Difficulty: Easy & Basic
What is the main purpose of regression analysis?
A Predict a variable
B Classify observations
C Measure central tendency
D Count categories
Answer: A. Predict a variable Regression analysis models how an outcome changes with predictors and supports estimates of that outcome.
Simple linear regression studies the relationship between how many variables?
A One variable
B Two variables
C Three variables
D Four variables
Answer: B. Two variables Simple linear regression uses one predictor and one outcome variable.
In regression of Y on X, which variable is predicted?
A X
B Both equally
C Y
D Neither variable
Answer: C. Y The regression of Y on X estimates the value of Y from a given value of X.
In regression of X on Y, which variable is predicted?
A Y
B Both equally
C Neither variable
D X
Answer: D. X The regression of X on Y estimates X for a given value of Y.
In y=a+bx, what does b represent?
A Slope coefficient
B Intercept
C Residual
D Predicted mean
Answer: A. Slope coefficient The coefficient b is the slope and gives the predicted change in y for a one-unit increase in x.
In y=a+bx, what does a represent?
A Slope
B Y-intercept
C Residual
D Correlation
Answer: B. Y-intercept The intercept a is the model’s predicted value of y when x equals zero.
What is a residual in regression?
A Predicted minus mean
B Slope times intercept
C Observed minus predicted
D Mean minus median
Answer: C. Observed minus predicted A residual is the observed outcome minus its fitted or predicted value.
What does the least-squares method minimize?
A Absolute slope
B Sample size
C Correlation sign
D Squared residual total
Answer: D. Squared residual total Least squares chooses coefficients that minimize the sum of squared residuals.
For regression of Y on X, which deviations are squared by least squares?
A Vertical residuals
B Horizontal residuals
C Rank differences
D Class boundaries
Answer: A. Vertical residuals Y-on-X regression minimizes squared differences between observed and fitted Y values.
A regression line with an intercept passes through which point?
A The origin always
B The two variable means
C The two maxima
D The two medians always
Answer: B. The two variable means The least-squares regression line passes through the point formed by the sample means.
Where do the two regression lines intersect?
A At the origin always
B At maximum values
C At the mean point
D At minimum values
Answer: C. At the mean point Both regression lines pass through (xΜ,Θ³), so they intersect at that point.
Which line estimates Y for a given X?
A Regression of X on Y
B Both lines equally
C Neither regression line
D Regression of Y on X
Answer: D. Regression of Y on X The Y-on-X equation has Y as the outcome being estimated from X.
Which line estimates X for a given Y?
A Regression of X on Y
B Regression of Y on X
C Both lines equally
D Neither regression line
Answer: A. Regression of X on Y The X-on-Y equation estimates X using the observed value of Y.
How does regression differ from correlation?
A Correlation predicts outcomes
B Regression predicts outcomes
C Both prove causation
D Neither describes variables
Answer: B. Regression predicts outcomes Regression models prediction, while correlation summarizes association between variables.
Does a regression equation by itself prove causation?
A Yes, always
B Only with positive slope
C No
D Only with large samples
Answer: C. No Prediction from an observed association does not alone establish a causal effect.
Why can extrapolating a regression line be risky?
A It changes the sample mean
B It removes residuals
C It reverses all ranks
D It extends beyond observed data
Answer: D. It extends beyond observed data A fitted trend may not continue beyond the range of observed predictor values.
A one-unit increase in X changes predicted Y by what?
A The slope amount
B The intercept amount
C The residual amount
D The correlation amount
Answer: A. The slope amount In a linear regression, the slope is the predicted change in Y for one additional unit of X.
What are the units of a Y-on-X slope?
A X units per Y unit
B Y units per X unit
C Squared Y units
D No units
Answer: B. Y units per X unit The slope compares a change in Y with a one-unit change in X.
When can a regression intercept lack practical meaning?
A When the slope is positive
B When the sample is large
C When X zero is irrelevant
D When residuals sum zero
Answer: C. When X zero is irrelevant The intercept is mathematical, but may not describe a realistic case if X=0 is outside context.
For ordinary least squares with an intercept, what is the residual sum?
A One
B The slope
C The variance
D Zero
Answer: D. Zero With an intercept, the least-squares residuals sum to zero.
What pattern in a residual plot supports a simple linear model?
A No systematic pattern
B A clear curve
C A widening funnel
D A steady upward trend
Answer: A. No systematic pattern Random scatter without structure is consistent with the linear model’s form.
What does a negative regression slope indicate?
A Y rises as X rises
B Y falls as X rises
C Y remains constant
D X must be negative
Answer: B. Y falls as X rises A negative slope means larger X values correspond to lower predicted Y values.
What is the outcome variable also called?
A Predictor variable
B Control limit
C Dependent variable
D Residual variable
Answer: C. Dependent variable The outcome is called dependent because its modeled value depends on the predictor.
What is X commonly called in Y-on-X regression?
A Outcome variable
B Residual variable
C Response error
D Predictor variable
Answer: D. Predictor variable X is the predictor used to estimate the dependent outcome Y.
What can a large residual indicate for one observation?
A A poor individual fit
B Perfect prediction
C Zero slope
D No sample variation
Answer: A. A poor individual fit A large residual means the observed outcome lies far from its fitted value.