Ice cream sales and drowning incidents both increase in summer. This shows: — A positive association, but no causal link (both increase with summer heat).Classic example of correlation without causation — both are driven by summer heat and increased outdoor activity.
Two scatter plots both show positive association. How can you tell which is STRONGER? — The one where points cluster closer to the trend line is stronger.Strength of association is about how tightly data clusters around the trend, not slope steepness or data volume.
A student says "since temperature and ice cream sales are positively correlated, buying more ice cream causes higher temperatures." What is wrong? — Correlation does not imply causation; temperature likely drives both.Classic causation fallacy. Higher temperature causes more ice cream buying — not the reverse. Correlation never proves causation.
A scatter plot has clusters of points in the upper-left and lower-right. This pattern suggests: — Negative association.Upper-left and lower-right clustering means when x is small y is large, and when x is large y is small — negative association.
The line y = 1.5x + 4 fits data with x ranging from 0 to 20. Which prediction is most reliable? — y at x = 10.x = 10 is within the data range (0–20) — this is interpolation, which is most reliable.
The line of best fit for age (x, years) vs. height (y, cm) in children is y = 7x + 50. Predict height at age 40. — Unreliable — extreme extrapolation.Adults stop growing. Extrapolating to age 40 would give 330 cm — clearly unrealistic. Extreme extrapolation is unreliable.
A residual is the difference between an actual data value and the: — Predicted value from the line of best fit.Residual = actual y − predicted y. Positive residuals mean the actual value is above the line; negative means below.
Two students draw lines of best fit for the same scatter plot. Student A's line has slope 2.5, Student B's has slope 2.8. Who drew the better fit? — Need to check which has points more equally above/below.The better fit has points most equally distributed above and below the line. Slope alone doesn't determine which is better.
What does a line of best fit with slope = 0 tell you about the association? — No linear association (the variables are unrelated linearly).Slope = 0 means the line is horizontal — y doesn't change as x changes. This indicates no linear association between the variables.
A two-way table shows 200 students: 80 boys and 120 girls. 60 boys and 90 girls prefer basketball. Are preferences independent of gender? — Yes — 60/80 = 75% and 90/120 = 75%: same rate for both genders.60/80 = 75% of boys; 90/120 = 75% of girls. Equal proportions → gender and preference appear independent.
What does it mean if two categorical variables are ASSOCIATED in a two-way table? — The proportions for one variable differ across categories of the other variable.Association means the relative frequencies differ across groups. If boys and girls have different proportions of sport preference, gender and sport preference are associated.
In a two-way table, how do you find the overall relative frequency for a joint cell? — Cell ÷ grand total.Overall (or joint) relative frequency = cell count ÷ grand total. This shows what proportion of ALL data points fall in both categories.
A conditional relative frequency answers which type of question? — Given that a student is in group A, what proportion prefer X?.Conditional relative frequency answers "given one category, what is the proportion in another?" — it divides by row or column total, not grand total.
In a survey of 200 people: 120 are adults, 80 are teens. 72 adults and 56 teens prefer video streaming. What proportion of teens prefer streaming? — 56/80 = 70%.Conditional: given teen (80 total), 56 prefer streaming. 56/80 = 70%.