Math I · S-ID.8

Computing and Interpreting the Correlation Coefficient

This objective gives students a numerical way to describe how strongly two quantitative variables follow a linear pattern. It helps them move beyond “the graph looks kind of related” into more precise data language.

Concept Statistics and Probability
Domain Interpreting Categorical and Quantitative Data
Read time 11 minutes

What this learning objective is really asking you to learn

This learning objective asks students to attach a number to the strength and direction of a linear relationship. A scatter plot gives a visual impression. A fitted line gives a model. The correlation coefficient, usually written as \(r\), gives a standardized numerical summary of how closely the points follow a linear pattern.

The correlation coefficient ranges from -1 to 1. A value near 1 indicates a strong positive linear association: as the input variable increases, the output variable tends to increase, and the points lie close to an upward-sloping line. A value near -1 indicates a strong negative linear association: as the input increases, the output tends to decrease, and the points lie close to a downward-sloping line. A value near 0 indicates little or no linear association. That last phrase is important. A correlation near zero does not always mean there is no relationship at all. The data might have a curved relationship, a cluster pattern, or a relationship hidden by subgroups. Correlation measures linear association.

Students are not expected at this level to calculate \(r\) by hand from the full formula. The standard explicitly points students toward technology. That is sensible because the computation is tedious and error-prone. But students must understand what the technology is producing. A calculator, spreadsheet, or statistical program can compute \(r\), but the student must interpret it.

A correlation coefficient has two main pieces of information: sign and magnitude. The sign tells direction. Positive \(r\) means the variables tend to increase together. Negative \(r\) means one variable tends to decrease as the other increases. The magnitude, or absolute value, tells strength. Values closer to 1 or -1 indicate points closer to a line. Values closer to 0 indicate a weaker linear pattern.

The word standardized matters. Correlation has no units. It does not matter whether height is measured in inches or centimeters; the correlation between height and arm span will be the same. This is different from slope, which does depend on units. If you convert inches to centimeters, the slope changes because the units change. Correlation focuses on how consistently the variables move together in a linear way, not on the exact rate of change in original units.

Correlation is closely related to standard deviation and standardized variables. Conceptually, \(r\) compares how far each \(x\) value is from its mean with how far each corresponding \(y\) value is from its mean. If points that are above average in \(x\) also tend to be above average in \(y\), the correlation is positive. If points above average in \(x\) tend to be below average in \(y\), the correlation is negative. If there is no consistent pairing, the correlation is near zero.

This objective also asks students to use technology responsibly. Entering data correctly, choosing the right command, reading the output, and interpreting the number in context are all part of the task. Technology makes computation faster, but it does not decide whether a linear model is appropriate. Students still need to inspect the scatter plot. A single number can hide outliers, clusters, curvature, and data mistakes.

Why students should learn this math

Students should learn correlation because modern life is full of claims about relationships between variables. People ask whether more sleep is associated with better grades, whether more advertising is associated with more sales, whether more training is associated with fewer injuries, whether income is associated with education, whether pollution is associated with illness, whether screen time is associated with anxiety, or whether practice time is associated with performance. Correlation gives one tool for describing such relationships.

Without correlation, people often rely on vague visual language: “it looks related,” “there seems to be a trend,” or “the graph goes up.” Those statements may be useful as first impressions, but they are imprecise. A correlation coefficient helps quantify the strength of the linear pattern. A correlation of 0.92 tells a different story from a correlation of 0.28, even if both are positive. A correlation of -0.75 tells a different story from -0.10.

Correlation is especially useful when comparing relationships. Suppose a coach studies the relationship between practice minutes and performance improvement for several skills. One skill may show a strong positive correlation, while another shows a weak correlation. That does not automatically prove practice causes improvement, but it helps identify where the linear association is stronger. A business might compare the correlation between customer wait time and satisfaction across different locations. A scientist might compare the correlation between environmental exposure and health outcome across several variables.

