r2_mlm: A command for computing R-squared measures for models fit by mixed

The article by Rights and Sterba (2023, Multivariate Behavioral Research 58: 340-367) provides a comprehensive framework for computing <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:msup> <mml:mi>R</mml:mi> <mml:mn>2</mml:mn> </mml:msup> </mml:math> measures for (linear) multilevel models. In this article, we introduce r2_mlm , a postestimation command for mixed that computes <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:msup> <mml:mi>R</mml:mi> <mml:mn>2</mml:mn> </mml:msup> </mml:math> measures using Rights and Sterba’s framework. We explain how this <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:msup> <mml:mi>R</mml:mi> <mml:mn>2</mml:mn> </mml:msup> </mml:math> framework works and demonstrate how r2_mlm can be used to compute various <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:msup> <mml:mi>R</mml:mi> <mml:mn>2</mml:mn> </mml:msup> </mml:math> measures for models fit by mixed . One of the most useful features of r2_mlm is that it will produce warning messages if the user specifies the model in a way that may lead to conflation bias (an easily overlooked issue). Finally, we walk through a simple example and explain how to interpret the various <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:msup> <mml:mi>R</mml:mi> <mml:mn>2</mml:mn> </mml:msup> </mml:math> measures.