corentinchoisy
/
simul_confusion
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
Rasch-based Cohort study simulations via stata/simIRT
You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
5 Commits
1 Branch
0 Tags
16 MiB
Stata 98.3%
SAS 1.6%
 
 
master
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '31d0743f4d'
${ noResults }
Go to file
Corentin Choisy 31d0743f4d Corrected typo in catalogue 10 months ago
Modules Initial commit 10 months ago
.gitignore Initial commit 10 months ago
DATA_LICENSE DATA LICENSE 10 months ago
LICENSE LICENSE 10 months ago
README.md Updated catalogue and readme 10 months ago
autorun.sh Initial commit 10 months ago
catalogue.md Corrected typo in catalogue 10 months ago
prepare_file_structure.sh Initial commit 10 months ago

README.md
Unescape Escape

Cohort simulations

This repository contains all code files related to our ROSALI/Resiuals Cohort study simulation project. In order to save disk space, data files are not stored on this server and are instead available on https://osf.io.

File Structure

📦 simul_these
├─ catalogue.md           - List and description of scenarios
├─ 🗂️ Analysis               - ANALYSIS RESULTS
├─ 🗂️ Data                   - GENERATED DATASETS
│  ├─ 🗂️ DIF                 - DATASETS WITH DIF
│  └─ 🗂️ noDIF               - DATASETS WITHOUT DIF
├─ 🗂️ Modules                - R AND STATA MODULES
│  ├─ 🗂️ rosali_custom       - DATASETS WITH DIF
├─ 🗂️ RProject               - R SCRIPTS FOR VARIOUS TASKS
└─ 🗂️ Scripts                - R AND STATA SCRIPTS
   ├─ 🗂️ Analysis            - PCM ANALYSIS SCRIPTS
   └─ 🗂️ Data_generation     - SIMULATION SCENARIO SCRIPTS
      ├─ 🗂️ DIF
      └─ 🗂️ noDIF

Naming conventions

Initial Datasets

XXX_N - Scenario XXX / N individuals per group

Analyzed Datasets

noDIF / XXX_N.csv - Analysis for scenario XXX_N by PCM without accounting for DIF and confusion accounted for DIF / XXX_N.xls - Analysis for scenario XXX_N by PCM with DIF and confusion accounted for

noDIF_prop / XXX_N.csv - Analysis for scenario XXX_N by PCM without accounting for DIF and confusion accounted for by propensity score DIF_prop / XXX_N.xls - Analysis for scenario XXX_N by PCM with DIF accounted for and confusion accounted for by propensity score ROSALI-DIF_prop / XXX_N_original.xls - Analysis for scenario XXX_N by PCM with DIF accounted for after detection by ROSALI and confusion accounted for by propensity score RESIDUALS_prop / XXX_N_original.xls - Analysis for scenario XXX_N by PCM with DIF accounted for after detection by Andrich & Hagquist's residuals method and confusion accounted for by propensity score

Reproduction - TO BE MODIFIED

  1. Run /Scripts/Data_generation/NoDIF/scenarios_noDIF_baseline.do to simulate no DIF data
  2. Run files in 🗂️ /Scripts/Data_generation/DIF/ to simulate DIF data
  3. Run /RProject/Scripts/Analysis/pcm_nodif.R to analyze without accounting for DIF
  4. Run files in 🗂️ /Scripts/Analysis/DIF/ to analyze while accounting for DIF
  5. Run /Scripts/Analysis/DIF-ROSALI/pcm_dif_rosali.do to analyze data after accounting for DIF as detected by ROSALI
  6. Run /RProject/Scripts/Analysis/resali_analysis.R to perform residuals DIF detection and prepare data for PCM analysis.
  7. Run /Scripts/Analysis/DIF-RESIDUALS/pcm_dif_residus.do to analyze data after accounting for DIF as detected by the residuals method
  8. Run /RProject/Scripts/Analysis/aggregation.R to compile and visualize results

OR

  1. [BASH ONLY] Run prepare_file_structure.sh (only on first run) and autorun.sh (by default, will take multiple weeks to run. Please modify to run in parrallel if necessary)