Correlation Module

Question 1

What is the difference between univariate and bivariate?

Answer 1

one variable and two variables respectively

Question 2

One of the methods that can be used to examine the correlation between two variables

Answer 2

Pearson’s R

Question 3

We can only use Pearson’s R when the data is _____

Answer 3

bivariate and continuous

Question 4

Does swapping the X and Y axis affect the value of the Pearson’s R and what is the name of the relationship

Answer 4

No, swapping the x and y axis does not affect the r value as pearson’s r is bi-directional (i.e., correlation between A = B and correlation between B = A)

symmetrical relationship

Question 5

Explain positive, negative and no correlation linear

Answer 5

positive
r = between 0 - 1
the smaller the x value is, the smaller the y value is
X increases, Y also increases

negative
r = between 0 - -1
the smaller the x value is, the larger the y value is
x increases, y decreases

no correlation
r = 0
there is no association between x and y
x and y is independent

Question 6

What is the step after obtaining a r value

Answer 6

perform a null hypothesis significance testing to test whether the r value is consistent with the result of the NHST

assuming to rho (population correlation coefficient) is 0 (H0) > if H0 is rejected > H1 (there is a correlation between the two variables)

Question 7

What are the rough rules of the size of correlation, and are they universal?

Answer 7

.1 - .3 (small)
.3 - .5 (medium)
.5 - .7 (large)

However, they are not universal as the size of the correlation also depends on the context

Question 8

What are the three things that should be careful when looking at correlation?

Answer 8

correlation does not necessarily mean causation
extreme score (outliers) can affect the effect size both negatively and positively
regression towards the mean (the tendency that the next point will be at the different direction as the previous point)

Question 9

What are variability and covariability

Answer 9

variability = how much each variable varies observation by observation
covariability = how much do the two variables vary together

Question 10

what are the three things that have to bear in mind when looking at correlation

Answer 10

correlation doesn’t mean causation
extreme score (outliers) can affect the accuracy of the model
regression towards the mean

Question 11

What are the three characteristics of residuals

Answer 11

independent
normality: (errors are normally distributed with a mean of 0)
homoscedasticity: all values are equal