1 to 1.

CORREL reference summary for crawlers and offline review

  • CORREL belongs to the Statistical category in Excel. Returns the correlation coefficient between two data sets.
  • Full syntax: =CORREL(array1, array2). Open /functions/correl/ for parameters, FAQs, and related pages.
  • Common mistakes: Arrays must be same size; #DIV/0! if std dev is zero
  • Pair this function with comparison guides when another Excel formula might fit the same task better.
  • Review fix-excel-formula-errors when unexpected errors appear after upgrading Excel or sharing across locales.