Variance components between neighbourhoods | Parameter | Variance estimate (μ) | SD | df | χ^{2} | p value |
---|---|---|---|---|---|---|

Intercept | μ0 | Constrained to zero | – | – | – | – |

BMI (centred) | μ1 | 0.018 | 0.13 | 51 | 84.44 | 0.002 |

Within neighbourhood fixed effects | Parameter | Coefficient | SE | t ratio | p value | |
---|---|---|---|---|---|---|

Intercept | γ00 | 0.26 | 0.059 | 4.31 | 0.0001 | |

Age | γ20 | −0.013 | 0.006 | −2.08 | 0.038 | |

Individual affluence* | γ30 | −0.21 | 0.22 | −0.98 | 0.33 |

Between neighbourhood effects | Parameter | Coefficient | SE | t ratio | p value | |
---|---|---|---|---|---|---|

*Individual affluence = dummy variable: “low income” (≤$20000) v not (>$20000); †neighbourhood affluence = centred variable: mean average family income across census tract boundary. | ||||||

Intercept | γ10 | 0.27 | 0.027 | 10.15 | 0.0001 | |

Neighbourhood affluence† (centred) | γ11 | 0.000002 | 0.000001 | 2.59 | 0.013 | |

Final model
| ||||||

Level-1 | Body satisfaction = β_{0} + β_{1} (BMI) + β_{2} (age) + β_{2} (individual affluence*) | |||||

Level-2 | β_{0} = γ_{00} | |||||

β_{1} = γ_{10} + γ_{11} (neighbourhood affluence†) + μ_{1} | ||||||

β_{2} = γ_{20} | ||||||

β_{3} = γ_{30} |