Skip to main content

Table 4 BMI estimates (outcome equation) of the endogenous treatment effects model with an ordered outcome considering potential endogeneity between smoking and body weight

From: An empirical investigation of the impact of smoking on body weight using an endogenous treatment effects model approach: the role of food consumption patterns

Variables

BMI

Full sample

Males

Females

Smoker (smoking participation)

1.155***

1.236***

0.634*

(0.335)

(0.439)

(0.332)

Sociodemographic characteristicsa

 Gender (women)

0.329***

–

–

(0.070)

  

 30–60 years old

0.536***

0.554***

0.463***

(0.110)

(0.166)

(0.112)

 Over 60 years old

1.022***

0.948***

0.879***

(0.191)

(0.258)

(0.170)

 Married

0.332***

0.418***

0.206**

(0.101)

(0.153)

(0.102)

 Divorced/widowed

0.261**

0.158

0.222*

(0.114)

(0.180)

(0.118)

 Big city

−0.198***

−0.169

−0.188**

(0.075)

(0.110)

(0.081)

 Village/Rural area

0.124*

0.130

0.116

(0.071)

(0.107)

(0.079)

Frequent consumption of:

 Whole grain bread

0.014

−0.075

0.080

(0.059)

(0.087)

(0.070)

 White-toasted bread

−0.158**

−0.337***

0.040

(0.073)

(0.122)

(0.078)

 Fruits

0.107

0.115

0.077

(0.075)

(0.108)

(0.093)

 Vegetables

0.014

0.008

−0.005

(0.061)

(0.091)

(0.070)

 Meat products

0.318***

0.408***

0.234***

(0.083)

(0.153)

(0.076)

 Deep fried foods

−0.067

−0.216

0.140

(0.107)

(0.147)

(0.156)

 Confectionery

−0.160***

−0.217**

− 0.113**

(0.058)

(0.089)

(0.065)

μ1

−1.805

−2.273

−1.135

μ2

1.033

0.946

1.142

μ3

2.495

2.689

2.171

Log-Likelihood (both stages)

− 5423.587

− 2800.598

− 2584.326

  1. Standard errors are given in parentheses
  2. aAge: 18–29 years old (reference category). Marital status: single (reference category)
  3. Area of residence: town (reference category)
  4. *p < 0.1. **p < 0.05. ***p < 0.01 (Statistically significant variables are highlighted in bold)