Students also need correlation because it is often misused. A social media post may show a correlation and imply proof. A news article may report a correlation without explaining strength or limitations. An advertisement may claim that customers who use a product have better outcomes, even if those customers differ in other ways. Understanding correlation helps students ask, “How strong is the relationship? Is it linear? Are there outliers? What variables were measured? Does this prove cause?”

This objective builds data literacy in a world where technology can generate statistics instantly. Anyone can compute a correlation with a spreadsheet. The scarce skill is interpretation. Students need to know that \(r = 0.8\) is not “80 percent true,” that \(r = 0\) does not rule out all patterns, and that \(r\) does not prove causation. They also need to know that correlation is meaningful only when the data and context support the question being asked.

The historical machinery behind correlation

The modern correlation coefficient is closely associated with Francis Galton and Karl Pearson in the late nineteenth century. Galton studied relationships among biological traits and became interested in how measurements vary together. Pearson developed the mathematical formalization of the product-moment correlation coefficient that is still widely used today. The history is scientifically important but also ethically complicated, because some early statistical work was entangled with flawed and harmful ideas about heredity and society. Students do not need a full history of statistics to learn \(r\), but they should know that mathematical tools can be used well or badly depending on the questions, assumptions, and values behind them.

The technical need behind correlation was clear: researchers wanted to quantify association. Scatter plots could show a pattern, but scientists needed a number that described how tightly two variables moved together. Covariance was one step in that direction, but covariance depends on units. If height is measured in centimeters instead of inches, the covariance changes. Correlation solved this by standardizing. It scales the relationship so that the result always falls between -1 and 1.

This standardization made correlation portable. A correlation between height and arm span can be compared with a correlation between study time and score, even though the units are completely different. That portability is one reason correlation became so influential in statistics, psychology, biology, economics, education, and social science.

Correlation also became a stepping-stone to regression. Regression lines describe prediction. Correlation describes strength and direction of linear association. The two are related but not identical. A data set can have a steep slope and a moderate correlation, or a shallow slope and a strong correlation, depending on the scales and scatter. This distinction is one reason students must interpret both slope and correlation carefully.

Today, correlation is everywhere in data analysis. It appears in exploratory data analysis, feature selection in machine learning, finance, medicine, education research, climate science, sports analytics, and quality control. But its wide use comes with a warning: a simple statistic can create false confidence if separated from context and design. That warning becomes the core of Objective 059.

Technical execution: computing \(r\) with technology

A typical process begins with paired data. The data must consist of matched pairs: each \(x\) value belongs with a specific \(y\) value. For example, each student has both hours studied and test score; each car has both age and price; each day has both temperature and energy use. If the pairings are broken, the correlation is meaningless.

Next, create a scatter plot. This step should not be skipped. The plot reveals whether the relationship is roughly linear, whether there are outliers, whether clusters exist, and whether a single correlation coefficient is appropriate. A correlation coefficient should not be interpreted in isolation from the graph.

Then use technology. In a graphing calculator, students may enter the \(x\) values into one list and the \(y\) values into another list, run a linear regression command, and read the displayed value of \(r\). In a spreadsheet, students may use a correlation function on the two data columns. In statistical software, they may request a correlation matrix or regression output. The exact button sequence depends on the tool, but the conceptual sequence is the same: enter paired data, compute correlation, interpret.

After computing \(r\), interpret the sign and magnitude. If \(r = 0.87\), the association is positive and strong. If \(r = -0.64\), the association is negative and moderately strong. If \(r = 0.12\), the data show a weak positive linear association or almost no linear association. These adjectives are not absolute laws. Context matters. In some social-science settings, a correlation that looks modest may still be practically meaningful. In a tightly controlled physics lab, the same value might be considered weak. At Math I level, students should use reasonable language rather than pretend there are universal cutoffs.

