Understanding Positive Correlation: Key Concepts and Examples

Search Fintrev

Search our guides, reviews and glossary.

A positive correlation is a relationship between two variables in which they tend to move in the same direction: when one rises, the other tends to rise, and when one falls, the other tends to fall. In statistics, the strength of that relationship is usually summarized by a correlation coefficient between 0 and +1.0, where +1.0 is a perfect positive correlation and values near 0 mean little or no linear relationship.

What positive correlation means

Everyday examples make the pattern easy to see:

• Hours worked and pay. For an hourly employee, more hours mean a larger paycheck.
• Fuel prices and airfares. Fuel is a major airline cost, and higher fuel prices are often passed on to travelers through ticket prices.
• Vehicle demand and tire demand. Complementary products rise and fall together: more cars on the road means more tires needed.
• Demand and price. With supply held constant, rising demand pushes prices up.
• Study time and test scores (a common example in psychology and education research): students who study more tend to score higher, though the relationship is far from exact.

How positive correlation works

A perfect positive correlation (r = +1.0) means every data point lies exactly on an upward-sloping straight line: each change in one variable is matched by a proportional change in the other, every time. It rarely appears outside relationships fixed by definition or formula, such as a price in dollars and the same price in cents, or interest earned at a fixed rate and the account balance. Real-world economic and market data almost always contain enough noise to keep the coefficient below 1.0.

Most real relationships are imperfect. A coefficient of +0.7, for example, means the two variables move together fairly consistently, but with exceptions and noise. A coefficient of +0.1 means the tendency to move together is faint.

What a positive correlation looks like on a scatter plot

A scatter plot places one variable on the x-axis and the other on the y-axis, with each point representing one observation (one day, one stock, one person). A positive correlation appears as a cloud of points that trends upward from left to right.

• Strong positive correlation: the points cluster tightly around an upward-sloping line.
• Weak positive correlation: the points still drift upward, but they are widely scattered.
• Perfect positive correlation: every point sits on the line.
• No correlation: the points show no upward or downward trend.

The steepness of the trend is not the same as the strength of the correlation. A shallow line with points packed tightly around it shows a stronger correlation than a steep line with points scattered far from it. Steepness relates to slope (how much y changes per unit of x), while correlation measures how consistently the points follow a line.

How to measure positive correlation

The most common measure is the Pearson correlation coefficient:

r = Σ[(x − x̄)(y − ȳ)] ÷ √[Σ(x − x̄)² × Σ(y − ȳ)²]

where x̄ and ȳ are the averages of each variable. The numerator measures how the two variables deviate from their averages together. When both tend to be above average at the same time and below average at the same time, the products are positive and r is positive. The denominator scales the result so it always falls between -1.0 and +1.0.

Worked example (hypothetical data)

Suppose a small business records weekly advertising spend (x, in thousands of dollars) and weekly orders (y, in hundreds) over five weeks:

1. Averages: x̄ = 3, ȳ = 4.
2. Deviations: x − x̄ = -2, -1, 0, 1, 2; y − ȳ = -2, 0, 1, 0, 1.
3. Products of deviations: 4, 0, 0, 0, 2. Sum = 6.
4. Sum of squared x deviations: 4 + 1 + 0 + 1 + 4 = 10.
5. Sum of squared y deviations: 4 + 0 + 1 + 0 + 1 = 6.
6. Coefficient: r = 6 ÷ √(10 × 6) = 6 ÷ 7.75 ≈ 0.77.

The result, about +0.77, indicates a fairly strong positive correlation in this sample: weeks with more ad spend tended to have more orders. Squaring r gives the coefficient of determination, r² ≈ 0.60, meaning about 60% of the variation in orders is associated with the linear relationship with ad spend. With only five data points, the figure could easily change with more weeks of data.

Strong vs. weak correlation

There is no single official cutoff, and fields use different conventions. A common rough guide for the absolute value of r is:

• 0 to 0.3: weak
• 0.3 to 0.7: moderate
• 0.7 to 1.0: strong

A strong positive correlation means knowing one variable gives a reasonably good idea of the other. A correlation of 0.118 is positive but weak: its r² is about 0.014, so only around 1.4% of the variation in one variable is linearly associated with the other.

Statistical significance and the p-value

The coefficient describes a sample. The p-value helps judge whether the observed correlation could plausibly have appeared by random chance when no real relationship exists. A p-value of 0.05 or lower is the conventional threshold for calling a result statistically significant, though the threshold is a convention, not a law.

Sample size matters a great deal. With thousands of observations, even a weak correlation such as 0.118 can be statistically significant: the relationship is probably real, but small. With only a handful of observations, even a large r can fail to reach significance. Reading r and the p-value together tells you both how strong the relationship appears and how much confidence the data supports.

