Appropriate Test Statistic

Machine Learning and Modeling
1

I have a data described using variable or code as indicated below. I am torn between using ANOVA Test and Chi-Square Test. Can anyone suggest which is appropriate and the reason why I need to use either for my hypothesis testing at 5% significance level to see if clearance by 14 day is affected by the (i) antibiotic used, (ii) age group of the child or (iii) ear affected.?

COLUMN VARIABLE FORMAT OR CODE
1-3 ID
5 Clearance by 14 days 1=yes/0=no
7 antibiotic 1 = CEF/2 = AMO
9 Age 1 = <2 yrs/2=2 - 5 yrs 3=6+ yrs
11 Ear 1=1st ear/2= 2nd ear
2

This seems like a classic use case for logistic regression. Your response variable is binomial
Consider

fit <- glm(clearance ~ antibiotic  + age + ear, family = binomial, data = YourData)
summary(fit)

The summary will give you an initial read, but there's a wealth of hidden data in fit, such as confidence intervals. Save ANOVA for continuous data.

3

@technocrat Is there any other apart from logistic regression?

4

Well, it's hard for me to say without a syllabus and an understanding of your class rules of engagement.

Anova isn't designed for categorical variables. It depends on variance of continuous variables.

Chi-squared can be used for categorical variables, if you have a sufficient number of observations to be in Central Limit Theorem territory. If you have fewer than about 30 observations, you're probably in Student's t land.

1 Like
5

Please take a moment to familiarise yourself with our site's homework policy (see below):

1 Like
6

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.