2.6/2.7: Correlations

Question 1

What are z scores used for?

Answer 1

To compare statistics from different studies that may have different units being compared.

Question 2

What are correlational designs referred to as?

Answer 2

Observational

Question 3

What do correlational designs measure?

Answer 3

Only measurable variables

Question 4

What do correlational deigns occur with?

Answer 4

Association claims

Question 5

What is the goal of correlational designs?

Answer 5

To detect relationships in variables as they naturally occur.

Question 6

What are the 3 characteristics of Pearson’s correlation statistic?

Answer 6

1. Both variables are at least interval (continuous)
2. Both variables are normally distributed in the population
3. The association can be summarized with a straight line (linear)

Question 7

Correlation Statistic

Answer 7

Describes the strength and direction of the relationship between two quantitative variables.

Question 8

What are correlation statistics only used for?

Answer 8

Linear relationships that can be modeled with a straight line

Question 9

How is the strength of correlation visually estimated?

Answer 9

How close the points are to the trend line.

Question 10

What is the range for strength of correlation?

Answer 10

-1 to 1

Question 11

How is the direction of a linear model shown?

Answer 11

By the slope (pos/neg)

Question 12

“r”

Answer 12

A unitless descriptive statistic

Question 13

What does r not relfect

Answer 13

Original units of the variables

Question 14

What is true about effect studies?

Answer 14

They are comparable across different variables and studies

Question 15

What is the formula for “r”

Answer 15

(Z score of X * Z score of Y) / (N-1)

Question 16

Association Claims

Answer 16

Assert that two variables are related to each other

Question 17

What is true about the frequency in association claims?

Answer 17

The frequency of one variable is tied to the frequency of another.

Question 18

What does the Pearson correlation statistic support?

Answer 18

Association claims

Question 19

What do extreme values in a data set have an impact on?

Answer 19

The correlation value of the data set

Question 20

What can outliers do to the correlation value?

Answer 20

Increase, mute, or decrease the correlation

Question 21

When there is no correlation between variables, what value is the correlation around?

Answer 21

0 - Not exactly 0

Question 22

How are strong correlations presented in the graph?

Answer 22

Tightly bound to the line of slope

Question 23

What correlation score do vertical lines have?

Answer 23

Exactly 0

Question 24

What do vertical lines have a correlation score of exactly 0?

Answer 24

A relationship cannot be determined due to the same x value having multiple y values

Question 25

What 2 ways help you protect your data set from outliers?

Answer 25

1. Check one variable distribution for the present outliers first (identify) 2. Look at the scatter plots before interpreting the correlation value (identify outlier influence)

Question 26

What do Z Scores tell us?

Answer 26

How many units a value variates from the mean, in terms of standard deviation.

Question 27

What happens to the z score of a value when the value is below the mean?

Answer 27

It will be negative

Question 28

How do you calculate the z score of a value?

Answer 28

1. Find variation from the mean 2. Divide the variation by the standard deviation

Question 29

How do outliers in the same direction of the slope affect the correlation?

Answer 29

Increases the strength of the relationship (more + or -)

Question 30

How do outliers in the opposite direction of the slope affect the correlation?

Answer 30

Decreases the strength of the relationship (closer to 0)

Question 31

What effect does taking out extreme values have on the correlation value of the data set?

Answer 31

Increases the correlation (stronger)

Question 32

What should you do if the outlier is proved to be a mistake in the data?

Answer 32

1. Remove the value 2. Report the removal in the methods section; explain what impact it had on the data / why it was an outlier.