. 2014 Mar 19;9(3):e92563. doi: 10.1371/journal.pone.0092563

Table 2. Predicting emotions: results from multilevel modeling.

	Enjoyment			Pride			Anger			Anxiety
	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3
Level 1
Intercept (γ₀₀₀)	2.52^***	2.25^***	2.71^***	1.95^***	1.74^***	2.10^***	1.97^***	1.79^***	2.10^***	1.50^***	1.49^***	1.51^***
Trait (γ₁₀₀)	0.05	0.42^**	−0.22^*	0.56^***	1.02^***	0.22^*	0.77^***	1.09^***	0.54^***	0.28^**	0.48^***	0.12
Level 2
Self-concept (γ₀₁₀)	0.19^***	0.17^**	0.17^*	0.17^**	0.19^*	0.12	−0.14^**	−0.22^**	−0.10	−0.11^*	−0.09	−0.13
Cross-level interactions L1–L2
Self-concept × Trait (γ₁₁₀)	0.23^*	0.13	0.35^**	0.30^***	0.28^**	0.36^**	−0.34^***	−0.32^**	−0.33^**	−0.33^***	−0.41^***	−0.24^*
Level 3
CH_dummy (γ₀₀₁)		0.46^***			0.37^**			0.31^*			0.02
DE_dummy (γ₀₀₁)			−0.46^***			−0.37^**			−0.31^*			−0.02
Cross-level interactions L1–L3
Trait × CH_dummy (γ₁₀₁)		−0.63^***			−0.80^***			−0.55^**			−0.36^*
Trait × DE_dummy (γ₁₀₁)			0.63^***			0.80^***			0.55^**			0.36^*
Cross-level interactions L2–L3
Self-concept × CH_dummy (γ₀₁₁)		−0.01			−0.07			0.12			−0.04
Self-concept × DE_dummy (γ₀₁₁)			0.01			0.07			−0.12			0.04
Cross-level interactions L1–L2–L3
Trait × Self-concept × CH_dummy/(γ₁₁₁)		0.22			0.08			−0.01			0.17
Trait × Self-concept ×/DE_dummy (γ₁₁₁)			−0.22			−0.08			0.01			−0.17
Variance components
Within-student (L1) variance ( ²)	1.103	1.096		0.924	0.916		1.120	1.118		0.735	0.735
Intercept (L2) variance (τ₀₀)	0.419	0.391		0.460	0.429		0.419	0.404		0.254	0.255
Slope (L2) variance (τ₁₁)	0.174	0.122		0.252	0.128		0.243	0.173		0.258	0.217
Intercept-slope (L2) covariance (τ₀₁)	−0.269	−0.215		−0.320	−0.234		−0.230	−0.196		−0.109	−0.104
Intercept (L3) variance	0.034	0.014		0.059	0.051		0.046	0.042		0.021	0.020
Explanatory power	0.022	0.090		0.261	0.369		0.314	0.357		0.295	0.332

Model 1 = combined model; Model 2 = combined dataset with German sample as reference group; Model 3 = combined dataset with Swiss sample as reference group.

Description of variables: Trait = Trait dummy (0 = state, 1 = trait); CH_dummy = Swiss dummy (0 = German sample, 1 = Swiss sample); DE_dummy = German dummy (0 = Swiss sample, 1 = German sample).

German sample: N_{Level 1} = 509; N_{Level 2} = 94; N_{Level 3} = 39; Swiss sample: N_{Level 1} = 880; N_{Level 2} = 131; N_{Level 3} = 41.

Explanatory power refers to the proportion of slope variance explained by the Level 2 and Level 3 predictors. The slope variance of the models in which no cross-level interaction is included was: τ₁₁ = 0.178 for enjoyment in the combined model and τ₁₁ = 0.134 in the German/Swiss model; τ₁₁ = 0.341 for pride in the combined model and τ₁₁ = 0.203 in the German/Swiss model; τ₁₁ = 0.354 for anger in the combined model and τ₁₁ = 0.269 in the German/Swiss model; τ₁₁ = 0.366 for anxiety in the combined model and τ₁₁ = 0.325 in the German/Swiss model.

*p<.05,

**p<.01,

***p<.001.