I need to perform a linear regression analysis in SPSS where:
-the predictor is a continuous variable representing the SD of changes over time.
-the outcome is a continuous variable measured at one time point.
-other co-variates are measured repeatedly over time.
What would be the best approach to test the association between predictor and outcome while co-varying for the other time-varying variables?
Thank you
Let me see if I understand. You have a single variable as an outcome, and a bunch of sets of covariates, where each set is a single variable with multiple measurements. And then there is the SD of something. First, what is this SD of and is it there to address heteroskedascity? On the sets of covariates, is the issue how to deal with the collinearity within each of these sets? Are you expecting similar covariances within these sets (and are the times the same)? Are you after something like estimating the slope and mean of each set, and then using these? Or wanting some more model-ish thing? How many variables do you have, how big a sample, how many time points? Are you testing a specific hypothesis or just trying to find a model with good fit?
The usual situation with time-dependent covariates also has multiple measurements of the outcome variable at those same times. But you have only one measurement of the outcome. Therefore, you will have only one row of data for each subject. And therefore, the time-dependent covariates will have to be coded as separate variables (columns). And in that case, you could, if you wished, use REGRESSION or UNIANOVA/GLM rather than MIXED. HTH.
Let me see if I understand. You have a single variable as an outcome, and a bunch of sets of covariates, where each set is a single variable with multiple measurements. And then there is the SD of something. First, what is this SD of and is it there to address heteroskedascity? On the sets of covariates, is the issue how to deal with the collinearity within each of these sets? Are you expecting similar covariances within these sets (and are the times the same)? Are you after something like estimating the slope and mean of each set, and then using these? Or wanting some more model-ish thing? How many variables do you have, how big a sample, how many time points? Are you testing a specific hypothesis or just trying to find a model with good fit?
There could be many ways of analyzing your data, but without knowing the answers to Daniel's questions. I think the simplest solution would be to treat your co-variates the same way you did your predictor: calculate how each variable changes with time and plug the delta variables in a single linear model. This seems better than using a mixed effect model since your response variable is only based on one timepoint.
Robert Jirsaraie Hi, I don't understand how the covariates are measured at several times and the DV only once. Perhaps I am just dense today but if the dv is the sd of the covariates can't you just calculate that directly. Have I got the picture wrong? Thanks, David Booth
Daniel,
The SD is the independent variable! I am testing the hypothesize that changes in the independent variable (as measured by SD over time) is associated with the outcome.
The sets of co-variates are con-founders that I would like to adjust for. They are all measured at the same time, and in different time points (7 time points). Imagine measuring someones cholesterol several times from the age of 60 to 80. There is no significant correlation among the covariates.
I am looking for the slope of the independent variable, after adjusted for the covraiets.
Hope this answers your questions.
- Badges
- Science topic
- Similar topics
- Theoretical Physics
- Time