Which test should be used with 1 dependent variable and 7 independent variables?

#1
Hello everyone
I'm carrying out a research which tries to analyse factors affecting coffee consumption behavior, there are 7 independent variables on hand and 1 dependent variable which is coffee consumption behavior
Is Multiple regression analysis can be used while there is relation among independent variables as well? If not what other tests can I use?, if yes also what other additional tests can be used? your help will be much appreciated

The following are 7 hypothesis as well as the conceptual framework:
Untitled-4.jpg
 

obh

Active Member
#2
Hi Yama,

As I understand when the dependencies are only between the predictors, say no round dependencies with the DV like (X1=>Y and Y=> X1)
You can use the regression model.

When there is a high correlation between the predictors is too high (Multicollinearity) you may not know the effect of each variable
say the coefficients won't be accurate but the DV will be ok.

GretaGarbo what do you think?
 

Karabiner

TS Contributor
#3
Your 7 hypothesis just require 7 correlation analyses (Pearson or Spearman, depending on the scale levels of the variables), as it seems. Do you 5hink you need something more elaborate?
How large is your sample size, by the way?

Maybe you should describe the context, the theoretical background, and the goals of your study in more detail.

With kind regards

Karabiner
 
#4
Hi Yama,

As I understand when the dependencies are only between the predictors, say no round dependencies with the DV like (X1=>Y and Y=> X1)
You can use the regression model.

When there is a high correlation between the predictors is too high (Multicollinearity) you may not know the effect of each variable
say the coefficients won't be accurate but the DV will be ok.

GretaGarbo what do you think?
Hi dear
sorry for the late reply as there was some issues i had to attend to and couldn't reply on time
sorry to sound ridiculous :(, but I don't know much about the statistical analysis,
Actually my initial model was as of figure 1, and my adviser suggested that if there is relation among independent variables, analysis will fail so I draw the figure 2 and he accepted it,
The actual conceptual model is as of figure 3 in my opinion, and I suggested it to the adviser but he said it will get too complected.
4 of independent variables, while affecting the dependent variable, also affects 2 independent variables.
I want to know which of the statistical analysis are enough to carry out with regards to the hypothesizes on hand
Untitled-4.jpg
 

Attachments

Last edited:
#5
Your 7 hypothesis just require 7 correlation analyses (Pearson or Spearman, depending on the scale levels of the variables), as it seems. Do you 5hink you need something more elaborate?
How large is your sample size, by the way?

Maybe you should describe the context, the theoretical background, and the goals of your study in more detail.

With kind regards

Karabiner
Hi dear Karabiner
Sorry for my late reply
The sample size is 267,
In my thesis titled (analyzing factors affecting coffee consumption behavior) I've got the 7 above mentioned variables, which are known to affect coffee consumption behavior, through exploring various conducted previous studies
The thesis has two sections, first part contains secondary data, and the second part primary data which is collected by using questionnaire as a tool , which I've done already and the data is ready to be analysed.
Now to be up front I don't know much about statistical analysis.
Actually my initial model was as of figure 1, and my adviser suggested that if there is relation among independent variables, analysis will fail so I draw the figure 2 and he accepted it,
The actual conceptual model is as of figure 3 in my opinion, and I suggested it to the adviser but he said it will get too complected.
Now what do you think which of the models conveys the concept best with regards to given hypothesizes ( we want to determine according the collected data, which of the factors have a positive effect on coffee consumption behavior)
It should be also mentioned that the 7 given variables as can be seen in the image bellow are placed in the survey form in 7 sections, each section having their own 3-4 statement with 1-5 degree of agreement levels.

Is the correlation analysis (Pearson) going to determine what is intended to? And can this analysis be carried out using SPSS?
I can provide the collected data from questionnaires if needed

Thanks for your help

Sincerely
Untitled-4.jpg
 
Last edited: