1/8
Looks like no tags are added yet.
Correlation
Asks whether two continuous variables are associated linearly. It measures the strength and direction of a linear relationship. Are X1 and X2 associated linearly? Is there a pattern?
Regression
Asks whether one variable predicts another and helps describe the relationship’s slope. Does X predict Y? What is the relationship? ex. How much does plant growth change as temperature increases
Pearson’s correlation coefficient ( r )
values range from -1 to +1
Pearson’s correlation assumptions
Sampling independence, linear relationship, and normally distributed bivariate variance
Spearman’s correlation
a non-parametric alternative to Pearson’s correlation; uses the ranks of the data rather than the original values
Non-parametric tests
Mann-Whitney-Wilcoxon test and Kruskal-Wallis test; these tests compare medians rather than means
What do you use instead of a t-test?
Mann-Whitney-Wilcoxin test
What do you use instead of an ANOVA test?
Kruskal-Wallis test
correlation matrix
shows correlation between multiple variables