Variable | Full ordered logistic model (N = 346) | Restricted generalized ordered logistic model (N = 349) | |||
---|---|---|---|---|---|

Coefficient (standard error) | P-value (P > [z]) | Coefficient (standard error) | P-value (P > [z]) | Odds ratio | |

landscaping | 0.17 (0.31) | 0.58 | 0.46 (0.22) | 0.04 | 1.58 |

nursery supply | −1.96 (0.72) | 0.00 | −1.73 (0.67) | 0.01 | 0.18 |

nursery tree | −0.37 (0.42) | 0.38 | |||

nursery stores | −0.17 (0.46) | 0.72 | |||

architecture | −0.26 (0.38) | 0.49 | |||

Log (jobs) | 0.14 (0.09) | 0.11 | 0.13 (0.08) | 0.09 | 1.15 |

Log (longevity) | −0.18 (0.15) | 0.24 | |||

UF jobs | 0.02 (0.01) | 0.00 | −0.01 (0.01)^{a} | 0.17 | 0.98 |

corporate | 0.37 (0.22) | 0.09 | 0.36 (0.21) | 0.09 | 1.43 |

R&D | 0.25 (0.13) | 0.05 | 0.20 (0.12) | 0.08 | 1.22 |

supply chain | −0.10 (0.13) | 0.43 | |||

recruitment | 0.08 (0.11) | 0.46 | |||

retention | −0.06 (0.12) | 0.63 | |||

The likelihood ratio test of proportionality of odds | 67.64 | 0.00 | 12.54 | 0.56 |

↵

^{a}The Gamma function in the generalized ordered logistic regression produces 4 Gamma coefficients for the variables that deviate from proportionality. So,*UF jobs*has Gamma 2 value of 0.02 (0.01), Gamma 3 value of 0.04 (0.01), and Gamma 4 value of 0.04 (0.01). All 3 Gamma estimates are statistically significant.