. Author manuscript; available in PMC: 2022 Jan 1.

Published in final edited form as: Ear Hear. 2021 Nov-Dec;42(6):1656–1667. doi: 10.1097/AUD.0000000000001072

Table 3.

Results of Bayesian analyses of linear and nonlinear AV word perception models.

Model	R²	params	BIC	BF Reduced vs. Full	BF Parabolic vs. Linear
full linear	.879	4	−211.8
reduced linear	.879	3	−216.4	9.979
					>150
full parabolic	.928	3	−269.7
reduced parabolic	.928	2	−274.1	9.677

Note: Params refers to the number of parameters in the model, Full and Reduced refer to models with and without Age as a predictor, and Linear refers to the Multiple Linear Regression model; BIC and BF refer, respectively, to the Bayes Information Criterion for a specific model and the Bayes factor for a comparison of corresponding full and reduced models. The BF for the Parabolic vs. Linear models compares the evidence for the reduced forms of these models.