. 2020 Nov 23;11:575032. doi: 10.3389/fpsyg.2020.575032

TABLE 5.

Estimated effects and coefficients for accentedness ratings.

Predictor		β estimate	Std. error	z value	p value
Testing Moment_POSTTEST		−0.055	0.066	−0.830	0.406
Gesture Condition_AV–P		0.066	0.223	0.295	0.768
Gesture Condition_AV–I		−0.058	0.225	−0.256	0.798
Phoneme_/u/		−0.139	0.356	−0.390	0.696
Testing Moment_POSTTEST * Gesture Condition_AV–P		0.428	0.093	4.478	0.000
Testing Moment_POSTTEST * Gesture Condition_AV–I		0.086	0.093	0.917	0.359
Testing Moment_POSTTEST * Phoneme_/u/		0.284	0.094	3.018	0.003
Gesture Condition_AV–P * Phoneme_/u/		0.337	0.093	3.608	0.000
Gesture Condition_AV–I * Phoneme_/u/		0.050	0.094	0.527	0.598
Testing Moment_POSTTEST * Gesture Condition_AV–P * Phoneme_/u/		−0.492	0.133	−3.704	0.000
Testing Moment_POSTTEST * Gesture Condition_AV–I * Phoneme_/u/		0.074	0.133	0.559	0.576

Random effects	Variance	Standard deviation

Participant	1.225	1.107
Speaker	0.159	0.399
Item	0.245	0.495

The intercept represents the following combination of variable levels: Testing Moment = Pretest, Gesture Condition = AV, Phoneme = /θ/. Asterisks (^∗) represent interactions, subscript signals the level of a categorical variable. Significant p-values are printed in bold. The model used in this analysis can be described as: Rating ∼ Testing Moment × Gesture Condition × Phoneme + (1| Participant) + (1 | Speaker) + (1 | Item).