Software Solutions

Excel

Cloud

R Package

SAS

---

AgreeStat360 for Excel / Windows

AgreeStat360 for Excel Windows is a self-automated workbook that contains a VBA (Visual Basic for Applications) program, and therefore requires no installation. It is a point-and-click program that is easy to use and can compute many well-known inter-rater reliability coefficients along with their standard errors.

Overview of AgreeStat360/xls Features

Have an  overview of AgreeStat360 capabilities before downloading a fully featured trial version to play with.   

---

Download the standard version of AgreeStat360 / Excel Windows here. It contains several sample datasets with detailed instructions for analysis. If your computer's regional settings are not those of the US or English Canada, you may need to temporarily adjust them (see instructions) for AgreeStat360 / Excel to work smoothly. 

Are you using comma as the decimal symbol? Check here!

---

Back to the top

AgreeStat360.com/Cloud-Based

If you do not use Excel then I still expect you to use a browser to surf the Internet. The browser is indeed all you need to run agreestat360.com, the cloud version of AgreeStat. You may get an overview of agreestat360.com here.

Whether you are using a PC (Windows, Linux, ...), a Mac, or even a Tablet, simply type agreestat360.com on your browser and you will have access to a convenient system for analyzing your inter-rater reliability data.

 AgreeTest / Cloud-Based

AgreeTest is an App, which you can access through your browser of choice. It is designed to test the diffference between 2 agreement coefficients for statistical significance. You can test correlated as well as uncorrelated coefficients.  Try it out.

Back to the top

R Packages & Functions

Users of the R language and computing envireonment can compute the same agreement coefficients as in AgreeStat/Excel.  There are 2 ways this can be achieved:

1. irrCAC and irrICC Packages
2. R functions

---

The irrCAC and irrICC packages were both developed by Dr. Gwet to analyze Chance-corrected Agreement Coefficients (CAC) and Intraclass Correlation Coefficients (ICC).  Package irrCAC is used for CAC analysis, while irrICC package is used for ICC analysis. Both packages can be installed either from CRAN as any other standard R package, or from github.

To install these packages from github, proceed as follows:

• devtools::install_github("kgwet/irrCAC")
• devtools::install_github("kgwet/irrICC")

---

Here is a list of downloadable R script files, each containing several R functions that can be used directly from R to compute various chance-corrected agreement coefficients:

1) agree.coeff2.r (for 2 raters only with ratings in a contingency table)
The functions in this script file compute various agreement coefficients and their standard errors when dealing with 2 raters, and ratings that are organized in the form of a contingency table (or a square matrix) showing counts of subjects by rater and category. You may use this format if each rater rated all subjects. Otherwise subjects rated by one rater and not by the other may not be properly classified.

2) agree.coeff3.dist.r (for 2 raters or more)
The functions in this script file compute various agreement coefficients and their standard errors when dealing with multiple raters, and ratings that are organized in the form of an n x q table showing counts of raters by subject and category. Here n represents the number of subjects and q the number of categories.

3) agree.coeff3.raw.r (for 2 raters or more - requires the script file "weights.gen.r" to be loaded as well.)
The functions in this script file compute various agreement coefficients and their standard errors when dealing with multiple raters, and rating data organized in the form of an n × r table showing the (alphanumeric) raw ratings that the raters assigned to the subjects. Here n represents the number of subjects and r the number of raters. The data is presented in the form of n rows containing r ratings each.

4) Weights.gen.r (this script is required to run the functions in the script file paired.ttest.r used for paired t-test.)
The functions in this script file generate various weights to be used when computing weighted agreement coefficients.

5) Paired t-test for agreement coefficients This script file contains functions that implement the paired t-test for testing the difference between 2 agreement coefficients for statistical significance, based on the linearization method of Gwet (2016).
In order to use any of the functions contained in these script files, you need to read this entire script file along with weights.gen.r into R using. This is accomplished using the source() function. A script file "weights.gen.r" for example is sourced as follows:

>source("C:\\YourDirectory\\weights.gen.r")

Back to the top

SAS System

SAS users who want to compute Cohen's kappa or Gwet's AC1 or AC2 coefficients for 2 raters, could do so using the FREQ procedure after specifying the proper parameters. The data must be in the form of a contingency table. Note that the AC1 option only became available in SAS/STAT version 14.2. For coefficients other than Cohen's kappa and Gwet's AC1, you may consider the use of SAS macro programs. More information can be found here.

Multiple-rater data are generally not organized in contingency tables and cannot be analyzed with the FREQ procedure. You may learn about the availability of SAS macro programs by visiting the webpage Inter-Rater Reliability using the SAS System. You will also find information about Intraclass Correlation Coefficients used for analyzing quantitative ratings.

Back to the top

---

Error Messages

You get an error message when launching AgreeStat / Windows?

This uncommon problem may be due to AgreeStat360 being incompatible with your regional settings. Tempporarily change your regional settings (see instructions here) 

Back to the top

Using Comma as the Decimal Symbol?

If you are from a country that uses the comma as the decimal symbol, AgreeStat360 may not work well.  Tempporarily change your regional settings (see instructions here)   

Back to the top

Additional Resources

Here are some additional resources many users may find to be of interest.

---

Inter-Rater Reliability for Stata Users

Stata users now have a convenient way to compute a wide variety of agreement coefficients within a general framework. The module KAPPAETC can be installed from within Stata and computes various measures of inter-rater agreement and associated standard errors and confidence intervals.

---

Implementing a general framework for assessing interrater agreement in Stata

This interesting article related to the Stata module KAPPAETC an entitled "Implementing a general framework for assessing interrater agreement in Stata" by Daniel Klein is a must read for Stata users. KAPPAETC is a remarkably well written Stata package, which I strongly recommend to all Stata users for calculating the the AC1, Kappa, Krippendorff agreement coefficients and associated standard errors, and confidence intervals.

Inter-Rater Reliability with Python

Researchers who conduct their data analysis in Python may want to look into the irrCAC library by Aris Fergadis on GitHub.