Table 2. HLM estimates of Model 2. Dependent variable: strength of music preference.
| Predictor | b | SE | t | p | df | Δσ2 |
|---|---|---|---|---|---|---|
| Part A: Level 2 predictors of the intercept (χ2 = 37.1, p < .001)a | ||||||
| Past functional experiences | 0.42 | 0.04 | 9.4 | < .001 | 113 | 49.1 |
| Control: age | -0.01 | 0.01 | -1.5 | .14 | 113 | |
| Control: sex | 0.09 | 0.18 | 0.5 | .63 | 113 | |
| Control: singing | 0.22 | 0.22 | 1.0 | .32 | 113 | |
| Control: instrument | 0.01 | 0.10 | 0.1 | .92 | 113 | |
| Control: music studies | -0.16 | 0.23 | -0.7 | .49 | 113 | |
| Control: musician | 0.37 | 0.22 | 1.7 | .10 | 113 | |
| Part B: Level 1 predictor (χ2 = 277.8, p < .001)b | ||||||
| Effects of music listening | 0.58 | 0.05 | 12.0 | < .001 | 113 | 20.5 |
| Part C: Level 2 predictors of the slope of the relationship between the effects of music listening and the strength of music preference (χ2 = 257.5, p < .001)c | ||||||
| Past functional experiences | 0.02 | 0.03 | 0.7 | .51 | 113 | 0.0 |
| Control: age | 0.00 | 0.05 | 0.2 | .81 | 113 | |
| Control: sex | 0.01 | 0.10 | 0.1 | .92 | 113 | |
| Control: singing | -0.10 | 0.10 | -1.0 | .30 | 113 | |
| Control: instrument | 0.23 | 0.10 | 2.4 | .02 | 113 | |
| Control: music studies | -0.08 | 0.11 | -0.7 | .49 | 113 | |
| Control: musician | 0.02 | 0.10 | 0.2 | .84 | 113 | |
Note. The b statistic is the unstandardized regression coefficient. Δσ2 is the amount of variance in the criterion the predictor is able to explain. The χ2 statistics are calculated from the difference between the deviance of the unconditional model and the deviance of the conditional model.
aSee Path B in Fig 1.
bSee Path C in Fig 1.
cSee Path D in Fig 1.