You can add all questions with the difference categories into one variable you will find it in spss

I don't think it is a good idea to add variables with such different content. If your goal is to create a single variable with 6 categories (2 x 3), then you can do with a set of IF statements.

But if you want to consider the effect of each variable plus the 2 x 3 combination, that would be considered an interaction effect. So, if you have an interval level dependent variable, you could use Analysis of Variance.

Hello Bi,

Here's one way to generate the six distinct categories you requested.

If sex (male, female) is coded numerically as 1, 2; and

If education (middle school, high school, university) is coded numerically as 1,2,3;

then this statement will create a new variable as desired:

compute category = (sex - 1) * 3 + education .

The resultant values for the new variable, "category," would be:

1 = Male, middle school

2 = Male, high school

3 = Male, university

4 = Female, middle school

5 = Female, high school

6 = Female, university

Please note, I'm not addressing whether this is better than a 2 x 3 framework for subsequent analysis.

Good luck with your work.

Thank you so much. I want to create a new variable with six categories( 1=male, middle school, 2= male, high school......), just exactly as David mentioned, however, I do not know how to compute the new variable in SPSS program. What kind of step need I do?

I will not recommend recoding the variables into one. But if you must I'd suggest you cross-tabulate the variables first.

To combine variables you go to the transform Menu and click compute. Type your new variable in the space Target Variable.

Find the variables in the list of all your variables under type& Label that you wish to add and click on it. Click on the arrow next to the list of variables. Your variable will move to the box Numeric Expression. Do this to all the variables. Then click Ok. You now have your new variable.

The Compute statement from David Morse does indeed work the way you want it to.

But if you want to do a regression analysis or Analysis of Variance, this is not the correct approach. And the same is true for chi-square.

As these questions are simple to respond to for participants as well as useful for analysis, keeping them separate makes sense. Is there another reason you are lookong to combine them?

Hello Dr. CohenMiller. Yes, you are right, it is better to keep them separate here. In my research, I actually want to combine another two variables, But I am worried that I will not be able to describe my problems, so I chose two simple variables to replace my problems. What I actually did was to connect the two events.

@ David L Morgan, Thank you very much. I actually want to connect two events together ( participate first and second). I worry that I cannot describe my problem clearly, so I use two simple variables. I want to use the logistics regression analysis since both depend and independent variables are categories.

Go for 2 way ANOVA.

I understand why you choose to give a simpler example. To start with, logistic regression requires that you have a 2-category dependent variable that you want to explain. Is that the case, and if so, what is it?