Students should avoid interpreting \(r\) as a slope. If \(r = 0.8\), that does not mean \(y\) increases by 0.8 for each increase of 1 in \(x\). Slope handles rate of change in units. Correlation handles strength and direction without units. Students should also avoid interpreting \(r\) as a percentage unless they are specifically discussing \(r^2\), and even then the interpretation must be careful and tied to variation explained by a linear model.

A concrete example

Suppose students collect data relating hours of sleep before a test to test score. A scatter plot shows a positive pattern: students who slept more tended to score higher, although there is plenty of variation. A spreadsheet gives \(r = 0.62\).

A good interpretation is: the data show a moderate positive linear association between sleep hours and test score. Students with more sleep tended to have higher scores, but the relationship is not perfect. Other factors likely affect score, such as preparation, prior knowledge, stress, and test difficulty.

A weak interpretation would be: “The correlation is 0.62, so sleep causes 62 percent of the score.” That is wrong. Correlation does not prove cause, and 0.62 is not a percent of causation. Another weak interpretation would be: “The slope is 0.62.” Unless the regression slope also equals 0.62 in score-points per hour, which would be a separate calculation, that statement confuses two different statistics.

A stronger analysis would also inspect the scatter plot for outliers. If one student slept very little and scored extremely high, or slept a lot and scored extremely low, that point might influence the correlation. The student might ask whether the data set is large enough, whether all students came from the same class, and whether the conclusion should be generalized.

What correlation can and cannot tell

Correlation can tell the direction of a linear association. It can tell whether the points are tightly or loosely arranged around a line. It can help compare the strength of different linear relationships. It can support prediction when a linear model is appropriate.

Correlation cannot prove causation. It cannot detect all non-linear relationships. It cannot protect against bad data. It cannot explain why variables are associated. It cannot decide whether a relationship is important in context. It cannot replace a scatter plot. It cannot make an unfair sample fair.

The classic danger is a curved relationship. Imagine data shaped like a U. Low and high values of \(x\) both correspond to high values of \(y\), while middle values of \(x\) correspond to low values of \(y\). The correlation might be near zero because the upward and downward parts cancel in a linear summary. But there is clearly a relationship. It is just not linear.

Another danger is outliers. A single extreme point can create a strong correlation where the main cluster has little relationship, or weaken a correlation where most points follow a clear pattern. That is why graphing comes first.

A third danger is mixing groups. Suppose a data set includes two different populations. Each group may have its own pattern, but the combined data may show a different correlation. This is one reason context and data collection matter. Students should not blindly trust one statistic without understanding what the data represent.

Where this objective fits on the full map of mathematics

On the full map, Objective 058 turns visual association into numerical evidence. Earlier objectives taught students to plot data, fit functions, analyze residuals, and interpret linear parameters. Correlation adds a standardized measure of how linear the pattern is. It is one of the first statistics students learn that is not simply about one variable but about the relationship between two variables.

This objective also previews later ideas in probability and inference. Correlation is a sample statistic. In more advanced courses, students may ask whether an observed correlation is statistically significant, whether it could arise by chance, how sample size affects confidence, and how to build models with multiple predictors. They may also learn about covariance matrices, regression coefficients, and causal inference. The Math I version is the doorway.

It also connects to geometry. Correlation is related to alignment in a coordinate plane. Points close to an upward line produce a positive value near 1; points close to a downward line produce a negative value near -1. It connects to algebra because the fitted line has slope and intercept. It connects to functions because the line is a rule used for prediction. It connects to statistics because variation around the line matters.

Common misconceptions and how to fix them

One misconception is that \(r = 0.7\) means “70 percent.” Correlation is not a percent. Another misconception is that a high correlation proves cause. It does not. A third misconception is that correlation measures any relationship. It specifically measures linear association. A strong curved relationship can have low correlation.

