Question

Factorial ANOVA

1. What’s the difference between a one-way ANOVA and factorial ANOVA?
2. What is a main effect? What is an interaction? Which should you interpret first?
3. Be sure you know how to look for main effects and interactions in graphs.

Correlation

4. What is a spurious correlation? Why do we care?
5. What types of questions can a correlation analysis answer?
6. What are the 3 main characteristics of a correlation?
7. How do outliers affect a correlational analysis?

Regression

8. Which variable goes on the x-axis and which goes on the y-axis? Does it matter which goes where?
9. Why do we use multiple regression?

Answer 1

What’s the difference between a one-way ANOVA and factorial ANOVA

In one way ANOVA we have only one factor and it has different levels,while in Factorial ANOVA we have more than one factors and it has unequal number of levels.

What is a main effect? What is an interaction? Which should you interpret first

Main effect means the effect of various treatment or level,while interaction effect is to check whether their is any interrelation different units of a particular level.

What is a spurious correlation? Why do we care?

spurious correlation also known as non sense correlation,for example, suppose we find that their is a higher correlation exist between rainfall in a country with the population of country.since this does not make any sense because there will never possible relation between these two variable.

What types of questions can a correlation analysis answer?

it will only tell about the linear relation between two variable.

What are the 3 main characteristics of a correlation?

a) it is unit free.

b)it is bounded.

c)easy to calculate.

Which variable goes on the x-axis and which goes on the y-axis? Does it matter which goes where?

since we two type of variable in regression analysis

a) study variable

b) independent/explanatory variable

study variable goes on y-axis and explanatory variable goes on x axis.

yes,it matters because it we interchange these variable then my fitted regression line will be different.

Why do we use multiple regression?

sometime it is not possible to explain all variability in study variable by a single explanatory variable so we involve more than one variable which results in multiple regression.