To aplly OR your table should be transformed in a 2x2 .  

Good!

I'll say club the two dependent variables. Now you have 3: Elementary, Secondary School and college. Make it binomial as Schooling education and Higher than School education with depressed and non depressed. Then it'll be 2x2 table.

Hope afterwards it will be easy to calculate and interpret too.

Some way, you can interpret this way- The odds of being depressed would be xxxx times higher among those undergone more than high school education (where your first schooling variable will be reference category).

Hope this will help you!

Bests

YAsh

Hi, 

See the document "OR-ResearchGate.doc".

Only "A" is relevant (statistic. significative)  because the value "1" does not belong to the 95% Confidence interval

I hope this will help you

Than you can not use OR. But you can analyse the relationship between the 2 variables using the decomposition of Likelihood ratio in df sub tables.

first row as following:

6/13 x 100 = 46.2%

7/13 x 100 = 53.8%

 Hi Hishan, 

Please see the file attached.

Elemenatry  is similar to secondary in terms of depression. The only difference is between college and the others 2.

If the outcome is a binary variable you can use logistic regression and calculate the hazards ratio for categorical variable with more than 2 categories