Variable | Category | Marginal effect | SE |
Gender | Male (base) Female | −0.0004 | 0.0024 |
Age | 0.0004* | 0.0002 | |
Education | None/elementary/middle school (base) | ||
High school/associate degrees | 0.0056 | 0.0044 | |
Bachelor/master degree | −0.0044 | 0.0037 | |
Occupation | Worker (base) Unemployed | 0.0125 | 0.0122 |
Retired/disabled/other | 0.0293 | 0.0318 | |
Student | 0.0635* | 0.0288 | |
Housekeeper | 0.0141 | 0.0149 | |
Smoking intensity | <0.0001 | <0.0001 | |
Age of smoking initiation | −0.0001 | 0.0003 | |
City | Bogotá (base) Medellín | −0.0040† | 0.0022 |
Cartagena | −0.0021 | 0.0036 | |
Cúcuta | 0.0619‡ | 0.0181 | |
Cali | 0.0009* | 0.0004 |
Marginal effects of probit estimates based on DEICS-COL-2016. Estimates from 1678 observations, less than the sample size because of missing values in age of initiation. Age, age of initiation and smoking intensity included linear and squared terms. Marginal effects calculated at the mean of covariates. Heteroskedasticity-robust SEs in the last column.
*At 5%.
†At 10%.
‡Denotes significance at 1%.
COL, Colombia; DEICS, Demand for Illicit Cigarettes Survey.