This web tool computes Lin’s agreement statistics for continuous data when the error structure is assumed constant (absolute difference) or proportional (% change). Agreement statistics include concordance correlation coefficient (CCC), precision and accuracy coefficients, total deviation index (TDI), coverage probability (CP), and their one-sided confidence limits. It also produces the agreement plot around the concordance line with the band of acceptable deviation allowance. Created by Dr. Lawrence Lin using the Python function converted from validated SAS macro, based on materials in Chapter 2 of Statistical Tools for Measuring Agreement, Springer, 2012 by Lin et al. To use the tool, first upload your data. Your data must be a .csv file. Then enter values for all parameters. After pressing the ‘Run Analysis’ button, it will present agreement statistics and plot. Pushing the back arrow, you may interactively update the results by changing values of parameters. Close and relaunch the tool for each new data set. Note that when the TDI allowance is set too large, you might get an error about s_T because the CP value is one and its standard error is zero.

How It Works

1. Upload your data: Go to Tool Page. Begin by importing your data set in .csv file into the platform. It is best to keep your file local at your computer for uploading.
2. Configure analysis: Set your preferred parameters, choose between absolute difference or percent change, and customize your statistical measures and plot. See Parameter Page for details.
3. Visualize results: Hit ‘Run Analysis’ button. Review agreement plot and statistics. See Statistics Page for details.