Two categorical and one continuous dependent variables + one categorical independent variable. Which test should I run?

#1
I am not sure if I should run MONOVA or some other test? I have collected data from two different sample groups. I have one categorical and two continuous dependent variables + one categorical independent variable. Because of the continuous dependent variables, I'm not too sure about using MANOVA. Would appreciate your help in this regard. Thanks in advance.

Edit: Details of the study: I'm sorry for not including all the information before. It's a netnographic study which will include analysis of 1000 different posts from two online communities. The independent variable is anonymity/non-anonymity of the community which is a categorical variable yes/no (At least that's what I think but I can be wrong, let me know if I am), the dependent variables are consumption(No. of followers in each community - 38,000 and 123,000 - I think this is also categorical?), contribution(no. of comments and likes on each post - continuous variable), and creation(whether the post in their own or forwarded from somewhere - 2 options yes/no so this is also categorical?). I want to check the effect of anonymity on each of the dependent variables and all of them together.
 
Last edited:

Karabiner

TS Contributor
#3
Could you please describe your actiual study? What is its topic, what are the research questions,
how was data collection designed, which variables did you include and how ere they measured,
how large is your sample size (cf. question above)? It is difficult to come up with solutions if
the problem is described without giving crucial information.

With kind regards

Karabiner
 
#4
I'm sorry for not including all the information before. It's a netnographic study which will include analysis of 1000 different posts from two online communities. The independent variable is anonymity/non-anonymity of the community which is a categorical variable yes/no (At least that's what I think but I can be wrong, let me know if I am), the dependent variables are consumption(No. of followers in each community - 38,000 and 123,000 - I think this is also categorical?), contribution(no. of comments and likes on each post - continuous variable), and creation(whether the post in their own or forwarded from somewhere - 2 options yes/no so this is also categorical?). I want to check the effect of anonymity on each of the dependent variables and all of them together.
 

staassis

Active Member
#5
Still do not understand variable "No. of followers". What are your reasons for considering the variable categorical?.. Is the variable defined for each post. My understanding is that, in your analysis, post is the observation unit, i.e. 1 observation = 1 post.
 

noetsi

Fortran must die
#6
logistic regression will work with your dependent categorical variable. Linear regression works with the interval dependent variables. Both have one and only one dependent variable at a time. MANOVA will work with multiple dependent variables but I know only of it, I have never used it. It may require only interval DV.