Plausible values multilevel analysis

[Lola Moreno-Herrero]

Hello everyone,

I am using data from the OECD project (PISA 2015). I am interested in analyzing (dependent variable) the results in mathematics of the students who participated in this wave. However, I find that there are 10 plausible values for the scores in mathematics. I would like to know how to deal with these values in a multilevel model in Stata. I would appreciate the support of someone on the list.

Regards

[Weiwen Ng]

Lola, there isn't enough information to help answer your question. The FAQ, especially section 12, gives some tips on how to provide a data example to help us help you. In particular, math scores could probably be modeled using a linear mixed effect model, because I imagine your individual-level results would be something like US SAT or GRE scores (e.g. they'd be within a certain range, say 200 to 800). It's unclear what you mean by 10 plausible values - there should be many more than 10. I could be wrong, of course! I have no clue what the PISA data actually look like. Further, I'm not an education researcher. It's possible that someone here will have worked specifically with that dataset and will tell you that normally, people do X. But you can't count on that, and people on Statalist come from many fields of research.

[Luis Muelle]

Lola,
Many people use PISA data since the early days (year 2000). Accordingly some software
has appeared to facilitate researcher’s work in education. SPSS and SAS were and are the
standards for OECD people and you can consult the Manuals that appeared in a 2009 second edition to analyze tables and pv values and are still available in their site.

In Stata, which I am a user, I have been working with some modules that need to be downloaded and installed on your computer.
I refer to my 3 favorites: pv written by McDonald; pisatools written by Procopek/Jabukosky; and repest written by Avvisatti/Kesslair. I know that repest at least has been updated to use 10 plausible values for PISA2015.
Hope this is helpful.
Regards/Saludos

[Lola Moreno-Herrero]

Thank you