Week 11 - correlation Flashcards
What is correlation?
measures the relationship between two continuous/ numerical variables
What is a scatterplot?
displays scores of two variables
two values for each ppt plotted as a single point
1st step of correlation analysis
Use of scatterplots?
identify outliers
familiarise ourselves with data
identify initial relationships + distributions of data
Scatterplot: strength and direction of relationship between two variables
strength : closeness of the points to the line of best fit
direction : positive, negative, null
Positive correlation
As X increases, Y also increases
perfect +ve correlation: r = 1
Negative correlation
As X increases, Y decreases
Perfect -ve correlation: r = -1
Null correlation
When X increases, Y shows no constant change
null correlation: r = 0
What is pearson’s r?
A statistic which quantifies the linear relationship between two variables, ranging from 1- to 1
Its an effect size
Alternative name for pearson’s r?
Full name of pearson’s r?
correlation coefficient
pearson’s product-moment correlation coefficient
Describe the general rules for weak/ moderate/ strong correlation coefficient values
Weak -> r > 0.1
Moderate -> r > 0.3
Strong -> r > 0.5
What is covariance?
the extent to which two variables vary together
value can be -/+ - suggests direction of variance
Issue with covarience?
How to fix it
size of covariance is affected by the size of the variance of the two variables
makes comparing between different samples difficult
fix : replace N with SD
r = covariance / SD(x) x SD(y)
What is the standardised version of covariance?
Pearson’s r
What is null hypothesis
How to know if correlation is significant or not?
No correlation
To be significant, correlation needs to be bigger than critical value
Coefficient of determination
aka…?
Calculation?
AKA r squared
tells us how much variance in one variable is accounted for by the other variable
r^2
x 100 (rounded to 0 decimals)
