Table 1. Descriptive statistics and univariate regression analyses.
| Variable Abbreviation | Definition | Mean (SD) | Regression coefficient† | Regression coefficient†† | ||||
| TB_INCI | 2003–2008 TB cumulative incidence | 0.0052 (0.0035) | – | – | ||||
| TB_INCI_6 | 2006–2008 TB cumulative incidence | 0.0024 (0.0016) | – | – | ||||
| ABOR_P | Aborigines % | 0.0775 (0.1966) | 1.60 *** | 1.46 *** | ||||
| BRIDE_P | % of population of brides from Southeast Asia | 0.0001 (0.0002) | −391.57 *** | −368.01 *** | ||||
| DENSITY | Township population/area (m2) | 0.0029 (0.0061) | −16.08** | −13.11** | ||||
| EDU1 | 8.2<Education years< = 8.7 (lower middle) | – | −0.06 | −0.05 | ||||
| EDU2 | 8.7<Education years< = 9.5 (middle) | – | −0.12 * | −0.09 | ||||
| EDU3 | Education years>9.5 (high) | – | −0.23 *** | −0.19** | ||||
| ELDER_P | % of Population >60 years old | 0.1413 (0.0398) | −0.20 | −0.83 | ||||
| HIV_INCI | 1984–2002 HIV cumulative incidence | 0.0001 (0.0001) | 410.19 * | 391.30 * | ||||
| HOU_PERS | Average number of persons per household | 3.5027 (0.4262) | 0.15** | 0.16** | ||||
| INCOME1 | 320<Average household income< = 440 (lower middle) | – | 0.08 | 0.06 | ||||
| INCOME2 | 440<Average household income< = 560 (middle) | – | −0.17** | −0.13 * | ||||
| INCOME3 | Average household income>560 (high) | – | −0.42 *** | −0.36 *** | ||||
| LABOR_P | % of population of laborers from Southeast Asia | 0.0109 (0.0152) | −6.93 *** | −5.09** | ||||
p<0.05 **p<0.01
p<0.001
Dependent variable: ln (TB_INCI)
Dependent variable: ln (TB_INCI_6).
ABOR_P, BRIDE_P, DENSITY, HIV_INCI, and LABORER_P were log transformed in regression analysis.
Average household income (in thousands Taiwan Dollars) was calculated using total income divided by number of households. The percentage of the population who received primary, junior high, senior high, bachelor’s, master’s and doctoral education were given. We used this information to calculate an average years of education for each township by giving a weight of 6, 9, 12, 16, 18 and 22 years to each education level. Because we were interested in how education level would affect the incidence of TB, we classified the average years of education into four groups using the quartiles as cutoff points. In this way, there were three dummy variables with the lowest serving as the reference group. The same procedure was performed for the average household income.