A fourth misconception is that correlation and slope are the same. Slope has units and describes predicted change in \(y\) for a one-unit change in \(x\). Correlation has no units and describes strength and direction of linear association. A fifth misconception is that technology output is automatically meaningful. Technology can compute correlation for data that should not be summarized by correlation.

Students can fix these misconceptions by always using a three-part routine: look at the scatter plot, compute \(r\), and interpret in context. The graph guards against blind calculation. The number adds precision. The context gives meaning.

Mastery looks like this

A student has mastered this objective when they can use technology to compute \(r\), describe the association as positive or negative, strong or weak, and linear or not clearly linear. They can explain why correlation has no units. They can distinguish correlation from slope. They can identify the danger of outliers and curved patterns. They can state clearly that correlation alone does not prove causation.

This objective gives students a powerful data-literacy tool. It helps them move from “I see a pattern” to “I can describe the pattern carefully.” In a data-saturated world, that carefulness is not optional. It is part of being mathematically awake.

Problem Library

Problems in the App From This Objective

144 problems across 12 archetypes in the app.

connect positive/negative r to association direction.
12 problems Warmup Practice Mixed Review Assessment
Problem 1

Interpret the sign of correlation coefficient r=0.78.

Problem 2

Interpret the sign of correlation coefficient r=-0.62.

Problem 3

Interpret the sign of correlation coefficient r=0.03.

Problem 4

Interpret the sign of correlation coefficient r=-0.10.

Open in simulator
Problem 5

Interpret the sign of correlation coefficient r=0.98.

Problem 6

Interpret the sign of correlation coefficient r=0.55.

Problem 7

Interpret the sign of correlation coefficient r=0.15.

Problem 8

Interpret the sign of correlation coefficient r=-0.95.

Problem 9

Interpret the sign of correlation coefficient r=-0.50.

Problem 10

Interpret the sign of correlation coefficient r=-0.20.

Problem 11

Interpret the sign of correlation coefficient r=0.00.

Problem 12

Interpret the sign of correlation coefficient r=-0.05.

connect closeness to 1 or -1 with strength.
12 problems Warmup Practice Mixed Review Assessment
Problem 13

Interpret the magnitude of correlation coefficient r=0.92.

Problem 14

Interpret the magnitude of correlation coefficient r=-0.88.

Problem 15

Interpret the magnitude of correlation coefficient r=0.35.

Open in simulator
Problem 16

Interpret the magnitude of correlation coefficient r=0.02.

Problem 17

Interpret the magnitude of correlation coefficient r=0.98.

Problem 18

Interpret the magnitude of correlation coefficient r=-0.95.

Problem 19

Interpret the magnitude of correlation coefficient r=0.65.

Problem 20

Interpret the magnitude of correlation coefficient r=-0.55.

Problem 21

Interpret the magnitude of correlation coefficient r=0.28.

Problem 22

Interpret the magnitude of correlation coefficient r=-0.15.

Problem 23

Interpret the magnitude of correlation coefficient r=1.00.

Problem 24

Interpret the magnitude of correlation coefficient r=-1.00.

estimate direction and strength visually.
12 problems Warmup Practice Mixed Review Assessment
Problem 25

Match scatter plot description tight upward linear pattern to likely correlation coefficient.

Problem 26

Match scatter plot description tight downward linear pattern to likely correlation coefficient.

Problem 27

Match scatter plot description widely scattered with slight upward trend to likely correlation coefficient.

Problem 28

Match scatter plot description no visible linear pattern to likely correlation coefficient.

Problem 29

Match scatter plot description moderate downward linear trend to likely correlation coefficient.

Open in simulator
Problem 30

Match scatter plot description strong upward linear pattern to likely correlation coefficient.

Problem 31

Match scatter plot description strong downward linear pattern to likely correlation coefficient.

Problem 32

Match scatter plot description moderate upward linear trend to likely correlation coefficient.

Problem 33

Match scatter plot description widely scattered with slight downward trend to likely correlation coefficient.

