Model sequence report

Overview

There are total of 29 distinct models, which can be organized into 4 groups:

• with fixed effects only (F): 9 models
• with 1 random term (R1): 9 models
• with 2 random terms (R2): 7 models
• with 3 random temrs (R3): 4 models

The following layout helps understand how each of the model group was constructed. The columns count the number of terms on the first level, while the rows indicate what predictors are added to the second level. Replace the stars in the name of the model with F for models with only the fixed effects and R1, R2, and R3 for models with the corresponding number of random terms in the second level.

For example, all models in group F are as follows.

For maps of R1, R2, and R3 model groups see the bottom of the report.

General form

The general form of the models can be expressed with LCM notation of Bollen & Curran (2006) or MLM notation of Snijders & Boskers (2011)

F-group : fixed effects

m0_F

m1_F

m2_F

m0a_F

m1a_F

m2a_F

m1b_F

m2b_F

m2c_F

R1-group : 1 random effect

m0_R1

m1_R1

m2_R1

m0a_R1

m1a_R1

m2a_R1

m1b_R1

m2b_R1

m2c_R1

R2-group : 2 random effects

m1_R2

m2_R2

m1a_R2

m2a_R2

m1b_R2

m2b_R2

m2c_R2

R3-group : 3 random effects

m2_R3

m2a_R3

m2b_R3

m2c_R3

Fit Information

Models w/ fixed only

F by row

F by column

Models with 1 random term

R1 by row

R1 by column

Models with 2 random terms

R2 by row

R2 by column

Models with 3 random terms

R3 by row

R3 by column

# General form (alternative) The general form of the models can be expressed with LCM notation of Bollen & Curran (2006) or MLM notation of Snijders & Boskers (2011)