The tittle of my research project is the relationship between physical exercise, sleep quality, mental health and suicide ideation. I'm confusing with my IV and DV. Is it the IV would be the condition of mental health and DV is suicide ideation? The measure of IV is physical exercise and sleep quality, is it appropriate?
Please correct me if I'm wrong, thank you.
You do not have to put everything in one sole analysis.
Sleep quality is a good measure of feeling well = being mentally healthy.
Why not put Sleep quality as dependent variable explained by the independents physical exercise and depression and suicidal ideation.
Your first IV is positively correlated with your DV and depression and suicidal ideation. are negatively correlated with sleep quality.
Your title is spelled with one t.
It can be something like:
Sleep quality is disturbed by suicidal ideation
Suicidal ideation reflects in sleep quality, depression and lack of physical exercise.
Finally the order of your variables are going to determine your title. You can have a working title and the in the end you change the title to your final head of your work.
Hard to say without knowing the research hypothesis or research question. Can you share your research question?
From the wording of your question, I assume that you are planning to use a regression type procedure for your analysis. In such analysis, variables should only be allocated roles of being IV(s) or DV(s) if there are clear grounds for regarding the DV as the causal outcome of the IVs. Or if it is made clear that the results of the analyses rely on that assumption of the causal ordering, and might not be valid if that assumption is not correct.
Looking at the set of variables whose interrelationships you plan to investigate, there does not appear to be obvious unidirectional causal influences between variables, which would justify them being given the roles of IV or DV. Intuitively, reciprocal causation would seem the most plausible pattern of causation (for example, it is plausible that depression would lead to the lack of motivation to exercise, and also more exercise could be beneficial in decreasing levels of depression.)
Ideally, your could analyse your data using structural equation modelling which would allow the examination of reciprocal causation between variables. Model refinement/simplification could lead to some of these being changed to one-way causal effects if suggested by fit indices. However, unless you have a very large sample, convergence can be difficult to achieve.
A simpler alternative approach would be to use procedures, such as partial correlations, which allow the inclusion of control variables, but which do not make assumptions about one-way causal ordering of variables, in the way that regression-type analyses do.
I suggest using an exploratory analysis method to find the best model fitting your data.