Descriptive Statistics 2

Question 1

What is skewness?

Answer 1

Asymmetry

Question 2

What is positive skewness?

Answer 2

When the data ‘skews’ out to the right

Question 3

What is negative skewness?

Answer 3

When the data ‘skews’ out to the left.

Question 4

What is kurtosis?

Answer 4

‘pointiness’

Question 5

What is the R function to get lots of descriptive stats?

Answer 5

describe()

Question 6

If kurtosis is greater than 0 it means that

Answer 6

The graph is too pointy (focused around the center)

compared to normal

Question 7

If kurtosis is less than 0 it means that

Answer 7

The graph is ‘too flat’ (compared to normal)

Question 8

What does Tukey’s ‘five number summary’ consist of?

Answer 8

Minimum value (0th percentile)

25% percentile

Median (50th percentile)

75th percentile

Maximum (100th percentile)

Question 9

A boxplot shows everything that is a part of?

Answer 9

Tukey’s five number summary

Question 10

What do scatter plots do?

Answer 10

Show the relationship between numeric variables

(plot one variable on x-axis, plot two on y-axis)

Question 11

What is a positive relationship?

What does a positive relationship look like in a scatter plot?

Answer 11

When one variable goes up, the other goes up too.

looks like going up a hill.

Question 12

What is a negative relationship?

What does it look like in a scatterplot?

Answer 12

When one variable goes up, the other goes down.

looks like going down a hill

Question 13

What does a strong correlational relationship tell us?

Answer 13

If we know the value of one variable it tells you a lot about the other one.

Question 14

What does a weak correlational relationship tell us?

Answer 14

If we know the value of one variable it tells us very little about the other variable.

Question 15

How do you make a scatter plot using R?

Answer 15

plot (x, y)

Question 16

What is a Pearson’s correlation?

Answer 16

Measures the strength and direction of a relationship.

It assumes the relationship is linear.

Question 17

Can Pearson’s work with non-linear correlations?

Answer 17

No

Question 18

How do you do a Pearson’s correlational analysis in R?

Answer 18

cor.test(x, y)

Question 19

What do you use Spearman correlations (Spearman’s Rho) for?

Answer 19

To measure non-linear relationships.

Question 20

In statistics, ρ refers to?

Answer 20

Spearman’s rho

Question 21

What are the characteristics of Spearman’s correlation?

Answer 21

It’s a number from -1 (perfect negative) to 1 (perfect positive)

It doesn’t assume the relationship is linear

All it assumes is monotonicity

Question 22

A curve is monotonic when?

Answer 22

If it always goes up or always goes down.

(Doesn’t have to be linear though)

Question 23

A curve is non-monotonic when?

Answer 23

It goes up and down at certain points.

Question 24

How do you do a correlational analysis in R?

Answer 24

library(lsr)

correlate()