# SPSS INDEPENDENT T TEST PROBLEM

#### rhoda

##### New Member
Hello,

I don't know what's wrong but I've been trying to compare the means of two independent groups using the independent t test in SPSS but I fail each time I try.

This is the message I get:

Warnings
The Independent Samples table is not produced.

a. t cannot be computed because at least one of the groups is empty. Also, it says a.t cannot be computed because the standard deviation of both groups are 0

Anyone help me solve this problem?

Last edited:

#### Dason

Is one of your groups empty

#### Stats fan

##### Member
Have you entered the correct values in "Define Groups..."?

#### rhoda

##### New Member
Have you entered the correct values in "Define Groups..."?
Yes, I have entered the correct values in define groups

#### Stats fan

##### Member
In the data view, have you set the variable type of both variables to numeric? And have you put your independent (grouping) variable below Grouping Variable and your dependent variable below Test Variable(s) (and not the other way around)? And how large is the standard deviation of the dependent variable in both groups?

#### rhoda

##### New Member
Group Statistics
gender N Mean Std. Deviation Std. Error Mean
infidelity Male 44 1.00 .000a .000
female 48 1.00 .000a .000
very disappoving and very accepting of infidelity Male 5 2.00 .000a .000
female 2 2.00 .000a .000
the greater your accetance of infidelity Male 1 3.00 . .
female 0b . . .
a t cannot be computed because the standard deviations of both groups are 0.
b t cannot be computed because at least one of the groups is empty.

In the data view, have you set the variable type of both variables to numeric? And have you put your independent (grouping) variable below Grouping Variable and your dependent variable below Test Variable(s) (and not the other way around)? And how large is the standard deviation of the dependent variable in both groups?

#### Stats fan

##### Member
When you look at the scores of the variable 'infidelity' in your data set, are they all equal to 1? And when you look at the scores of the variable 'very disappoving and very accepting of infidelity' in your data set, are they all equal to 2?

And is it correct that only one male as an observed score for the variable 'the greater your accetance of infidelity', and none of the females have an observed score for this variable?

#### Berley

##### Member
You don't use a t-test with nominal and ordinal data. You appear to be trying to compare groups of people on a yes/no or likert scale. Try a chi square not a t-test. T-test is not the right approach.

#### Dr.Yossef

##### New Member
If you face this problem in SPSS
Try to transfer your data to new columns and start Analyze in SPSS again.