Hello! I am working on longitudinal data (measured at T0, T1, and T2).
I aim to: 1) Analyse longitudinal differences on the outcome variable and 2) Test what variables measured at T0 can predict the outcome variable longitudinaly (i.e., T0, T1, and T2).
Someone suggested I should use multilevel modeling to do so. However, after reading about it, I realised that multilevel modeling is used when data is grouped (e.g., students within universities), which is not the case.
Can anyone explain if I can use multilevel modeling in SPSS with just 1 level, please? Does this make sense? Or is there another alternative?
Thank you very much!
Random coefficient models can handle repeated measures data and this includes longitudinal data with occasions at level 1 nested within respondents atlevel2 (who may be further nested).
This will give you the type of structures that can be handed by multilevel models:
http://www.bristol.ac.uk/cmm/learning/multilevel-models/data-structures.html
and this is a example of a two level model:
http://www.bristol.ac.uk/media-library/sites/cmm/migrated/documents/repeated-measures.pdf
and this is an accessible book length account of using multilevel models in longitudinal analysis
https://www.pilesofvariance.com/
See also:
Peugh, J. L., & Enders, C. K. (2005). Using the SPSS mixed procedure to fit cross-sectional and longitudinal multilevel models. Educational and psychological measurement, 65(5), 717-741.
https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=71d0a46249aa95f77ae387dece4aacc53c5b4d93
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis