Scatter Plots and Correlation
Line of Best Fit
Residuals
Two-Way Tables
Causation Caution
100

As points go from left to right, if they generally rise, the correlation is...

Positive

100

What is slope? Give an interpretation.

How much the line rises for every increase in x. Rise over run. 

100

What is a residual plot that looks like a cloud of points randomly scattered near the horizontal line at zero

Good linear fit, the residuals are low

100

Out of 240 students surveyed, 96 said they own a dog. What is the proportion who own a dog.

96/240=0.4=40%

100

What is the one thing correlation NEVER proves.

Causation

200

The car-insurance scatter plot had r = –0.951. Is this correlation strong or weak?

Very strong (close to –1)

200

If the correlation is negative, is the number multiplying x in the best fit line positive or negative

Negative (the slope will always be in the same direction as the correlation)

200

The formula used to calculate a residual.

Actual y - Predicted y

200

In a two-way table, the ______ is found by dividing a cell by its row or column total.

Relative frequency

200

Ice cream sales and drowning deaths are positively correlated in summer months. What is the likely lurking (confounding) variable.

The temperature

300

A hat company finds that the more of a certain type of hat they make, the lower the average price of that type of hat. This is a _____ correlation

Negative

300

In a scatter plot where x represents hours worked and y represents money made, the line of best fit is y=12x+41. How much money is expected to be made if you work for 8 hours?

96 dollars

300

A residual of –8 for a data point means this:

The actual value was 8 less than the predicted value

300

A table shows 60% of males and 60% of females surveyed like pepperoni pizza. What type of association exists between gender and pizza preference.

There is no association

300

A principal sees a strong positive correlation between number of absences and failing grades, then says “absences cause students to fail.” Why is this claim not justified?

Correlation does not imply causation; there may be lurking variables or the relationship could be reversed

400

A study of hours studied vs. test score gives r ≈ 0. A reasonable conclusion would be:

There is no correlation between hours studied and test score

400

In our scatter plot of the number of orange vs. the weight of the box, what does the slope represent?

The average weight of the orange (or similar answer)

400

A residual plot that forms a U-shape instead of a random cloud: good or bad?

Bad: should be close to 0 line

400

A survey of 400 people: 120 drink coffee and exercise, 80 drink coffee and do not exercise. What is the proportion of coffee drinkers who exercise.

120/200=0.6=60%

400

Data reveal a strong positive correlation between the number of storks observed nesting in a European country and its human birth rate over 20 years. Why “storks deliver babies” is nonsense, and the real reason for the pattern.

Both stork nesting and birth rates have declined due to modernization / urbanization / fewer rural buildings, and increase in rural settings

500

A data set has r = 0.45. Changing from feet to inches for the one of the variables will do this to r:

Nothing: the relationship will stay the same, just changing units

500

In a scatter plot, the x value represents a student's height and the y value represents a student's weight. The y-intercept was at (0,50). Interpret this value and whether it makes sense.

It is extrapolating the data too far, the y-intercept doesn't make physical sense, a student who is 0 feet tall weights 50 pounds????

500

The residual plot for a good linear model fit looks like:

Residual points near the horizontal 0 line

500

A completed two-way table with totals 500 students is said to show no association between wearing glasses and playing a sport. The row percentages for “plays sport” must be what, compared to the row percentages for "doesn't play sport".

The percentages should be approximately the same

500

Researchers find a strong positive correlation between children’s shoe size and their reading level in elementary school. Explain why larger shoes do not cause better reading.

Age (or maturity/grade level) is a lurking variable that affects both shoe size and reading ability

M
e
n
u