Skip to main content

Table 4 Fractional logistic regression models for the median scores across by category

From: State licenses for medical marijuana dispensaries: neighborhood-level determinants of applicant quality in Missouri

 

(1)

Dispensary Supplement

(0-100%)

(2)

Business Plan

(0-100%)

(3)

Security

(0-100%)

(4)

Economic Impact

(0-100%)

(5)

Character

(0-100%)

Independent variables

Marginal

Effect

[95% C.I.]

Marginal

Effect

[95% C.I.]

Marginal

Effect

[95% C.I.]

Marginal

Effect

[95% C.I.]

Marginal

Effect

[95% C.I.]

Continuous variables

Semi-elasticity coefficients: (ΔScore)/(%ΔX)

Population density (in thousands)

0.011*

[0.003, 0.020]

0.012**

[0.004, 0.019]

0.010*

[0.001, 0.020]

0.009

[–0.002, 0.021]

0.002

[–0.002, 0.006]

Age dependency ratio x 100

0.005

[–0.021, 0.030]

–0.007

[–0.032, 0.017]

–0.007

[–0.037, 0.024]

0.005

[–0.028, 0.039]

0.004

[–0.009, 0.017]

Socioeconomic vulnerability index

–0.018

[–0.038, 0.002]

–0.010

[–0.027, 0.007]

–0.010

[–0.032, 0.013]

–0.015

[–0.038, 0.008]

–0.003

[–0.012, 0.006]

% Retail industry employment

0.002

[– 0.017, 0.021]

0.002

[–0.017, 0.020]

0.003

[–0.019, 0.025]

0.007

[–0.016, 0.029]

0.002

[–0.007, 0.011]

Binary variables

Level-level effects: ΔPr(Y = 1)/(ΔX)

Black or Hispanic majority (0,1)

0.022

[–0.003, 0.048]

0.007

[–0.019, 0.032]

–0.009

[–0.041, 0.024]

0.010

[–0.029, 0.049]

–0.005

[–0.018, 0.009]

No racial or ethnic majority (0, 1)

0.026*

[0.004, 0.048]

0.035**

[0.012, 0.059]

0.034

[–0.006, 0.075]

0.053**

[0.015, 0.090]

0.006

[–0.009, 0.020]

K-12 public school (0,1)

–0.001

[–0.018, 0.017]

–0.005

[–0.020, 0.010]

–0.004

[–0.024, 0.017]

0.000

[–0.021, 0.021]

–0.003

[–0.011, 0.005]

Wald χ2

18.59**

27.42**

11.39

13.56

4.38

p-value (Prob. > χ2)

0.010

0.000

0.122

0.060

0.735

Sample size

606

606

606

606

606

  1. Note Average marginal effects and confidence intervals are reported for fractional logistic models. For continuous variables, semi-elasticity estimates indicate the unit change in the predicted outcome given a 1% change in X. For categorical X variables, the marginal effects are shown in level-level form indicating the unit change in the predicted outcome given a one-unit change in X. Statistical significance is denoted by * and ** for significance at the 5% and 1% levels, respectively