## analysis jobsatisfaction small company

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

woody83
Posts: 2
Joined: Fri Dec 13, 2013 1:53 pm

### analysis jobsatisfaction small company

10 years later after higher education I'm getting back interested in statistics and market research.
I have a completed survey of a small company. There has been informed to general information such as gender, age, type of job... and specific questions about job satisfaction. The questions about job satisfaction are each quoted on a scale of 10 points.
My first question (and hopefully a good one) is how to check in spss if the personel is significant un-/satisfied at their job in each question?
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

### Re: analysis jobsatisfaction small company

Well, it depends on how you define (un)satisfied.

One possible way may be to say: everything that differs significantly form a certain value, I call significantly (un) satisfied.

For instance, is the scale goes from 1 to 10, the mean is 5.5 .
You could argue that when on a certain question they differ significantly from 5.5, they are significantly (un)satisfied.

If this is the way you want to go, you can use a one-sample t-test with 5.5 as the test value.

If this is not how you want to define it, you should indicate here what you mean by "significantly (un)satisfied", and we can go from there.
woody83
Posts: 2
Joined: Fri Dec 13, 2013 1:53 pm

### Re: analysis jobsatisfaction small company

In satisfaction researches with scales on 10 I read levels of 6 = your customers run away, 7 = ok, 8 = is very good, 8,5 = excellent and a level of 9/10 is almost untouchable. I wonder how these levels are defined. And is the t-test still the right tool in this kind of research?
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

### Re: analysis jobsatisfaction small company

Well there is a difference between theoretical decisions (e.g. what do I think defines satisfactory) and statistical questions, we can probably only help you with the latter.
So you need to make decisions about what it actually is you think you measured, what defines satisfactory and what defines unsatisfactory.