Correlation

Covers correlation, Pearson's and Spearman's rho

What is a correaltion used for

To determine whether there is a significant relationship between two variables

What does correlation assess

Strength and direction of an association

What is a linear relationship

How well can the relationship be described by a straight line

What are the 2 test outputs of a correlation

Correlation coefficient and corresponding p-value

What is a correlation coefficient

The strength and direction of the relationship

What is the p-value

Is the relationship statistically significant

What are the directions of a correlation

Positive and negative

What does a positive correlation show

Both variables move in the same direction

What type of relationship is a positive correlation

Direct relationship

What does a negative correlation show

Variables move in opposite directions

What type of relationship is a negative correlation

Inverse relationship

How are correlations visualised

scatterplots

What should a scatterplot have

A line of best fit, a pos/neg distribution and a general impression of strength of correlation

How is the strength of a correlation measured

Correlation coefficients from -1 to +1

What is another name for correlation coefficient

effect size

How is a Pearson’s effect size denoted

r

How is a Spearman’s effect size denoted

p

What does a one tailed hypothesis for a correlation mean

expect to find either a positive or a negative effect

What does a two tailed hypothesis for a correlation mean

expect to find a relationship but don't specify direction

What is the default hypothesis to use

Two tailed

What does correlation not imply

Causation

What are the benefits of correlations

• Explore relationships

• Test hypotheses or naturally occurring associations

• Make predictions

• When experimentation isn't possible

• Guide further research

• Starting point for experiments

• Simplifies complex relationships

What is Pearson’s

A test of correlation

When is a Pearson’s test used

For a correlation when the assumptions are met

What type of test is Pearson’s

Parametric

Why are parametric tests deemed better

• More efficient and powerful

• More precise and accurate

• Higher chance of detecting a true effect

What are the assumptions of correlations

• Linearity

• Normality

• Types of data

• Outliers

What is linearity

Is there a linear relationship seen on a scatter graph

What is normality

Is the data approximately normally distributed

How can normality be tested

Visually in distributions or in a Shapiro Wilk test

What types of data can be used in a Pearson’s

Interval, ratio or ordinal if it has meaningful order

What are outliers

Points that don’t follow the trend as seen in scatterplots

What are the results after running Pearson’s in JASP

A Pearson’s r and p

What aspects need to be included when writing up a correlation

• Test type (Pearson's or Spearman’s)

• Variables (include range/unit)

• Test statistic (r or p)

• Degrees of freedom

• P-value

• Significance

• Direction

What elements of a correlation write up have to follow a strict order

Test statistic, Df and p-value

How should correlations be written out

r or p (df) = xx.xx, p = .xxx

How do you calculate sample size for a correlation

Sample size - 2

When is a Spearman’s rho used

When testing a correlation and assumptions are not met

What is Spearman’s an example of

A non-parametric test

What is necessary for a Spearman’s rho

a Monotonic trend Wh

What is a monotonic trend

Consistent movement of data in a curve, steep or flat not a U

What are the results after running Spearman’s in JASP

Spearman’s p and a p-value

When do you need to write up assumptions

When they are violated or if asked to