1/13
Looks like no tags are added yet.
What is correlation?
Measuring the degree of a relationship between two independent variables (x and y)
Used to confirm the relationship between two variables
Measures strength and direction of relationship between two variables
What are two examples of correlation tests and when would you use them?
If data are normally distributed: Pearson’s
If data are not normally distributed: Spearman’s
What is regression?
Regression measures how one variable affects another (i.e. independant and dependant variable)
Used to show how one variable affects another
What do you put on the X and Y axis?
Variable which causes the relationship is called the independent variable, which you plot on the x axis
Dependent variable is plotted on the y axis
What is a confounder?
A confounder is an external variable related to both the independent variable and the outcome, potentially creating a misleading association between them.
What are the three parameters of a Linear regression model?
Slope: strength of the relationship
Intercept: ‘starting value’
Coefficient of determination (r2): reliability of the relationship
What is the least squares method?
The ‘‘best’’ fit is defined as the line that minimizes the squared vertical distances between the data points and the line.
What is a regression line?
Line that minimizes the sum of the squared vertical distances between the points and the line.
What is the importance of the R squared value and P value?
Determines the reliability of the relationship
P-value: if there is a high p value then we are less sure about a relationship but if the p value is less than 0.05 then we know that there is a strong correlation/relationship between the variables.
What does a high R squared value determine?
There is less variability around data points and the data is more reliable.
What is the reason that significant regression does not establish causation?
There maybe a confounding factor that could influence both the independent and dependent variable which would suggest that there is a strong relationship when there is not.
The analysis shows how one variable depends on another, but this dependency does not mean one variable causes another to change.
What question does correlation answer that differs from prediction (linear regression)?
Correlation establishes/discovers the relationship between the two variables.
Linear regression is used to predict one variable based on another.
What does a coefficient of an independent variable represent?
The expected change in the dependent variable for a one-unit change in the independent variable.
What is homoscedasticity?
Homoscedasticity is a statistical assumption meaning “same variance,” where the variance of errors or residuals remains constant across all levels of the independent variable.