STATS 1 CORRELATION

Question 1

What are the advantages of correlational research?

Answer 1

Useful for looking at variables we cannot control experimentally

Question 2

What are the disadvantages of correlational research?

Answer 2

Correlation does not imply causation
Sensitive to outliers
Attenuation due to unreliability
Restriction of range can underestimate pop. correlation severely

Question 3

What is covariance?

Answer 3

Covariance is the degree to which two variables vary together

Question 4

What is the Pearson Correlation coefficient?

Answer 4

A standardised measure of the degree of relationship (correlation) between two variables using covariance and standard deviations

Question 5

What does the Pearson correlation coefficient reflect?

Answer 5

The strength of the relationship, NOT the significance

Question 6

How do we test the significance of the correlation coefficient?

Answer 6

Assume null hypothesis, so population mean should be 0, then obtain the probably for the correlation coefficient to emerge from a population with correlation 0

Question 7

Where do we obtain the probability for our correlation to emerge from a population with correlation zero?

Answer 7

The t-distribution

Question 8

What is the difference between statistical significance and effect size of a correlation?

Answer 8

Effect size of a correlation is the strength of the correlation, the significance of a correlation is the probability that the correlation would emerge from a population with correlation 0

Question 9

What are the effect sizes of correlation coefficient?

Answer 9

.1 - small
.3 - medium
.5 - large

Question 10

What is the coefficient of determination?

Answer 10

R squared - gives the proportion of variance shared between the two variables

Question 11

What is restriction of range?

Answer 11

Problem in interpreting correlation - restricting the range of sample may underestimate the population correlation severely

Question 12

What is attenuation due to unreliability?

Answer 12

Occurs when the measures of X and Y are unreliable, which reduces the correlation coefficient

Question 13

What impact do outliers have?

Answer 13

Inflate or deflate the correlation coefficient

Question 14

What is Spearman’s RHO?

Answer 14

Used for non-parametric correlations (data that is not equal-intervals)

Question 15

Why would you consider using Spearman’s RHO for parametric data?

Answer 15

If you have outliers it can be useful because it is less sensitive to outliers

Question 16

What is Kendall’s tau-b?

Answer 16

If you have a small sample with many tied ranks, you should use Kendall’s tau-b

Question 17

What do you have to do to your data before you can implement Spearman’s RHO or Kendall’s tau-b?

Answer 17

Rank it

Question 18

What is a point-biserial correlation?

Answer 18

Used to describe the relationship between dichotomous variable and continuous variable (e.g. sex and IQ)