Hi I would appreciate some guidance on my proposed quantitative methodology and design. As per the uploaded diagram, my model has four independent variables (IV’s) and two dependent variables (DV’s). All of the variables will be measured using summated scales. I intend to use the averages of the summated scale scores to represent each of the variables. I then intend to analyse the relationships between the IV’s and DV’s using multiple regression. My research question does not extent to examining the relationships between the two DV’s (these are two desirable but different outcomes). If my research questions extended to covarience/relationships believe the DV’s then I believe that SEM would be more appropriate. However in this case, although I have two DV’s in my model as they are entirely separate, I believe I can test this using two runs of a multiple regression in SPSS (one for the IV’s on DV A and one for the IV’s on DV B). but I would appreciate any feedback. Many thanks Nick
Is there a relationship between IV and DV? Use ANOVA, MANOVA, or that type of analysis.
What is the relationship between IV and DV? Use regression. Maybe you want to predict outcomes for unmeasured values of the IV within the range of observed values, or simply gain a better understanding of the relationship between DV and IV.
Hi yes, based on literature review and qualitative exploration I have hypothesized relationships between the predictors (IV's) and the outcomes (DV's) notwithstanding the need to avoid overstating causation as this is cross-sectional survey. I am aiming to understand the significance and strength of the relationships.
As Vincent wrote, if you want to test that many DVs, you need an SEM setup. But I would start with two simultaneous equations in regression and then cross compare with SEM results.
It's fine to use two distinct models with the same data to test two different research questions. If you want to test the interrelationships between the two dependent variables, then you need to do more, but that depends on how much you expect the two dependent variables to be interrelated. Also, even if you expect the two to be interrelated, testing two independent models might be a good first step (depending on what you wish to test).
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running