RTP workshops

Largescale data analysis
 
Slides are here.
Here are some data for the multilevel part (rename the extension to .zip, many websites don’t allow zip files 🙂
I suggest the following useful pre-reading:
  1. For an overview of what PISA is:
    Hopfenbeck, T., Lenket, J., El Masri, Y., Cantrell, K., Ryan, J., & Baird, J-A. (2017). Lessons Learned from PISA: A Systematic Review of Peer-Reviewed Articles on the Programme for International Student Assessment. Scandinavian Journal of Educational Research. Link
  2. For the multilevel modeling section, it is useful to have a look at modules 1 to 5 from https://www.cmm.bris.ac.uk/lemma/course/ (register). Some of the materials of module 5 will feature; furthermore, the rest of the course is excellent as well.
It is useful to install some software beforehand, and perhaps already download some data.
  • Make sure you have SPSS, see software.soton.ac.uk. Don’t forget to run the license wizard.
  • IDB analyzer: this is software from IEA, you will need to register: https://www.iea.nl/data. It is used to deal with the datasets and generates SPSS scripts which will take into account the ‘complex sampling design’.
    Note that unfortunately IDB analyzer is not available in Mac format. As it creates SPSS code files. I will post these during the session so that they still can be tried out. There are R packages as well, which I will highlight but won’t use.
     
  • Multilevel software MLwin via http://www.bristol.ac.uk/cmm/software/mlwin/. There are more packages like those in R, HLM and Mplus but this one should be available to anyone with an ac.uk account. Unfortunately MLwin doesn’t have a Mac version either, but see here for information (quite technical, sorry!). You could also use the Southampton Virtual Environment. Login in with your university credentials (Note: strangely enough MLwin doesn’t always appear!).
  • A spreadsheet like Excel. Although care is needed, still useful for managing data.
  • Some data we might use, and therefore might be useful to already download as the files are rather large, is PISA 2018 data (see https://www.oecd.org/pisa/data/2018database/) and TIMSS 2015 data (at https://timssandpirls.bc.edu/timss2015/international-database/. This data is in several parts, note the filenames, as well as for grade 4 and grade 8).
  • Note that all other data files can be downloaded during the session.
  • IMPORTANT: we use this data for teaching purposes, when you use the data for research purposes, you must do an ethics application for secondary data!

R workshop

  1. Slides.
  2. R script and data.