Measuring the Dependent Variable

#1
Hi all,

I am a PhD student conducting research related to student satisfaction. I am struggling to find a way to measure my dependent variable. In my operationalization table, I have 5 constructs (Independent variables) as A1, A2, A3, A4 and A5. My dependent variable student satisfaction (X) is not included in the operationalization table and it is measured by the 5 independent variables (A1, A2, A3, A4 and A5). In my questionnaire (5-point Likert Scale), A1 has 4 items; A2 has 4 items; A3 has 6 items; A4 has 4 items and A5 has 9 items. But student satisfaction (X) does not have separate questions.
What is the correct way of measuring X by using A1, A2, A3, A4 and A5? Can I use the mean values of the 5 constructs to measure X?

Thank you.
 

Karabiner

TS Contributor
#2
How large is your sample size?
Why did you choose A1 to A5 as possible indicators for student satisfaction, did you have references in the literature for this?

How will X be used in the further course of the study, what will you need it for?

With kind regards

Karabiner
 
#3
Thank you very much for your reply.
I have 52 responses in my dataset.
Actually, I developed the operationalization table by referring to the literature and I have literature saying that the student satisfaction can be measured by those constructs. Further, my target is to develop a model for a effective course design. For that, I want to know whether the students are satisfied with the constructs.
I don't have separate questions to measure student satisfaction directly and want to measure that by my constructs.
 

Karabiner

TS Contributor
#4
So why do you not use just one of these constructs, if literature supports each of them?
Why do you feel the need to combine them?

With kind regards

Karabiner
 
#5
The 5 constructs are course design aspects. I want to measure how those course design aspects relate with student satisfaction. So, I don't have a way to take one of the constructs as a dependent variable. That is why I thought to combine those. But I'm not sure whether it is correct or not.

Thank you.
 

Karabiner

TS Contributor
#6
Is this one-group a pre-post study or something like that? If the constructs represent different aspects,
then maybe it would be more informative to analyse them separately, instead of collapsing them. It is
problematic to develop and check a measurement instrument within the very same sample in which
the final instrument will be used to answer the substantial research questions.

With kind regards

Karabiner