The role of positive correlation in finance

Investors and analysts measure correlation between individual stocks, between asset classes, and between a holding and the broader market. Companies in the same industry face many of the same risks, so their share prices tend to be more positively correlated than those of unrelated businesses: two retailers would generally move together more closely than an online retailer and an auto repair chain.

In finance, correlation is usually measured on returns (percentage changes), not on price levels. Price levels of many assets trend upward over long periods, which can produce high correlations simply because both series grow over time.

How positive correlation affects diversification

Modern portfolio theory, developed by Harry Markowitz in the 1950s, rests on the observation that combining assets that do not move in lockstep lowers portfolio risk without necessarily lowering expected return. Positive correlation works against that benefit. The more closely holdings move together, the less one can offset losses in another.

The effect shows up directly in the formula for the variance of a two-asset portfolio:

σp² = w₁²σ₁² + w₂²σ₂² + 2w₁w₂ρσ₁σ₂

where w is each asset’s weight, σ is each asset’s volatility (standard deviation of returns), and ρ is the correlation between them.

Hypothetical example. Two assets each have annual volatility of 20%, and the portfolio holds 50% in each.

With a perfect positive correlation, combining the two assets provides no reduction in volatility: the portfolio is as risky as either asset alone. As correlation falls, volatility drops even though neither asset became less risky on its own. This is why concentrated portfolios of highly correlated holdings, such as several stocks from the same sector, can carry more risk than their number of positions suggests.

Beta and its relationship to correlation

Beta measures how much a stock’s returns tend to move relative to a benchmark, often the S&P 500. It is closely related to correlation but is not the same thing:

β = ρ × (σ_stock ÷ σ_market)

Beta combines the correlation with the market and the stock’s volatility relative to the market. Two stocks with the same correlation to the index can have very different betas if one is far more volatile.

• Beta of 1.0: the stock has tended to move with the market by about the same proportion.
• Beta below 1.0: the stock has tended to move less than the market, as utility stocks often do.
• Beta above 1.0: the stock has tended to move more than the market. A beta of 1.2 implies moves about 20% larger than the benchmark’s, on average; technology and small-cap stocks often fall here.
• Negative beta: the stock has tended to move opposite the market, as put options and inverse ETFs are designed to do.

Beta captures systematic (market-wide) risk only. A stock with a beta of 1.0 still carries company-specific risk that beta does not measure. Returns beyond what beta would predict are measured separately as alpha.

Positive vs. negative correlation

A negative correlation, also called an inverse correlation, describes two variables that move in opposite directions: when one rises, the other tends to fall. Its coefficient runs from 0 to -1.0, with -1.0 a perfect negative correlation.

Stocks and bonds are often described as negatively correlated in theory, with bonds tending to do well when stocks struggle. That relationship is not fixed. The correlation between the two asset classes has changed sign across different periods, which is a reminder that historical correlations describe the past, not a guarantee.

Positive is also not the same as strong. “Positive” refers only to direction and “strong” only to magnitude: +0.2 is positive but weak, while -0.9 is strong but negative.

Correlation does not imply causation

Assuming that one variable causes the other because they move together is a common logical error. A positive correlation can arise in several ways:

1. Direct cause: one variable drives the other, as fuel costs feed into airfares.
2. A shared third factor: commodity prices, interest rates or economic growth can lift several markets at once.
3. Common trends over time: the number of U.S. internet users and the price of oil both rose over long stretches of recent decades, producing a positive correlation with no meaningful link. The shared factor is simply time.
4. Coincidence: in a small sample, two unrelated variables can line up by chance.

Even when a causal link is plausible, correlation alone cannot show which way it runs. Employment and inflation are often cited as positively correlated: tight labor markets push employers to raise wages, which can feed into prices, while high unemployment tends to weaken demand and ease price pressure. Establishing the mechanism takes economic analysis, not just a coefficient.

Limitations of positive correlation

• It measures only linear relationships. Pearson’s r can be near zero for variables with a strong curved relationship, such as one that rises and then falls.
• Outliers can distort it. A single extreme observation can create or hide a correlation, especially in small samples.
• It ignores magnitude. A correlation of +1.0 means the variables change proportionally along a straight line, not by the same amount. Beta or a regression slope is needed to describe size.
• It changes over time. Asset correlations can shift between periods and often rise during broad selloffs, when diversification is most wanted.
• It depends on the data window and frequency. Daily, monthly and annual returns for the same two assets can give different coefficients, and a coefficient from a limited range of data may not hold outside it.

FAQs

The Bottom Line

Positive correlation matters most to investors building portfolios and to anyone reading data that shows two things rising together. Highly correlated holdings, such as several stocks from one sector, offer less diversification than their count suggests, and correlations can rise during selloffs. Read the coefficient alongside its p-value and sample size, measure assets on returns, and treat any correlation as evidence of association rather than proof of cause.