#1
Hello, suppose i have example of buyer's age from 3 product

Notebook:35,25,40,30
Smartphone:20,21,25,30,25
Watch:35,20,30,45,35

I want to make a hypothesis like this:
"Is buyer's age will significanly affect product choices?"

So the questions:
1. I confuse how to choose independent variable. Is Product choices will be dependent variable?
Can ANOVA calculate where dependent variable is categorical?

2. Is the hypothesis will be Same with:
"From 3 product choices, are there significanly buyer's age?"

Thank you for your attention. Sorry for my english
 

obh

Active Member
#2
hI Ichsan,

You may compare the average age per product to see if the difference in average age is significant.
In this case, you will use the one way ANOVA:

H0: μ_Note = μ_Smart =μ_Watch
H1: otherwise

In this case, the product is the IV while the age is the DV

Another option is to estimate the probability to buy a product based on age.
In this case, you will use the multinomial logistic regression

example of the result: on age 22: 0.31 Notebook, 0.2 Smartphone, 0.49 Watch

In this case, the age is the IV is the product while the product is the DV
 
Last edited: