Using Generalized Linear Mixed Models in the Analysis of Count and Rate Data in Single-case Eperimental Designs: A Step-by-step Tutorial

Generalized linear mixed models (GLMMs) have great potential to deal with count and rate data in single-case experimental designs (SCEDs). However, applied researchers face challenges to apply such an advanced approach in their own studies. Hence, our study aimed to provide a tutorial and demonstrate a step-by-step procedure of using GLMMs to handle SCED count and rate outcomes. We utilized an empirical examplewith a purpose to examine the effect of prelinguistic milieu teaching (PMT) on prelinguistic intentional communication for six school-age children with autism. The outcomes were sustained intentional communication (frequency count) and initiated intentional communication (rate). A step-by-step analytical approach with GLMMs was illustrated and associated R and SAS code was provided. The results overall supported the original conclusions of the effectiveness of PMT, whereas additional evidence regarding the precise estimate of the individual treatment effect and between-case variation of the treatment effect were also interpreted. The implications of the similarities and differences between the findings based on GLMMs and from the original study were discussed.