ABSTRACT
Model assessment involves choices between competing models in terms of best fit, and checks to ensure model adequacy. For example, even if one model has a superior fit, it still needs to be established whether predictions from that model check with, namely, reproduce satisfactorily, the observed data. Checking may also seek to establish whether model assumptions (e.g. normality of random effects) are justified, whether the model reproduces particular aspects of the data, and whether particular observations are poorly fit (Sinharay and Stern, 2003; Berkhof et al., 2000; Kelly and Smith, 2011; Lucy, 2018; Conn et al., 2018; Park et al., 2015).