Problem 34

Match scatter plot description very weak upward linear pattern to likely correlation coefficient.

Problem 35

Match scatter plot description very weak downward linear pattern to likely correlation coefficient.

Problem 36

Match scatter plot description nearly perfect upward linear pattern to likely correlation coefficient.

compare absolute values of r.
12 problems Warmup Practice Mixed Review Assessment
Problem 37

Choose the strongest linear association from r-values 0.4, -0.8, 0.6.

Problem 38

Choose the strongest linear association from r-values 0.91, -0.75, 0.12.

Problem 39

Choose the strongest linear association from r-values -0.2, 0.3, -0.35.

Problem 40

Choose the strongest linear association from r-values 0.05, -0.04, 0.01.

Open in simulator
Problem 41

Choose the strongest linear association from r-values 0.1, 0.5, 0.2.

Problem 42

Choose the strongest linear association from r-values -0.3, -0.9, -0.6.

Problem 43

Choose the strongest linear association from r-values 0.7, -0.4, 0.1.

Problem 44

Choose the strongest linear association from r-values 0.25, -0.85, 0.5.

Problem 45

Choose the strongest linear association from r-values -0.15, 0.08, -0.1.

Problem 46

Choose the strongest linear association from r-values 0.6, -0.7, 0.3, -0.1.

Problem 47

Choose the strongest linear association from r-values 0.99, -0.95, 0.8.

Problem 48

Choose the strongest linear association from r-values -0.55, 0.62, -0.38.

read r from regression/statistics output.
12 problems Warmup Practice Mixed Review Assessment
Problem 49

Read correlation coefficient from technology output LinReg y=ax+b: a=2.1, b=4.5, r=0.82.

Problem 50

Read correlation coefficient from technology output Regression output: slope=-1.4, intercept=20, r=-0.76, r^2=0.58.

Problem 51

Read correlation coefficient from technology output STAT CALC shows r = 0.15 and r^2 = 0.0225.

Problem 52

Read correlation coefficient from technology output Linear Regression Results: r = -0.934, p-value = 0.001.

Problem 53

Read correlation coefficient from technology output r = 0.55, slope = 1.2, intercept = 3.4.

Problem 54

Read correlation coefficient from technology output Summary Statistics: Mean=10, StDev=2, N=50, r=0.08.

Problem 55

Read correlation coefficient from technology output TI-84 Output: a=5, b=10, r=1.

Problem 56

Read correlation coefficient from technology output Software results: r = -1, r-squared = 1.0.

Problem 57

Read correlation coefficient from technology output Correlation coefficient r: 0.012.

Problem 58

Read correlation coefficient from technology output The correlation coefficient is r = 0.7891.

Open in simulator
Problem 59

Read correlation coefficient from technology output Model fit: R-sq = 0.64, R-sq(adj) = 0.62, r = -0.80.

Problem 60

Read correlation coefficient from technology output Calc output: r-value = 0.456, p = 0.02.

describe linear association only.
12 problems Warmup Practice Mixed Review Assessment
Problem 61

Interpret r=0.84 in context hours practiced and performance score without implying causation.

Problem 62

Interpret r=-0.70 in context age of a car and resale value without implying causation.

Problem 63

Interpret r=0.12 in context shoe size and quiz score without implying causation.

Problem 64

Interpret r=0.90 in context daily temperature and ice cream sales without implying causation.

Open in simulator
Problem 65

Interpret r=0.65 in context study time and exam grades without implying causation.

Problem 66

Interpret r=0.30 in context height and income without implying causation.

Problem 67

Interpret r=-0.92 in context number of hours of sleep and caffeine consumption without implying causation.

Problem 68

Interpret r=-0.55 in context exercise frequency and body fat percentage without implying causation.

Problem 69

Interpret r=-0.25 in context distance from school and GPA without implying causation.

Problem 70

Interpret r=0.05 in context favorite color and IQ score without implying causation.

