Double condition (if)

[#1]

Hello, 

I have a data set with two randomization (code 1 or 2) arm and 5 types (1,2...,5) of interventions and my interest variable is a scale pain (1,... to 10).

How can I make a khi2 test analysis between the two arm and my scale pain for 1 of the five intervention.

I create a new variable "Chir Abdo - Rando" with formula to make analysis X,Y "Chir Abdo - Rando" in function "EVA <4".

For example I begin (but whithout succeded) like this :

But doesn't work ...

Here the result I'm trying to have :

You can find my data base attached. There is color in regard of the mentionned variables.

"Bras prise en charge habituelle = variable "Groupe randomisation" = 1

"Bras "toucher-détente" = "Groupe randomisation" = 2

"EVA < 4" = variable "EVA < 4" = 1

"EVA > or = 4" = "EVA < 4" with code = 0

Thank you ++++

[#2]

Maybe something like this?

if (
     intervention == 1 & random == 1 , 1, // a test, and an answer for the test
     intervention == 1 & random != 0 , 2, // another test, and an answer
     0  // if none of the previous tests, use this answer
)

If Secrets might help too. Watch out for missing values.

[#3]

Here the solution, it works.

Thank you Craige !

[#4]

Great! 

Make sure the answer is correct when intervention is not 1. The picture is missing a value for the last else.

[#5]

It is only if I leave a value for what else that it work.

I don't want to take in account the other type of (chirurgical) intervention, and no attribute values, to make a khi2 test only on the sub saple of patients who underwent abdominal surgery.