TABLE 3.

Generalized linear mixed model fit by maximum likelihood (Laplace approximation) for the corpus data.

Formula: overlap ∼ 1 + group size × transition type + (1 + transition type | T2_ID)
Model summary
AIC	BIC	logLik	Deviance	df.resid
1,782.7	1,819.5	−884.4	1,768.7	1,408
Scaled residuals
Min	1st Quartile	Median	3rd Quartile	Max
−1.0534	−0.7232	−0.5812	1.2107	1.9376
Groups: Term	Variance	Std. Dev.	Corr
Random effects:
T2_ID: Intercept	0.12422	0.3525
T2_ID: transition_type [QR]	0.07033	0.2652	−1.00
Number of obs		1415
Number of groups (T2_ID)		60
	Estimate	Std. Err.	z	p
Fixed effects
(Intercept)	−0.74698	0.07852	−9.513	<2e-16***
group_size [S.dyadic]	−0.39172	0.15649	−2.503	0.0123*
transition_type [S.QR]	0.13720	0.12931	1.061	0.2887
group size [S.dyadic]:transition_type [S.QR]	0.55533	0.25734	2.158	0.0309*

The response family was binomial with a logit link. Group Size and Transition Type are sum coded (±0.5).