Problem 71

Interpret r=0.88 in context amount of fertilizer and crop yield without implying causation.

Problem 72

Interpret r=-0.68 in context number of hours watching TV and test scores without implying causation.

recognize that r measures linear association.
12 problems Warmup Practice Mixed Review Assessment
Problem 73

Determine whether correlation is appropriate for pattern points form a clear straight-line upward pattern.

Problem 74

Determine whether correlation is appropriate for pattern points form a strong U-shaped pattern.

Problem 75

Determine whether correlation is appropriate for pattern points follow a curved exponential pattern.

Problem 76

Determine whether correlation is appropriate for pattern points show no pattern.

Problem 77

Determine whether correlation is appropriate for pattern points form a tight downward sloping line.

Problem 78

Determine whether correlation is appropriate for pattern points form an inverted U-shape.

Problem 79

Determine whether correlation is appropriate for pattern data points generally follow a straight line with some scatter.

Problem 80

Determine whether correlation is appropriate for pattern points follow a C-shaped curve.

Problem 81

Determine whether correlation is appropriate for pattern a clear linear trend decreasing from left to right.

Problem 82

Determine whether correlation is appropriate for pattern a scatter plot showing a logarithmic curve.

Open in simulator
Problem 83

Determine whether correlation is appropriate for pattern data points forming a parabolic arc.

Problem 84

Determine whether correlation is appropriate for pattern points are clustered moderately around an upward sloping line.

reason about leverage and association strength.
12 problems Warmup Practice Mixed Review Assessment
Problem 85

Explain how outlier one high-leverage point far to the right lies on the trend line can affect correlation.

Problem 86

Explain how outlier one point far from an otherwise strong positive line can affect correlation.

Problem 87

Explain how outlier removing one extreme point changes r from 0.85 to 0.30 can affect correlation.

Problem 88

Explain how outlier a single point in the upper left of a data set that otherwise shows a strong positive trend can affect correlation.

Problem 89

Explain how outlier a data point far from the mean of X and Y that falls perfectly on the regression line can affect correlation.

Problem 90

Explain how outlier a point far off the line of best fit in a generally linear data set can affect correlation.

Problem 91

Explain how outlier one extreme point in the top-right corner of a scattered plot with no clear trend can affect correlation.

Problem 92

Explain how outlier a point that strongly deviates from an otherwise clear negative linear trend can affect correlation.

Problem 93

Explain how outlier an observation with an extreme x-value that lies close to the regression line can affect correlation.

Problem 94

Explain how outlier a point with an unusual y-value but a typical x-value, far from the trend can affect correlation.

Problem 95

Explain how outlier a point at the far end of a curvilinear pattern that aligns with a straight line can affect correlation.

Open in simulator
Problem 96

Explain how outlier a single data point that lies in the opposite quadrant of the majority of a strong positive cluster can affect correlation.

compare direction and strength.
12 problems Warmup Practice Mixed Review Assessment
Problem 97

Compare correlation coefficients r1=0.82 and r2=0.45.

Problem 98

Compare correlation coefficients r1=-0.75 and r2=0.70.

Problem 99

Compare correlation coefficients r1=-0.20 and r2=-0.90.

Problem 100

Compare correlation coefficients r1=0.05 and r2=-0.04.

Problem 101

Compare correlation coefficients r1=0.30 and r2=0.90.

Problem 102

Compare correlation coefficients r1=-0.85 and r2=-0.40.

Problem 103

Compare correlation coefficients r1=0.10 and r2=-0.80.

Problem 104

Compare correlation coefficients r1=-0.95 and r2=0.08.

Problem 105

Compare correlation coefficients r1=0.98 and r2=-0.97.

Problem 106

Compare correlation coefficients r1=0.15 and r2=0.05.

Open in simulator
Problem 107

Compare correlation coefficients r1=-0.12 and r2=-0.07.

