Based on the above content I have certain doubts: 1) I have (1 Independent Variable and 3 Intermediary variable and 1 Dependent Variable in my model. Shall I count both (IV and intermediary Variable) for computation of degree of freedom? 2) Every construct has multiple items and likert scale responses are collected on them. Can use Liker scale responses for computation if Yes then Do I need to compute the aggregate score construct wise to apply this test or any alternative is there to test the outlier? Please share 3) I have an integrated framework, analysis will be done in 2 stages. Should I check MD for each stage? One of my variable look like a mediator - it may or it may not. It will be a DV in first stage and subsequently may be an IV in 2nd stage of the analysis. Stage 1 and stage 2 does not have any connection. If I consider both and compute the probability than probably each testing will generate different Prob_MD / outliers to be deleted. What to do in such a case. 4) Should I include demographic questions also as part for computation of degree of freedom, as the prob outliers results may vary when I omit or include the same. Is it relevance of using demographic questions as a component to compute degree of freedom? 5) Identifying & handling outliers are about independent variables. Does it mean that we don;t have to identify & handle the outliers of dependent variables?
You have to perform the multivariate analysis for each instrument
Check the link below for more details:
http://statwiki.kolobkreations.com/index.php?title=Data_screening
- Badges
- Science topic
- Similar topics
- Engineering
- Computer Engineering
- Computer