Problem 108

Compare correlation coefficients r1=0.70 and r2=-0.15.

know r must be between -1 and 1.
12 problems Warmup Practice Mixed Review Assessment
Problem 109

Identify whether r-value 1.2 is possible for a correlation coefficient.

Problem 110

Identify whether r-value -1.05 is possible for a correlation coefficient.

Problem 111

Identify whether r-value 0.99 is possible for a correlation coefficient.

Problem 112

Identify whether r-value -1 is possible for a correlation coefficient.

Problem 113

Identify whether r-value 0 is possible for a correlation coefficient.

Problem 114

Identify whether r-value 0.5 is possible for a correlation coefficient.

Open in simulator
Problem 115

Identify whether r-value -0.75 is possible for a correlation coefficient.

Problem 116

Identify whether r-value 1 is possible for a correlation coefficient.

Problem 117

Identify whether r-value 2 is possible for a correlation coefficient.

Problem 118

Identify whether r-value -2 is possible for a correlation coefficient.

Problem 119

Identify whether r-value 150% is possible for a correlation coefficient.

Problem 120

Identify whether r-value -150% is possible for a correlation coefficient.

distinguish linear association from nonlinear association.
12 problems Warmup Practice Mixed Review Assessment
Problem 121

Explain why r near zero for pattern points form a strong U-shape does not always mean no relationship.

Problem 122

Explain why r near zero for pattern points follow a cycle rising and falling does not always mean no relationship.

Problem 123

Explain why r near zero for pattern two separate clusters create no single linear trend does not always mean no relationship.

Problem 124

Explain why r near zero for pattern points form a strong inverted U-shape does not always mean no relationship.

Problem 125

Explain why r near zero for pattern points trace out a clear sine wave does not always mean no relationship.

Problem 126

Explain why r near zero for pattern points form a distinct circular pattern does not always mean no relationship.

Problem 127

Explain why r near zero for pattern points form an X-shape does not always mean no relationship.

Problem 128

Explain why r near zero for pattern points form a distinct V-shape does not always mean no relationship.

Problem 129

Explain why r near zero for pattern points are grouped into three distinct horizontal lines does not always mean no relationship.

Open in simulator
Problem 130

Explain why r near zero for pattern points show exponential decay followed by exponential growth does not always mean no relationship.

Problem 131

Explain why r near zero for pattern points outline a square does not always mean no relationship.

Problem 132

Explain why r near zero for pattern points form a sideways parabolic shape does not always mean no relationship.

catch sign, strength, causation, or nonlinearity mistakes.
12 problems Warmup Practice Mixed Review Assessment
Problem 133

Correct the correlation interpretation error in r=-0.9 means there is a weak relationship because the number is negative.

Problem 134

Correct the correlation interpretation error in r=0.8 proves x causes y.

Open in simulator
Problem 135

Correct the correlation interpretation error in r=1.3 shows a very strong correlation.

Problem 136

Correct the correlation interpretation error in r=0 means the variables have no relationship at all.

Problem 137

Correct the correlation interpretation error in r=0.4 indicates a strong positive correlation.

Problem 138

Correct the correlation interpretation error in r=0.1 shows a strong positive relationship.

Problem 139

Correct the correlation interpretation error in A correlation coefficient of r=-1.5 is possible for a very strong negative relationship.

Problem 140

Correct the correlation interpretation error in The strong positive correlation between ice cream sales and crime rates means that eating ice cream causes crime.

Problem 141

Correct the correlation interpretation error in r=-0.2 represents a strong negative correlation.

Problem 142

Correct the correlation interpretation error in A correlation coefficient of r=0 means that the variables are completely unrelated and independent.

Problem 143

Correct the correlation interpretation error in r=-0.7 means a weak positive relationship.

Problem 144

Correct the correlation interpretation error in A strong negative correlation between exercise and weight proves that exercise directly causes weight loss.